Movatterモバイル変換


[0]ホーム

URL:


CN113435833A - City three-dimensional model collaborative management method and system for smart community - Google Patents

City three-dimensional model collaborative management method and system for smart community
Download PDF

Info

Publication number
CN113435833A
CN113435833ACN202110651680.3ACN202110651680ACN113435833ACN 113435833 ACN113435833 ACN 113435833ACN 202110651680 ACN202110651680 ACN 202110651680ACN 113435833 ACN113435833 ACN 113435833A
Authority
CN
China
Prior art keywords
dimensional model
monitoring center
site
model
urban
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110651680.3A
Other languages
Chinese (zh)
Other versions
CN113435833B (en
Inventor
刘俊伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Terry Digital Technology Beijing Co ltd
Original Assignee
Terra It Technology Beijing Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Terra It Technology Beijing Co ltdfiledCriticalTerra It Technology Beijing Co ltd
Priority to CN202110651680.3ApriorityCriticalpatent/CN113435833B/en
Publication of CN113435833ApublicationCriticalpatent/CN113435833A/en
Application grantedgrantedCritical
Publication of CN113435833BpublicationCriticalpatent/CN113435833B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Images

Classifications

Landscapes

Abstract

The invention provides a city three-dimensional model collaborative management method of a smart community, which is characterized by comprising the steps of S1, setting a server S and a monitoring center C, setting a main monitoring center CC, setting a server SS of the main monitoring center CC, and setting a city smart community site three-dimensional model; s2, the mobile terminals of various personnel and users are equipped; a device capable of recording whether the person is located in the intelligent community site or not is arranged in the intelligent community site; s3, establishing an artificial intelligence model AI for searching the urban intelligent community site; s4, searching for smart community sites at real time and fixed points in the three-dimensional model established in the step S1 by using the AI model established in the step S3 and the app established in the step S2, and marking the smart community sites in the three-dimensional model; s5, when finding out the site J of the urban intelligent community, the three-dimensional model established in the step S1 can be used for communication and connection with the site J, so that the person can rush to the fixed point to perform medical aid service. The method can provide the same coordinated management and timely and visible emergency treatment scheme for urban medical care and sanitary work, other property, business service and smart home.

Description

City three-dimensional model collaborative management method and system for smart community
Technical Field
The invention relates to a cooperative management method of a smart community, in particular to a three-dimensional model cooperative management method and a three-dimensional model cooperative management system of the smart community, and belongs to the field of site management.
Background
With the development of population aging forms, the smart community should also be more able to afford the health and prevention of the middle-aged and the elderly. However, in the existing situations, such as a nursing home, an elderly college, hospitals and nursing centers at all levels, as well as schools of social enterprises and public institutions with scattered sanitation rooms, and pharmacies and drug stores in cities, the smart community sites generally have respective management schemes, and how to cooperate is not considered.
When problems occur, such as accidents like old people stroke, heart attack, faint and fall of unknown reasons, and industrial accidents, the enterprise and public institutions can only think of contacting hospitals at all levels for rescue and treatment, and usually describe the positions and the situations of patients in the places in the telephone communication, sometimes the positions cannot be known exactly, and the time is increased for thinking about where the help-seeking person is. These details all cause untimely times and potentially over the rescue prime period. Although a relatively perfect management system is arranged in each level of hospital care center, and intellectualization is kept up to the latest in recent years, sudden accidents of middle-aged and elderly people in the society and other people are not actively monitored, so that medical treatment and treatment services are only passively provided. While the scattered health rooms and urban prescriptions and pharmacies are not involved in active discovery and monitoring, but only provide passive primary medical and pharmaceutical products to patients, and thus the resources are not well utilized.
Not only do smart communities consider using the internet and even internet of things to provide convenient and timely services for community members in medical terms, but also other aspects of life, such as property, business services (e.g., express delivery sites, online shopping brick and mortar stores, traditional brick and mortar stores, food wholesale and dish markets), and smart homes.
Therefore, in the prior art, various intelligent community sites cannot coordinate services of various aspects of the society into a system on the whole, so that a timely, visible and efficient service management scheme cannot be provided for specific objects such as middle-aged and old people, emergent accident crowds and daily life, and prediction work cannot be well done through big data analysis.
Disclosure of Invention
In order to solve the problem that the intelligent community sites lack resource integration in the passive and scattered management situation in the prior art, the invention considers the following solution: firstly, carry out the mark and the location of map formula to information such as all kinds of wisdom community sites and all kinds of medical treatment in the city, nurse, the state of an illness, the prevention, the property, business service, wisdom house, the second, calculate through artificial intelligence and find nearest all kinds of wisdom community site service sites through the crowd of being concerned with (like old person and disease in endowment site and the medical site, because of public business poor, and the crowd that needs accomplish special operation, and ordinary socioer) place and realize the communication, thirdly, realize the daily cooperative management of the visual wisdom community site in city.
The various intelligent community sites include but are not limited to various endowment medical and health institutions, such as sites and organizations related to medical treatment, endowment, professional medical treatment and beauty and health care services, including hospitals at all levels, nursing centers, various professional medical and beauty and health care sites, social enterprises and public institutions and scattered health rooms of schools, pharmacies and drug stores at all levels of cities, and service office places in the aspects of property, business service and intelligent home furnishing. Personnel in the smart community site refer to working or temporary workers in the smart community site, and can provide various professional services for the society. For various medical and health institutions for the aged, including but not limited to doctors with qualified medical sites, trainees, nurses, doctors in all levels of pharmacy and pharmacy, and various scattered persons (free medical personnel with qualified certification) who can provide certain medical treatment; for the property, business service and smart home industries, the smart community sites respectively comprise a property office building, an express delivery site, an online shopping entity store, a traditional entity store, food wholesale, a vegetable market and the like, and a house where each community owner is located. The user include all kinds of wisdom community website outsiders, total monitoring center CC staff includes the sampling personnel for AI modeling at least. The management in the collaborative management includes emergency and daily management, and the prevention work is done according to the data for scientific research and analysis. The fixed point refers to the current geographic position of a person, a worker or a user. The app generally refers to an application program installed on an intelligent mobile terminal such as a mobile phone, and may also be application software installed on a desktop computer.
Therefore, the invention provides a city three-dimensional model collaborative management method of an intelligent community site, which is characterized by comprising the following steps:
s1, a server S and a monitoring center C are set in the smart community site, a main monitoring center CC and a server SS of the main monitoring center CC are set, and a city three-dimensional model of the city smart community is set;
s2, arranging mobile terminals provided with the urban three-dimensional model collaborative management application program app of the smart community for personnel and external community users in the smart community site; providing a device capable of recording whether the person is currently located in the intelligent community site, wherein the device is used for indicating whether the community is in a state that the person capable of providing service is sufficient;
s3, establishing an artificial intelligence AI model for searching the urban intelligent community site;
s4 uses the AI model established in step S3 and the app of step S2 to realize real-time fixed-point finding of smart community sites in the three-dimensional model established in step S1 and marking in the three-dimensional model.
In one embodiment, step S4 is followed by:
when finding the site J of the urban intelligent community, the S5 can contact the site J through the three-dimensional model communication established in the step S1, so that the person can attend to the fixed point for service.
About S1
The S1 specifically includes:
s1-1, setting an independent server S and an independent monitoring center C in various intelligent community sites, setting a main monitoring center CC to perform real-time processing and analysis on data of the independent server S and the independent monitoring center C, and performing cooperative management on various intelligent community sites, wherein a server SS of the main monitoring center CC is used for three-dimensional model modeling and AI model modeling;
s1-2, establishing an urban road network model of the artificial intelligent network;
s1-3, building a city building network model of the artificial intelligent network;
s1-4, fusing the models established in the steps S1-2 and S1-3 to form a two-dimensional model of the urban road and the building
Figure 465155DEST_PATH_IMAGE001
S1-5 two-dimensional model of urban road and building based on step S1-4
Figure 996499DEST_PATH_IMAGE001
Constructing the unified three-dimensional model
Figure 166581DEST_PATH_IMAGE002
Wherein, step S1-2 specifically includes:
s1-2-1, establishing an urban geographic coordinate system E, wherein an XOY plane represents the ground, generating road continuous nodes by a node generator comprising an encoder and a decoder by utilizing an RNN recurrent neural network algorithm based on urban remote sensing images, connecting the two nodes before and after generation in the generation process, inputting the new generated nodes into the node generator to continuously generate new nodes, continuously connecting the generated new nodes, and circularly connecting the nodes to form a road network;
s1-2-2, widening all lines in a road network according to a preset width w to form road width lines with a certain width, and accordingly obtaining an urban road network model, wherein w is widened according to the corresponding road width in the remote sensing image, preferably, w is 0.5-1.5 times of the average value of all road widths in the remote sensing image, more preferably, 0.5-1 times of motor vehicle roads and non-motor vehicle roads and 1-1.5 times of pedestrian roads. It is understood that a pedestrian road shall include a road that can be traveled by a person or a road vehicle (e.g., an ambulance, a volunteer service vehicle, etc.) within an area of a building complex such as a street in a city, a sidewalk beside a non-motorized lane, a cell or a factory building.
In one embodiment, the widening is done on both sides with the lines forming the road network as the central axis.
Step S1-3 specifically includes:
s1-3-1, based on the urban remote sensing image in the step S1-2-1, extracting a series of feature maps obtained by different convolutional layers by using a VGG-16 algorithm without an added layer as a CNN main network, wherein the feature maps are 1/2-1/10, preferably 1/8 of the size of an input image;
meanwhile, a characteristic pyramid is constructed by using different layers of a CNN main network through an image pyramid algorithm FPN, and the frames of a plurality of buildings are predicted,
s1-3-2, for each building in the plurality of buildings, obtaining a local feature map F of the building by using a RoIAlign algorithm on the feature maps obtained by the series of different convolutional layers and the corresponding frame of the building;
s1-3-2, forming a polygonal boundary cover M by adopting convolution layer processing on the local feature map F of each building, and then forming a plurality of predicted vertexes P of the boundary cover M by utilizing convolution layer processing; wherein polygonal bounding box M refers specifically to the vertical projection of the XOY plane describing the building in E;
s1-3-3, selecting the point with the highest probability in P as the starting point
Figure 203807DEST_PATH_IMAGE003
Performing multi-step prediction by using a multi-layer RNN algorithm of convolution long-short term memory ConvLSTM to obtain multiple prediction points
Figure 311702DEST_PATH_IMAGE004
(t is step number) closed building boundary polygons to form an urban building network model;
s1-3-4 representing index points of the building for intersection of longest and next longest diagonal lines in each polygon boundary
Figure 850000DEST_PATH_IMAGE005
And arranging a road guide point at every preset distance on the road network
Figure 468063DEST_PATH_IMAGE005
And integrating the building index points and the road index points into an index library K (each index point in the index library K is represented by a coordinate corresponding to the position E).
S1-4 specifically comprises the step of fusing the models established in the steps S1-2 and S1-3 according to the relative coordinate positions of the buildings and the roads in the remote sensing image under the urban geographic coordinate system E to form a two-dimensional model of the urban roads and the buildings
Figure 817136DEST_PATH_IMAGE001
S1-5 presetting the two-dimensional model
Figure 206136DEST_PATH_IMAGE001
Representing the height of the building layer by the polygon boundaries in the Z coordinate H under E, marking the building index points of various intelligent community sites with geometric figures with different colors to form a three-dimensional model of the urban intelligent community site
Figure 157911DEST_PATH_IMAGE002
(ii) a The mobile terminal can inquire a specified urban smart community site through the app and can display a corresponding three-dimensional model
Figure 771426DEST_PATH_IMAGE002
Displaying corresponding geometric figure marks and site names, geographic coordinates of the sites and names of roads of the sites; the mobile terminal, the monitoring center C of various smart community sites and the main monitoring center CC interface can be manually clicked respectively, and a cursor moves to the geometric figure marks with different colors or is clicked by the cursor to display and/or enter the smart community service information interface, realize a service reservation function, realize emergency contact by pressing the emergency contact function for a long time, and/or register or enter a smart management system in the corresponding site or view other smart community sites nearby. Preferably, the long press time is 2-4 seconds and a confirmation dialog box pops up.
It should be appreciated that the smart community service information interface can be viewed or accessed for use by the app respective application module alone. The different colors are used to distinguish between the site buildings and the non-site buildings, and may be the same as or different from the first color or gray scale or the second color or gray scale defined in the later detection circle forming process.
The intelligent community service information comprises registration information, pricing and charging information, medical equipment (such as an X-ray machine, a CT (computed tomography), an NMR (nuclear magnetic resonance) machine, an expiration detection machine and the like) queuing reminding information, current personnel information, hospitalization or sitting room information, medicine use information, other nearby intelligent community site information, medical records of patients and nursed objects, medical examination data and records and current health data in the aspect of medical service. The current personnel information comprises the brief introduction of personnel, the schedule of sitting treatment, whether sitting treatment is currently performed or not, whether emergency tasks are performed or not when the personnel go out and the like. The medicine use information comprises medicine use methods, cautions, information for reminding the use of the medicine, and information for re-diagnosis reminding or on-line consultation and re-diagnosis communication and the like. And the other properties, business services and smart homes are provided with corresponding service information for reference according to the characteristics of various industries.
About S2
The step S2 specifically includes:
s2-1, arranging mobile terminals for personnel in the smart community site and staff of a main monitoring center CC, installing a professional version of a city three-dimensional model collaborative management application program app of the smart community in the mobile terminals, arranging mobile terminals for users, and installing a user version of the city three-dimensional model collaborative management application program app of the smart community in the mobile terminals; the professional app version can register or enter an intelligent management system in a corresponding site, and the user app version only keeps displaying or enters an intelligent community service information interface and realizes the functions of service reservation and emergency contact; wherein, the CC staff of the main monitoring center is equipped with a mobile terminal for AI modeling;
s2-2, installing face recognition device on each entrance and exit door in the intelligent community site, and when a person goes out, indicating that the person has gone out of the site through face recognitionRecording and reducing a number of real-time emergency tasks, or installing an emergency prompt information module on the app, sending rescue arrangement and reminding information to the app through a monitoring center C or a main monitoring center CC, and recording that a clicked person starts to execute the emergency task by clicking the emergency prompt information module, so that a number of real-time emergency tasks are reduced in the site; the monitoring center C and the main monitoring center CC store the records and the records are stored in the three-dimensional model
Figure 150455DEST_PATH_IMAGE002
And correspondingly, updating the record of whether the travel executes the emergency task in the intelligent community service information so as to indicate whether the site is in a state of sufficient personnel capable of providing service.
The emergency contact function and the emergency prompt information module may be the same or different app function buttons.
About S3
The S3 specifically includes:
s3-1, positioning an initial current geographic position by the aid of a mobile terminal equipped by staff of the CC at preset intervals, and finding out a corresponding three-dimensional model in the indexing library K according to the current geographic position
Figure 934740DEST_PATH_IMAGE002
The road index point with the shortest straight line distance with the road index point; preferably, the predetermined time period is between 1 day and 1 month;
s3-2, using the road sign guide point with the shortest straight line distance as the center of circle, in the three-dimensional model
Figure 690207DEST_PATH_IMAGE002
Finding the geometric figure marks with different colors representing the smart community site buildings in a circle with a preset radius R and acquiring whether the personnel corresponding to each mark are in sufficient state; if no different color geometric figure mark is found which satisfies the sufficient state of the person, enlarging the radius R by a specified step length until the found; the specified step length is 50-5000 m;
s3-3, according to the circle center and the geometric figure marks with different colors meeting the sufficient state of people in the R range, finding the geometric figure mark which is closest to the circle center in a straight line distance and meets the sufficient state of people, and marking the geometric figure mark which is closest to the straight line distance and meets the sufficient state of people as a first color or gray scale, wherein the geometric figure marks which represent other intelligent community buildings with different colors in the circle range are unified as a second color or gray scale, and other areas in the remaining circles are unified as a third color or gray scale, so that a detection circle is formed;
s3-4 urban three-dimensional model
Figure 112223DEST_PATH_IMAGE002
Repeating the steps S3-1-S3-3 while continuously changing the new geographic position to obtain a plurality of detection circles, preferably, the number of the detection circles is 1000-; wherein the number ratio of the plurality of detection circles divided into the training set and the verification set is 10:1-1: 1.
S3-5, establishing an artificial intelligence model by the multiple detection circles and the searched urban intelligent community site J according to the geographic coordinates of the circle centers of the detection circles under E. Preferably, the AI model includes a convolutional neural network CNN, an SVM support vector machine, and generates a countermeasure network GAN. And obtaining a final AI model through the training model and the model verification.
It can be understood that, because the samples of the model training are samples of whether the person meets the sufficient state in the long-term real-time scene, and are not samples in which various sufficient states are randomly assumed, the obtained AI model can more truly improve the hit finding efficiency. If the sampling is random to assume sufficient state, the amount of training is significantly increased. This is because most of the samples come from three-dimensional models that do not truly reflect the abundant state of the person
Figure 662153DEST_PATH_IMAGE002
Real-time distribution of states.
About S4
The S4 specifically includes:
s4-1, finding the current position by the user through the app, and acquiring the coordinate of the current user under E and acquiring a detection circle O by the app;
s4-2, substituting the coordinates and the detection circle O into the AI model, finding the intelligent community site J and marking the geometric figure with the corresponding color in the three-dimensional model.
About S5
The S5 specifically includes:
when finding out the city smart community site J, the three-dimensional model established in the step S1 can be clicked
Figure 730472DEST_PATH_IMAGE002
The geometric indicia of (a) is communicatively linked to J to allow the person to rush to the fixed point for service.
In one embodiment, after the step S5, the step S5-1 and the step S5-2 of finding the user by navigation are further included before the person reaches the fixed point, which specifically includes:
s5-1, when receiving service information, starting from a corresponding station, identifying a human face and/or clicking the emergency prompt information module when passing through a door in the station, and displaying a navigation route established between a guide point where the current station is located and a road guide point which is closest to the geographical position of the user in a straight line on apps of the personnel and the user, and/or displaying the current position of the personnel starting from the corresponding station;
s5-2 the user prompts the user 'S current location to the person on the app to maintain communication with the person so that the person arrives at the user' S current location accurately;
s5-3 the person uses the navigation route to find the user' S current location.
It will be appreciated that when a person is familiar with a route, he can travel directly to the service site without navigation.
The navigation route establishing method comprises the following steps:
s5-1-1, according to the coordinate position of the index point where the current station building is located, between the coordinate position and the target location which is the road index point with the straight line distance to the geographical position of the user being the nearestEstablishing a straight line segment L, and setting a three-dimensional model needing to pass through
Figure 696154DEST_PATH_IMAGE002
A navigation planning suggested path is established between two adjacent points in the plurality of intermediate points, and the plurality of intermediate points, the representation symbols of the target point and the navigation planning suggested path are displayed in a personnel mobile terminal or a vehicle driven by personnel;
s5-1-2 follows the displayed representation symbol and the navigation plan suggested path to finish the navigation until the target location.
Setting a straight line segment L between the coordinate position and a target position which is a road index point with the closest straight line distance to the geographic position of the user, wherein the straight line segment L needs to pass through the three-dimensional model
Figure 557363DEST_PATH_IMAGE002
The plurality of intermediate locations in (a) specifically include:
s5-1-1-1, sending the geographic coordinates of the user in the input coordinate system E to the server S at the position, and utilizing the server S to perform three-dimensional modeling
Figure 278194DEST_PATH_IMAGE002
Establishing a straight line segment L between said location in (a) and said target location;
s5-1-1-2 equally dividing the straight line segment L into a plurality of straight line segments pi (i =1,2.. k-1, k is a preset number of parts) in a preset number of parts, for each of which the abscissa and ordinate under E are gradually searched for using a plurality of moving points (2 k for the midpoint, 2 (k-1) for the bisector) moving away from the midpoint or the bisector in directions of both sides of L starting from the midpoint thereof or starting with each bisector of the straight line segment L;
s5-1-1-3, continuously calculating the distances of the nearby road index points in the index library K from the plurality of moving points under E obtained by step-by-step search in the step 5-1-1-2, and taking the nearest one as a middle point corresponding to each straight line segment pi or each bisector when the distance is the nearest one.
The two-side direction is preferably two-side directions parallel to the X or Y axis of E, and may also be a direction perpendicular to the straight line segment L, or a multi-direction simultaneous search of two-side directions parallel to the X axis and Y axis of E and three-direction directions perpendicular to the straight line segment L. It will be appreciated that when the straight line segment L is parallel to the X-axis or Y-axis, only two simultaneous searches of the X-axis and Y-axis parallel to E are required. When a multi-directional search is selected, each step of the stepwise search results is an average of the abscissa and ordinate in all directions, and the average includes an arithmetic average or a weighted average.
The invention also provides a city three-dimensional model collaborative management system of the smart community for realizing the method, which is characterized by comprising the following steps: servers S, a monitoring center C, a main monitoring center CC and a server SS of the main monitoring center CC which are arranged in various intelligent community sites, and mobile terminals equipped for personnel, users and staff of the main monitoring center CC in the intelligent community sites, wherein,
the server S is used for receiving the geographic positioning of the user, detecting the circle, providing the intelligent community service information, and receiving the updated and perfect intelligent community service information and three-dimensional model sent by the server SS of the main monitoring center CC
Figure 584542DEST_PATH_IMAGE002
And an AI model;
the monitoring center C is used for receiving the geographical positioning information and the three-dimensional model of the user sent by the server S
Figure 603182DEST_PATH_IMAGE002
And AI model to send site J in three-dimensional model to user mobile terminal
Figure 904851DEST_PATH_IMAGE002
The geometric figure marks the search result, and reports the intelligent community service information and the search result to the main monitoring center CC;
the main monitoring center CC is used for analyzing and processing the historical and real-time data in the servers S and the monitoring center C, and performing analysis and processing on each server S and each monitoring center CCollaborative management, said data including intelligent community service information, service subscription and emergency contact information, data in an intelligent management system, and three-dimensional models
Figure 937529DEST_PATH_IMAGE002
And an AI model;
the server SS of the main monitoring center CC is used for three-dimensional model modeling and AI model modeling and is used for modeling a three-dimensional model Mod in the server S3DUpdating the AI model;
the method comprises the steps that an urban three-dimensional model collaborative management application program app of the smart community is installed in mobile terminals equipped by people and users in the smart community site respectively, preferably, a professional version of the urban three-dimensional model collaborative management application program app of the smart community is installed in the mobile terminal for the people, and a user version of the urban three-dimensional model collaborative management application program app of the smart community is installed in the mobile terminal equipped by the users.
The present invention also provides a computer-readable non-transitory storage medium, in which a program that can implement the above-mentioned city three-dimensional model collaborative management method for a smart community site by the operations of the server S, the server SS, and the monitoring center C, and the master monitoring center CC is stored.
It should be noted that the three-dimensional model according to the present invention may be a remote sensing image map-based model or a semantization model.
The intelligent community city three-dimensional model collaborative management method and system have the advantages that the city three-dimensional model collaborative management method and system are based on the city accurate large three-dimensional model, so that data information, daily management and emergency treatment of all sites become overall and coordinated, online medical treatment and information lookup can be visually carried out through the real-time geographic position of the mobile terminal app, and timely traceable medical rescue emergency service is provided.
Drawings
FIG. 1 is a flow chart of a city three-dimensional model collaborative management method of an old-age care medical and health institution of the invention,
FIG. 2 three-dimensional model of example 1
Figure 278642DEST_PATH_IMAGE002
Wherein, marking the index point of the main building of a family rest home in the city
Figure 914023DEST_PATH_IMAGE005
The index circle with the index points of the buildings as the circle centers of the hos of the hospital opposite to the road is marked, the colors of the index circles of the nursing home and the hos of the hospital are different, and the colors of the index circles are different from the colors of the black index circles of other buildings, and in addition, road index points are marked (one point is a road index point)
Figure 476723DEST_PATH_IMAGE005
) And searching for the index point of the home building of the nursing home used by the nearest hospital nearby
Figure 804936DEST_PATH_IMAGE005
Is a partial circumference with the circle center R as the radius,
FIG. 3a is a schematic diagram of an RNN recurrent neural network algorithm process extracted from an urban road network and an urban road network generation process,
FIG. 3b is a schematic diagram of local road network broadening within the circle of the generated urban road network in FIG. 3a,
FIG. 4 is a flow chart of the extraction of the building index points and the multi-layer RNN building boundaries of the convolutional long short term memory ConvLSTM based on the CNN backbone network,
figure 5a is a schematic diagram of a user version of an app displayed in a user smart phone,
FIG. 5b is an interface schematic of a monitoring center C professional version of the app for hospital hos,
figure 5c shows an interface schematic of a nearby courier site open at the point in the user version app schematic shown in the user smart phone;
FIG. 5d is an interface schematic diagram of a nearby express delivery site opened at the midpoint of a professional edition app of a monitoring center C of Hos in Hospital;
FIG. 6 is a schematic diagram of an interface of professional-version app on a CC staff handset in AI modeling in a total monitoring center,
figure 7 is a schematic diagram of a process for detecting circle formation,
FIG. 8 is an interface diagram showing the navigation route of the smart phone of the user and the current location of the ambulance when the emergency doctor visits thelandmark 4 after the emergency contact with the hospital J,
FIG. 9 City three-dimensional model of straight line L between intermediate location on app of emergency doctor smartphone andindex point 4 as target location
Figure 69564DEST_PATH_IMAGE002
Is shown in (a).
Wherein, the reference numeral, 1 user's smart mobile phone, 2 master control center CC is smart mobile phone for staff, 3AI is the current geographical position of selection or user when modeling, 4 is from AI is the road sign guide point of selection or user's current geographical position straight line distance nearest when modeling, 5AI is the circle of radius R that uses when looking for hospital J when modeling, 6 is the index circle that hospital J building is located, 7 is the index circle that other website building is located, 8 detects the circle, 9 navigation route, 10 emergency doctor or ambulance is located at present position, 11 is different from the caregiver smart mobile phone of another home care hospital of embodiment 1-4, 22 emergency doctor smart mobile phone.
Detailed Description
The following describes a typical embodiment of the intelligent community service in the field of the elderly care.
Example 1
FIG. 1 is a schematic flow chart of a cooperative management method of a city three-dimensional model of an aged care medical and health institution according to the present invention, and FIG. 2 shows an index point of a home building of an aged care hospital
Figure 774215DEST_PATH_IMAGE005
And a hospital hos across the road. In fig. 2, each building index point is marked with an index circle as a circle center, and the color of the index circle of the nursing home and the hos of the hospital is different, and is also different from the color of the black index circle of other buildings. In FIG. 2, the road sign points (one of which is
Figure 191421DEST_PATH_IMAGE006
) And searching for the index point of the home building of the nursing home used by the nearest hospital nearby
Figure 893798DEST_PATH_IMAGE005
Is a partial circumference with the circle center R as the radius.
The method comprises five steps, namely:
s1, setting a server S and a monitoring center C (not shown in figure 2) in the home building of the nursing home and hos of the hospital, setting a main monitoring center CC (not shown in figure 2) and a server SS of the main monitoring center CC in other places in the city, and establishing a three-dimensional model of the urban nursing medical and health institution;
s2, nursing staff in a home building of an old care hospital, family members of the cared old people and emergency doctors in hos of the hospital are all equipped with smart phones provided with urban three-dimensional model collaborative management application apps of the old care medical and health institution; providing an identification device in a hos building gate of a hospital;
s3, establishing an artificial intelligence AI model for searching urban endowment medical and health institutions;
s4, utilizing the AI model established in the step S3 and the app of the step S2 to realize the real-time fixed-point finding of the hos of the hospital in the three-dimensional model established in the step S1 and marking in the three-dimensional model;
s5, when finding out the hos of the hospital, the hos of the hospital can be communicated through the indexing circle of the hos in the three-dimensional model established in the step S1, so that an emergency doctor can rush to the home building of the nursing home to perform medical care on the old.
Wherein the S1 specifically includes:
s1-1, an independent server S and an independent monitoring center C are set in the home building of the nursing home, a main monitoring center CC is set to process and analyze the data of the independent server S and the independent monitoring center C in real time and to cooperatively manage various nursing medical and health institutions, and a server SS of the main monitoring center CC is used for three-dimensional model modeling and AI model modeling.
Example 2
Example 1 further includes after S1-1:
s1-2, establishing an urban road network model of the artificial intelligent network;
s1-3, building a city building network model of the artificial intelligent network;
s1-4, fusing the models established in the steps S1-2 and S1-3 to form a two-dimensional model of the urban road and the building
Figure 521088DEST_PATH_IMAGE001
S1-5 two-dimensional model of urban road and building based on step S1-4
Figure 121440DEST_PATH_IMAGE001
Constructing the three-dimensional model
Figure 48945DEST_PATH_IMAGE002
Wherein, step S1-2 specifically includes:
s1-2-1 establishes a city geographic coordinate system E (as shown in FIG. 2), and the XOY plane represents the ground (the X direction is north).
Referring to FIG. 3a, based on the city remote sensing image, using RNN recurrent neural network algorithm, defining step length l (selected from 1-5m according to the total length of the road) and vector direction r as attribute vector V, and using each initial node and K incident path directions
Figure 594327DEST_PATH_IMAGE007
Figure 912176DEST_PATH_IMAGE008
The points of (A) are used as input points (K initial attribute vectors correspond to K points and the corresponding initial points), K +1 input points and the attribute vector V are input into an encoder, and a decoder generates a new node; in particular for each direction of each starting point
Figure 693050DEST_PATH_IMAGE007
Correspond to the coordinates under E
Figure 334116DEST_PATH_IMAGE009
The attribute vector V corresponding to coordinate increments
Figure 440612DEST_PATH_IMAGE010
WhereintThe sequence number representing the current input point (0 for the start point and 1 for the first new input point), the coordinate and attribute vector V are input to the encoder, and the decoder will emit a new node generated under E
Figure 917861DEST_PATH_IMAGE011
Wherein
Figure 502426DEST_PATH_IMAGE012
Figure 76627DEST_PATH_IMAGE013
. Fig. 3a shows an exemplary road network generation process with 100 node generation cycles at intervals of 20 node generation cycles;
s1-2-2 is a schematic diagram of the local road network widening within the circle in FIG. 3a as shown in FIG. 3 b. And (3) widening the local road network of the graph 3b towards two sides by taking the generated road network line as a central axis according to a preset width w to form a road width line with a certain width, thereby obtaining an urban road network model, wherein the w is 1-1.1 times of the average value of the widths of all roads in the remote sensing image.
Referring to fig. 4, the step S1-3 of establishing the city architectural network model of the artificial intelligence network specifically includes:
s1-3-1, based on the urban remote sensing image in the step S1-2, extracting a series of feature maps obtained by different convolutional layers by using a VGG-16 algorithm without an added layer as a CNN main network, wherein the feature maps are 1/8 of the size of an input image;
meanwhile, a characteristic pyramid is constructed by using different layers of a CNN main network through an image pyramid algorithm FPN, and the frames of a plurality of buildings are predicted,
s1-3-2, for each building in the plurality of buildings, obtaining a local feature map F of the building by using a RoIAlign algorithm on the feature maps obtained by the series of different convolutional layers and the corresponding frame of the building;
s1-3-2, forming a hexagonal boundary cover M by adopting convolutional layer processing on the local feature map F of each building, and forming 6 prediction vertexes P of the boundary cover M by utilizing convolutional layer processing; wherein hexagonal bounding box M refers specifically to the vertical projection of the XOY plane describing the building in E;
s1-3-3, selecting the point with the highest probability in P as the starting point
Figure 714544DEST_PATH_IMAGE003
6-step prediction is carried out by utilizing a multilayer RNN algorithm of convolution long-short term memory ConvLSTM to obtain 6 prediction points
Figure 538143DEST_PATH_IMAGE004
Figure 67345DEST_PATH_IMAGE014
Closed building boundary polygons, forming a city building net model (as shown in figure 2);
s1-3-4 representing index points of the building for intersection of longest and next longest diagonal lines in each polygon boundary
Figure 292790DEST_PATH_IMAGE005
And arranging a road guide point at every preset distance on the road network
Figure 600143DEST_PATH_IMAGE006
Integrating the building index points and the road index points into an index library K;
s1-4 specifically comprises the step of fusing the models established in the steps S1-2 and S1-3 according to the relative coordinate positions of the buildings and the roads in the remote sensing image under the urban geographic coordinate system E to form a two-dimensional model of the urban roads and the buildings
Figure 176618DEST_PATH_IMAGE001
S1-5 presetting the two-dimensional model
Figure 978352DEST_PATH_IMAGE001
The Z coordinate H of the boundary of each polygon in the city is under E and represents the height of the building layer, and the building index points of various medical institutions are marked with index circles in different colors to form a three-dimensional model of the urban endowment medical and health institution
Figure 261566DEST_PATH_IMAGE002
(ii) a The smart phone can inquire the appointed urban old-age medical and health institution through the app and can perform corresponding three-dimensional model
Figure 880766DEST_PATH_IMAGE002
The corresponding index circle is displayed, together with the name of the organization, the geographical coordinates of the location and the name of the road on which the location is located (not shown in fig. 2 and 5a and 5 b).
Example 3
The step S2 specifically includes:
s2-1, a nursing staff in an elderly care hospital is equipped with a smart phone (namely, a user mobile phone), a family member of the cared old man (also belonging to the user mobile phone) and an emergency doctor in the hospital are equipped with the smart phone, and the emergency doctor is equipped with the smart phone and is provided with a professional version of the urban three-dimensional model collaborative management application program app of the elderly care medical and health institution and is provided with a user version of the urban three-dimensional model collaborative management application program app of the elderly care medical and health institution; the app professional edition can register or enter a medical and management system in a corresponding organization, and the app user edition only keeps displaying or enters a medical information interface and realizes the functions of medical treatment registration and call for help contact; then, a main monitoring center CC worker is provided with an intelligent mobile phone for AI modeling;
s2-2, a face recognition device is installed on each access door in the nursing medical and health institution, and a distress call prompt information module is installed on the app, wherein the distress call prompt information module is a distress call contact button in the embodiment.
When an emergency doctor goes out of a hos main building gate of a hospital, the face recognition device recognizes the face of the emergency doctor, the face recognition device indicates that the person goes out of the office and records that a real-time emergency task is reduced, the monitoring center C or the master monitoring center CC sends rescue arrangement and reminding information to the app, and the person clicks a call-for-help contact button to record that the clicked person starts to execute the emergency task, so that a real-time emergency task is reduced in the office; the monitoring center C and the main monitoring center CC store the records and the three-dimensional model Mod3DAnd correspondingly, the medical information updates the record that the emergency doctor has gone out to perform the emergency task, so as to indicate whether the organization is in a state that the staff available for service is sufficient.
It should be emphasized that: clicking the call for help contact button and face recognition of the emergency doctor together form a double insurance means for confirming that the emergency doctor goes out of the hospital hos. If people forget face recognition or the face recognition fails but the people do not pay attention to the recognition again, the outgoing record can be formed by clicking the call for help contact button; or on the contrary, forgetting to click the call for help contact button, but the supplementary recording can also be carried out through the face recognition. And the cases that both are forgotten are few, and the patient is regarded as a medical practice who does not operate according to a professional flow.
Wherein, fig. 5a and 5b respectively show three-dimensional models Mod of apps ofuser version 1 of smartphone and professional version C of app of monitoring center C of hos of hospital, which can be magnified (for the former, the former is realized by an operation mode that two conventional fingers simultaneously slide on the smartphone screen in opposite directions, and for the latter, the latter can be operated by a magnification button of professional version app) by clicking and moving the cursor to the index circles of different colors respectively3DAnd displaying a hos medical information menu interface of the hospital. The top column of the menu is marked with medical information, and the middle two columns of the menu are respectively marked with a reservation registration button and a call for help contact button. Whereas in fig. 5a the user pops up a dialog box (not shown in fig. 5 a) confirming the first aid by pressing the 2s call for help contact button too long; while in fig. 5b the call for help contact button is used on the emergency doctor smartphone, the emergency doctor is also additionally assigned an emergency task by the monitoring center C and the pop-up menu is lit up, at which time the emergency doctor is ready to respond to a call for help in the opposite-side nursing home on the road in fig. 2 by pressing the call for help contact button to indicate that the task is accepted and the emergency doctor leaves the hospital hos.
In fig. 5a the user version of the app, which is the other neighboring health care institution, and in fig. 5b the professional version of the app, which is the medical and management system, show different contents in the last column of the menu. The user and the emergency doctor can enter the system for daily sitting and examining operation by clicking respectively.
Wherein the medical information includes registration information, billing information, CT queuing reminder information, current staff information, hospitalization information, and a view details button for viewing detailed medical information and/or for updating entered medical information in both fig. 5a and fig. 5 b. Wherein the current person information comprises: whether a doctor is currently sitting on a doctor, and after clicking the view details button, the brief introduction of the emergency doctor in the additional hospital hos, the sitting schedule of other hospital personnel and whether to go to perform emergency tasks can be seen. The drug use information in the user edition app comprises information for reminding drug use and a review reminder.
Also shown in fig. 5a and 5b is a nearby courier station with a purple building index circle below the interface. Fig. 5c and 5d show that the user and the person manually click and cursor click on the purple index circle on the express site building respectively on the user-version app and professional-version app three-dimensional models of fig. 5a and 5b, and then pop up detailed button options about the name of the express site, the courier to be sent, the current package to be viewed, and the historical data of the courier to be consulted. In fig. 5a, a menu popped up after a hos main building index circle is clicked on the user version app, and other community service site options are added, so that the user clicks and enters to view other community sites representing other nearby non-hospital hos in the three-dimensional model of the urban intelligent community. The indexing circles of the site buildings are all different colors.
Example 4
The S3 specifically includes:
s3-1, every 1 month, as shown in FIG. 6, an initial current geographic position 3 (marked by a blue circle) is located in the city through asmart phone 2 equipped by a staff member of a general monitoring center CC, and a corresponding three-dimensional model is found in the indexing library K according to the currentgeographic position 3
Figure 573565DEST_PATH_IMAGE002
Theroad index point 4 with the shortest straight line distance;
s3-2, using theroad index point 4 with the shortest straight line distance as the center of circle, in the three-dimensional model
Figure 38044DEST_PATH_IMAGE002
Finding the index circles of different colors representing the construction of the old care medical and health institution and acquiring whether the personnel corresponding to each index circle are in sufficient state or not in thecircle 5 with the preset radius R; if no different color index circle satisfying the sufficient state of the person is found, enlarging the radius R by a specified step size of 100m until found;
s3-3 as shown in fig. 6, in thesmartphone 2 equipped by the staff in the total monitoring center CC, according to the center of the circle and the index circles of different colors in the range of R that satisfy the sufficient status of the staff, the index circle that satisfies the sufficient status of the staff that is closest to the center of the circle is found, and oneindex circle 6 that satisfies the sufficient status of the staff that is closest to the center of the circle and the straight line is marked red and green, respectively, while the other twoindex circles 7 in the range of the circle are all uniformly black (indicating an insufficient staff mechanism, such as a pharmacy or a school health room, and once there is an emergency doctor in at least one of them, they are also marked green), and the other areas in the remaining circles are uniformly yellow, thereby forming a three-color green, black, and yellow detection circle. When there is a suitable emergency doctor in the lower one of the twoindexing circles 7, it turns green, while thegreen indexing circle 6 turns black due to being farther from theindexing point 4 than theindexing circle 7 that turned green.
As shown in fig. 7, the master monitoring center CC staff is equipped with asmartphone 2 with aleft circle 5 obtained in one sampling. Wherein there is a certain currentgeographic position 3, said currentgeographic position 3 finding the corresponding three-dimensional model in said indexing library K
Figure 113448DEST_PATH_IMAGE002
Theroad index point 4 with the shortest straight-line distance therebetween, agreen index circle 6 satisfying the sufficient state of the person, and ablack index circle 7 of the mechanism in which the person is insufficient, the circle areas other than thegreen index circle 6 and theblack index circle 7 are set to yellow, thereby obtaining a three-color detection circle 8 of green, black, and yellow (the right side of fig. 7).
S3-4 urban three-dimensional model
Figure 106812DEST_PATH_IMAGE002
Repeating the steps S3-1-S3-3 while continuously changing the new geographic position to obtain 10000 detection circles;
s3-5, establishing an artificial intelligence model according to the geographic coordinates with the circle center under E, the detection circles and the searched urban endowment medical and health institution J. The AI model convolutional neural network CNN. The number ratio of the detection circles in the training set to the detection circles in the verification set is 5: 1. And obtaining a final AI model through the training model and the model verification.
Example 5
This embodiment is an application scenario of another nursing home in a city and a hospital J located far away from it, unlike the scenarios ofembodiments 1 to 4. In this scenario:
the S4 specifically includes:
s4-1, as shown in fig. 8, the caregiver in the nursing home finds thecurrent position 3 through the app of thesmart phone 11, and the app obtains the coordinate of thecurrent position 3 of the current user under E and obtains a detection circle O (not shown in the figure, but acircle center 4 is given, i.e. a road guidance point where the straight distance from theposition 3 of the previous user is the shortest);
s4-2, the coordinates and the detection circle O are sent to the aged care server S to be substituted into the AI model, a hospital J is found, and a green index circle J is marked in the three-dimensional model.
The S5 specifically includes:
when finding the hospital J, the three-dimensional model established by the step S1 can be clicked
Figure 126720DEST_PATH_IMAGE002
Green in the index circle J, or first enlarging the three-dimensional model
Figure 50683DEST_PATH_IMAGE002
Finding out the enlarged three-dimensional model
Figure 105226DEST_PATH_IMAGE002
Clicking the three-dimensional model established in step S1
Figure 676016DEST_PATH_IMAGE002
The green index circle J in the middle pops up a menu as shown in fig. 5a, and the emergency doctor can communicate with the hospital J by pressing the call for help contact button and selecting and determining in the popped dialog box, so as to lead the emergency doctor to rush to the fixed point for medical rescue service.
Example 6
Inembodiment 5, after the step S5 of communicating with the user to reach the hospital J, the method further includes steps S5-1 and S5-2 of finding the user by navigation, including:
s5-1, when receiving the rescue information from the monitoring center C, the emergency medical practitioner starts from hospital J, recognizes the face and clicks the call for help contact button (clicking on the emergency medical practitioner 'S smart phone or on the emergency medical practitioner' S desktop computer) when passing through the door of the institution, and both the caregiver in the nursing home and the app of the emergency medical practitioner 'S smart phone can display thenavigation route 9 established between the index point where the current hospital J is located and theroad index point 4 that is the closest to the caregiver' Scurrent position 3 in a straight line, and display the current location 10 (indicated by a five-pointed star) where the emergency medical practitioner starts from hospital J (see fig. 8);
s5-2 prompting the caregivercurrent location 3 of the emergency medical practitioner on the caregiver smartphone app so that the user remains in communication with the personnel so that the personnel arrive exactly at the caregivercurrent location 3;
s5-3 the emergency medical practitioner utilizes thenavigation route 9 to find the caregiver' Scurrent location 3 using an ambulance.
Example 7
The method for establishing thenavigation route 9 in fig. 8 comprises the following steps:
as shown in fig. 9, S5-1-1 establishes a straight line segment L between the coordinate position of the index point where the current hospital J main building is located and a target point that is aroad index point 4 closest to the current position 3 (see fig. 8) of the caregiver and sets a three-dimensional model to be passed through
Figure 652062DEST_PATH_IMAGE002
In two ofAn intermediate location for which a navigation planning suggestedpath 9 is established between the two intermediate locations and which may be displayed in the app of theemergency doctor smartphone 22 or in the ambulance together with the representation of theindex point 4 as the target location and the navigation planning suggested path; wherein a and b are two mutually perpendicular coordinate axes of a coordinate system F established on the ambulance;
s5-1-2 follows the displayed representation symbol and the navigation plan suggested path to finish the navigation until the target location.
Wherein, a straight line segment L between the coordinate position and a target position which is aroad index point 4 with the closest straight line distance to thegeographic position 3 of the user is set to pass through the three-dimensional model
Figure 723924DEST_PATH_IMAGE002
Specifically, the two intermediate locations include:
s5-1-1-1, sending the geographic coordinates of the user in the input coordinate system E to the server S at the position, and utilizing the server S to perform three-dimensional modeling
Figure 524652DEST_PATH_IMAGE002
A straight line segment L is established between the road markingguide point 4 and a marking guide point where a hospital J building is located;
s5-1-1-2, equally dividing the straight line segment L into 3 straight line segments in preset number, starting from each equally dividing point of the straight line segment L, searching equally dividing points close to theindex point 4 at intervals of 2.5m in the directions of two sides parallel to the X axis of the E, and searching equally dividing points far away from theindex point 4 at intervals of 2.5m in the directions of two sides parallel to the Y axis of the E for each straight line segment;
s5-1-1-3 calculates the moving points under E obtained by the step-by-step search in step 5-1-1-2, and continuously calculates the distances between the neighboring road index points in the index library K, and when the distance is the closest, the distance is used as the intermediate points D and G (shown in fig. 9) corresponding to each bisector.
Example 8
An urban three-dimensional model collaborative management system for an endowment medical and health institution, which implements the method of theabove embodiments 1 to 6, comprising: the system comprises a server S, a monitoring center C, a main monitoring center CC, a server SS of the main monitoring center CC, and smart phones equipped for personnel in the nursing medical and health institutions, users and staff of the main monitoring center CC in various nursing medical and health institutions, wherein,
the server S is used for receiving the geographic positioning of the user, detecting the circle, providing medical information and receiving updated and perfected medical information and three-dimensional models sent by the server SS of the main monitoring center CC
Figure 656556DEST_PATH_IMAGE002
And an AI model;
the monitoring center C is used for receiving the geographical positioning information and the three-dimensional model of the user sent by the server S
Figure 854319DEST_PATH_IMAGE002
And AI model and send mechanism J in three-dimensional model to user smart phone
Figure 870816DEST_PATH_IMAGE002
The geometric figure marks the search result, and reports the medical information and the search result to the main monitoring center CC;
the main monitoring center CC is used for analyzing and processing the historical and real-time data in the servers S and the monitoring center C and cooperatively managing the servers S and the monitoring center C, wherein the data comprises medical information, information of medical treatment registration and call for help, data in a medical treatment and management system and a three-dimensional model
Figure 899952DEST_PATH_IMAGE002
And an AI model;
the server SS of the main monitoring center CC is used for three-dimensional model modeling and AI model modeling and is used for modeling a three-dimensional model Mod in the server S3DUpdating the AI model;
and an urban three-dimensional model collaborative management application program app of the endowment medical and health institution is respectively installed in the smart phones equipped by the personnel and the users in the endowment medical and health institution. The personnel installs a professional version of the urban three-dimensional model collaborative management application app of the endowment medical and health institution in the smart phone, installs a user version of the urban three-dimensional model collaborative management application app of the endowment medical and health institution in the smart phone equipped by the user, and is equipped by a main monitoring center CC.

Claims (13)

1. A city three-dimensional model collaborative management method of a smart community is characterized by comprising the following steps:
s1, a server S and a monitoring center C are set in the smart community site, a main monitoring center CC and a server SS of the main monitoring center CC are set, and a city three-dimensional model of the city smart community is set;
s2, arranging mobile terminals provided with the urban three-dimensional model collaborative management application program app of the smart community for personnel in the smart community site and users outside the site; providing a device capable of recording whether the person is currently located in the intelligent community site, wherein the device is used for indicating whether the site is in a state that the person capable of providing service is sufficient;
s3, establishing an artificial intelligence AI model for searching the urban intelligent community site;
s4, searching for smart community sites at real time and fixed points in the three-dimensional model established in the step S1 by using the AI model established in the step S3 and the app established in the step S2, and marking the smart community sites in the three-dimensional model;
the three-dimensional model is a remote sensing image map-based model or a semantic model.
2. The method according to claim 1, wherein step S4 is followed by further comprising:
when finding the site J of the urban intelligent community, the S5 can contact the site J through the three-dimensional model communication established in the step S1, so that the person can attend to the fixed point for service.
3. The method according to claim 2, wherein the S1 specifically includes:
s1-1, setting an independent server S and an independent monitoring center C in each intelligent community site, setting a main monitoring center CC to perform real-time processing and analysis on data of the independent server S and the independent monitoring center C, and performing cooperative management on each intelligent community site, wherein a server SS of the main monitoring center CC is used for three-dimensional model modeling and AI model modeling;
s1-2, establishing an urban road network model of the artificial intelligent network;
s1-3, building a city building network model of the artificial intelligent network;
s1-4, fusing the models established in the steps S1-2 and S1-3 to form a two-dimensional model of the urban road and the building
Figure 193099DEST_PATH_IMAGE001
S1-5 two-dimensional model of urban road and building based on step S1-4
Figure 143737DEST_PATH_IMAGE001
Constructing the unified three-dimensional model
Figure 557401DEST_PATH_IMAGE002
Wherein, step S1-2 specifically includes:
s1-2-1, establishing an urban geographic coordinate system E, wherein an XOY plane represents the ground, generating road continuous nodes by a node generator comprising an encoder and a decoder by utilizing an RNN recurrent neural network algorithm based on urban remote sensing images, connecting the two nodes before and after generation in the generation process, inputting the new generated nodes into the node generator to continuously generate new nodes, continuously connecting the generated new nodes, and circularly connecting the nodes to form a road network;
s1-2-2, widening all lines in a road network according to a preset width w to form road width lines with a certain width, and thus obtaining an urban road network model, wherein w is widened according to the corresponding road width in the remote sensing image, and is 0.5-1.5 times of the average value of the widths of all roads in the remote sensing image;
step S1-3 specifically includes:
s1-3-1, based on the urban remote sensing image in the step S1-2-1, extracting a series of feature maps obtained by different convolutional layers by using a VGG-16 algorithm without an added layer as a CNN main network, wherein the feature maps are 1/2-1/10 of the size of an input image;
meanwhile, a feature pyramid is constructed by using different layers of a CNN backbone network through an image pyramid algorithm FPN, and the borders of a plurality of buildings are predicted, and S1-3-2, for each building in the plurality of buildings, a local feature map F of the building is obtained by using a RoIAlign algorithm for feature maps obtained by the series of different convolution layers and the corresponding border of the building;
s1-3-2, forming a polygonal boundary cover M by adopting convolution layer processing on the local feature map F of each building, and then forming a plurality of predicted vertexes P of the boundary cover M by utilizing convolution layer processing; wherein polygonal bounding box M refers specifically to the vertical projection of the XOY plane describing the building in E; s1-3-3, selecting the point with the highest probability in P as the starting point
Figure 332721DEST_PATH_IMAGE003
Performing multi-step prediction by using a multi-layer RNN algorithm of convolution long-short term memory ConvLSTM to obtain multiple prediction points
Figure 540849DEST_PATH_IMAGE004
(t is step number) closed building boundary polygons to form an urban building network model;
s1-3-4 representing index points of the building for intersection of longest and next longest diagonal lines in each polygon boundary
Figure 713204DEST_PATH_IMAGE005
And arranging a road guide point at every preset distance on the road network
Figure 805925DEST_PATH_IMAGE006
And integrating the building index points and the road index points into an index library K.
4. The method according to claim 3, characterized in thatStep S1-4 specifically includes: under the urban geographic coordinate system E, according to the relative coordinate positions of the buildings and the roads in the remote sensing image, the models established in the steps S1-2 and S1-3 are fused to form a two-dimensional model of the urban roads and the buildings
Figure 544074DEST_PATH_IMAGE001
S1-5 specifically comprises: presetting the two-dimensional model
Figure 578895DEST_PATH_IMAGE001
Representing the height of the building layer by the polygon boundaries in the Z coordinate H under E, marking the building index points of various intelligent community sites with geometric figures with different colors to form a three-dimensional model of the urban intelligent community site
Figure 35284DEST_PATH_IMAGE002
(ii) a The mobile terminal can inquire a specified urban smart community site through the app and can display a corresponding three-dimensional model
Figure 666117DEST_PATH_IMAGE002
Displaying corresponding geometric figure marks and site names, geographic coordinates of the sites and names of roads of the sites; the mobile terminal, the monitoring center C of various smart community sites and the main monitoring center CC interface can be manually clicked respectively, and a cursor moves to the geometric figure marks with different colors or is clicked by the cursor to display and/or enter the smart community service information interface, realize a service reservation function, realize emergency contact by long-time pressing of the emergency contact function, and/or register or enter a smart management system in a corresponding site or view other smart community sites nearby, wherein the long-time pressing is 2-4 seconds, and a confirmation dialog box pops up.
5. The method according to claim 3 or 4, wherein the step S2 specifically comprises:
s2-1, arranging mobile terminals for personnel in the smart community site and staff of a main monitoring center CC, installing a professional version of a city three-dimensional model collaborative management application program app of the smart community in the mobile terminals, arranging mobile terminals for users, and installing a user version of the city three-dimensional model collaborative management application program app of the smart community in the mobile terminals; the app professional version can register or enter an intelligent management system in a corresponding site, and the app user version only keeps displaying or enters an intelligent community service information interface, so that the functions of service reservation and emergency contact are realized, and other intelligent community sites nearby are checked; wherein, the CC staff of the main monitoring center is equipped with a mobile terminal for AI modeling;
s2-2, a face recognition device is installed on each entrance door in the smart community site, when people go out, the people are shown to have gone out of the site through face recognition, and people with a name of real-time emergency tasks are recorded and reduced, and/or an emergency prompt information module is installed on the app, rescue arrangement and prompt information are sent to the app through a monitoring center C or a main monitoring center CC, and the people click the emergency prompt information module to record that the clicked people start to execute the emergency tasks, so that people with a name of real-time emergency tasks are reduced in the site; the monitoring center C and the main monitoring center CC store the records and the three-dimensional model Mod3DAnd correspondingly, updating the record of whether the travel executes the emergency task in the intelligent community service information so as to indicate whether the site is in a state of sufficient personnel capable of providing service.
6. The method according to claim 3, wherein the S3 specifically comprises:
s3-1, positioning an initial current geographic position by the aid of a mobile terminal equipped by staff of the CC at preset intervals, and finding out a corresponding three-dimensional model in the indexing library K according to the current geographic position
Figure 727614DEST_PATH_IMAGE002
The road index point with the shortest straight line distance to the road index point, theThe predetermined time is between 1 day and 1 month;
s3-2, using the road sign guide point with the shortest straight line distance as the center of circle, in the three-dimensional model
Figure 808702DEST_PATH_IMAGE002
Finding the geometric figure marks with different colors representing the smart community site buildings in a circle with a preset radius R and acquiring whether the personnel corresponding to each mark are in sufficient state; if no different color geometric figure mark is found which satisfies the sufficient state of the person, enlarging the radius R by a specified step length until the found; the specified step length is 50-5000 m;
s3-3, according to the circle center and the geometric figure marks with different colors meeting the sufficient state of people in the R range, finding the geometric figure mark which is closest to the circle center in a straight line distance and meets the sufficient state of people, and marking the geometric figure mark which is closest to the straight line distance and meets the sufficient state of people as a first color or gray scale, wherein the geometric figure marks which represent other intelligent community buildings with different colors in the circle range are unified as a second color or gray scale, and other areas in the remaining circles are unified as a third color or gray scale, so that a detection circle is formed;
s3-4 urban three-dimensional model
Figure 109977DEST_PATH_IMAGE002
Repeating the steps S3-1-S3-3 while continuously changing the new geographic position to obtain a plurality of detection circles, wherein the number of the detection circles is preferably 1000-; wherein the number ratio of the plurality of detection circles divided into the training set and the verification set is 10:1-1: 1;
s3-5, establishing an artificial intelligence model according to the geographic coordinates of the circle center of each detection circle under E, the detection circles and the found urban intelligent community site J, wherein the AI model comprises a convolutional neural network CNN and an SVM (support vector machine), generating a countermeasure network GAN, and obtaining a final AI model through training model and model verification.
7. The method according to claim 5, wherein the S4 specifically includes:
s4-1, finding the current position by the user through the app, and acquiring the coordinate of the current user under E and acquiring a detection circle O by the app;
s4-2, substituting the coordinates and the detection circle O into the AI model, finding an intelligent community site J and marking the intelligent community site J with the geometric figure with a corresponding color in the three-dimensional model;
the S5 specifically includes:
when finding out the city smart community site J, the three-dimensional model established in the step S1 can be clicked
Figure 669134DEST_PATH_IMAGE002
The geometric indicia of (a) is communicatively linked to J to allow the person to rush to the fixed point for service.
8. The method of claim 7,
after the step S5, the step S5-1 and the step S5-2 of finding the user by navigation are further included before the person reaches the fixed point, which specifically includes:
s5-1, when receiving rescue information, starting from a corresponding station, identifying a human face and/or clicking the emergency prompt information module when passing through a door in the station, and displaying a navigation route established between a guide point where the current station is located and a road guide point which is closest to the current position of the user in a straight line manner on the app of the personnel and the user, and/or displaying the current position of the personnel starting from the corresponding station;
s5-2, the user prompts the current position of the user to the personnel on the app so as to keep communication with the personnel, so that the personnel can accurately reach the geographical position of the user;
s5-3 the person uses the navigation route to find the user' S current location.
9. The method of claim 8, wherein the navigation route is established by:
s5-1-1, according to the coordinate position of the index point where the current station building is located, establishing a straight line segment L between the coordinate position and a target location which is a road index point with the straight line distance closest to the geographic position of a user, and setting a three-dimensional model to be passed by
Figure 585138DEST_PATH_IMAGE002
A navigation planning suggested path is established between two adjacent points in the plurality of intermediate points, and the plurality of intermediate points, the representation symbols of the target point and the navigation planning suggested path are displayed in a personnel mobile terminal or a vehicle driven by personnel;
s5-1-2 follows the displayed representation symbol and the navigation plan suggested path to finish the navigation until the target location.
10. The method according to claim 9, wherein a straight line segment L between the coordinate position and a target position as a road index point having a straight line distance closest to the user's geographic position is set to pass through the three-dimensional model
Figure 446915DEST_PATH_IMAGE002
The plurality of intermediate locations in (a) specifically include:
s5-1-1-1 sends the geographic coordinates of the user in the input coordinate system E to the server S at the position, and the server S is used for the three-dimensional model Mod3DEstablishing a straight line segment L between said location in (a) and said target location;
s5-1-1-2 equally dividing the straight line segment L into a plurality of straight line segments pi (i =1,2.. k-1, k is a preset number of parts) in a preset number of parts, for each of the straight line segments pi, gradually searching for an abscissa and an ordinate under E using two moving points moving away from the midpoint or the bisector in directions on both sides of L, starting from a midpoint thereof, or starting using each bisector of the straight line segment L;
s5-1-1-3, continuously calculating the moving point under E obtained by the step-by-step search in the step 5-1-1-2 to obtain the distance of the nearby road index point in the index library K, and when the distance is the nearest point, using the nearest point as a middle point corresponding to each straight line segment pi or each equally divided point;
the two-side directions are preferably two-side directions parallel to the X or Y axis of the E, and can also be directions perpendicular to the straight line segment L, or multi-directional simultaneous search of the two-side directions parallel to the X axis and the Y axis of the E and three directions perpendicular to the direction of the straight line segment L; when a multi-directional search is selected, each step of the stepwise search results is an average of the abscissa and ordinate in all directions, and the average includes an arithmetic average or a weighted average.
11. A city three-dimensional model collaborative management system for smart communities implementing the method according to any one of claims 1-10, comprising: servers S, a monitoring center C, a main monitoring center CC and a server SS of the main monitoring center CC which are arranged in various intelligent community sites, and mobile terminals equipped for personnel, users and staff of the main monitoring center CC in the intelligent community sites, wherein,
the server S is used for receiving the geographic positioning of the user, detecting the circle, providing the intelligent community service information, and receiving the updated and perfect intelligent community service information and three-dimensional model sent by the server SS of the main monitoring center CC
Figure 143475DEST_PATH_IMAGE002
And an AI model;
the monitoring center C is used for receiving the geographical positioning information and the three-dimensional model Mod of the user sent by the server S3DAnd AI model to send site J in three-dimensional model to user mobile terminal
Figure 365378DEST_PATH_IMAGE002
The geometric figure marks the search result, and reports the intelligent community service information and the search result to the main monitoring center CC;
the main monitoring center CC is used for analyzing and processing the historical and real-time data in the servers S and the monitoring center C, and for each server S and each monitoring centerC, performing cooperative management, wherein the data comprises intelligent community service information, service reservation and emergency contact information, data in an intelligent management system, and a three-dimensional model
Figure 932625DEST_PATH_IMAGE002
And an AI model;
the server SS of the general monitoring center CC is used for three-dimensional model modeling and AI model modeling and carries out three-dimensional model modeling on the server S
Figure 965304DEST_PATH_IMAGE002
Updating the AI model;
and the urban three-dimensional model collaborative management application program app of the smart community is respectively installed in the mobile terminals equipped by the personnel and the users in the smart community site.
12. The system of claim 11, wherein the personnel uses a professional version of a smart community city three-dimensional model collaborative management application app installed in the mobile terminal, and a user version of the smart community city three-dimensional model collaborative management application app installed in the user-equipped mobile terminal.
13. A computer-readable non-transitory storage medium in which a program operable by the server S, the server SS, and the monitoring center C, and the total monitoring center CC to implement the method of any one of claims 1 to 10 is stored.
CN202110651680.3A2021-06-112021-06-11City three-dimensional model collaborative management method and system for smart communityActiveCN113435833B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202110651680.3ACN113435833B (en)2021-06-112021-06-11City three-dimensional model collaborative management method and system for smart community

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202110651680.3ACN113435833B (en)2021-06-112021-06-11City three-dimensional model collaborative management method and system for smart community

Publications (2)

Publication NumberPublication Date
CN113435833Atrue CN113435833A (en)2021-09-24
CN113435833B CN113435833B (en)2022-05-03

Family

ID=77755607

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202110651680.3AActiveCN113435833B (en)2021-06-112021-06-11City three-dimensional model collaborative management method and system for smart community

Country Status (1)

CountryLink
CN (1)CN113435833B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN113791195A (en)*2021-11-162021-12-14西南交通大学Device and method for obtaining landslide retaining structure test data
CN116150290A (en)*2022-05-072023-05-23西宁市测绘院 A method for attaching business attributes to three-dimensional data of geographical entities based on center point forced offset operation

Citations (12)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
US20160192315A1 (en)*2014-12-312016-06-30Line CorporationApparatuses, methods, and recording mediums for providing location sharing services
CN106558012A (en)*2016-11-162017-04-05四川省亚丁胡杨人力资源集团有限公司A kind of health services system based on intelligent Community
CN106558011A (en)*2016-11-162017-04-05四川省亚丁胡杨人力资源集团有限公司A kind of intelligent Community service system
CN109067822A (en)*2018-06-082018-12-21珠海欧麦斯通信科技有限公司The real-time mixed reality urban service realization method and system of on-line off-line fusion
CN109064482A (en)*2018-08-162018-12-21武大吉奥信息技术有限公司A kind of method and device obtaining outline of house automatically in three-dimensional tilt photographed data scene
CN109087230A (en)*2018-08-132018-12-25河南易商谷电子商务有限公司A kind of intelligence community service system and its application method
CN109635966A (en)*2019-01-082019-04-16博拉网络股份有限公司A kind of intelligence community modularization management system
US20190180501A1 (en)*2017-12-112019-06-13Locus Social Inc.System and Method of Highly-Scalable Mapping and 3D terrain Modeling with Aerial Images
CN111126166A (en)*2019-11-302020-05-08武汉汉达瑞科技有限公司Remote sensing image road extraction method and system
CN111259797A (en)*2020-01-162020-06-09南开大学Iterative remote sensing image road extraction method based on points
CN112712592A (en)*2021-03-262021-04-27泰瑞数创科技(北京)有限公司Building three-dimensional model semantization method
CN113435834A (en)*2021-06-112021-09-24泰瑞数创科技(北京)有限公司City three-dimensional model collaborative management method and system for community medical emergency organization

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
US20160192315A1 (en)*2014-12-312016-06-30Line CorporationApparatuses, methods, and recording mediums for providing location sharing services
CN106558012A (en)*2016-11-162017-04-05四川省亚丁胡杨人力资源集团有限公司A kind of health services system based on intelligent Community
CN106558011A (en)*2016-11-162017-04-05四川省亚丁胡杨人力资源集团有限公司A kind of intelligent Community service system
US20190180501A1 (en)*2017-12-112019-06-13Locus Social Inc.System and Method of Highly-Scalable Mapping and 3D terrain Modeling with Aerial Images
CN109067822A (en)*2018-06-082018-12-21珠海欧麦斯通信科技有限公司The real-time mixed reality urban service realization method and system of on-line off-line fusion
CN109087230A (en)*2018-08-132018-12-25河南易商谷电子商务有限公司A kind of intelligence community service system and its application method
CN109064482A (en)*2018-08-162018-12-21武大吉奥信息技术有限公司A kind of method and device obtaining outline of house automatically in three-dimensional tilt photographed data scene
CN109635966A (en)*2019-01-082019-04-16博拉网络股份有限公司A kind of intelligence community modularization management system
CN111126166A (en)*2019-11-302020-05-08武汉汉达瑞科技有限公司Remote sensing image road extraction method and system
CN111259797A (en)*2020-01-162020-06-09南开大学Iterative remote sensing image road extraction method based on points
CN112712592A (en)*2021-03-262021-04-27泰瑞数创科技(北京)有限公司Building three-dimensional model semantization method
CN113435834A (en)*2021-06-112021-09-24泰瑞数创科技(北京)有限公司City three-dimensional model collaborative management method and system for community medical emergency organization

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
MAO, ZHU等: "Deep Neural Networks for Road Sign Detection and Embedded Modeling Using Oblique Aerial Images", 《REMOTE SENSING》*
刘俊伟: "基于三维地理信息系统的移动数字城市方案", 《城市科学管理》*
李皓昱: "遥感影像道路自动提取系统组件化设计与实现", 《中国优秀硕士学位论文全文数据库(基础科技辑)》*

Cited By (4)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN113791195A (en)*2021-11-162021-12-14西南交通大学Device and method for obtaining landslide retaining structure test data
CN113791195B (en)*2021-11-162022-02-08西南交通大学Device and method for obtaining landslide retaining structure test data
CN116150290A (en)*2022-05-072023-05-23西宁市测绘院 A method for attaching business attributes to three-dimensional data of geographical entities based on center point forced offset operation
CN116150290B (en)*2022-05-072023-09-08西宁市测绘院 Method for attaching business attributes to 3D data of geographical entities based on forced offset operation of center point

Also Published As

Publication numberPublication date
CN113435833B (en)2022-05-03

Similar Documents

PublicationPublication DateTitle
BoulosTowards evidence-based, GIS-driven national spatial health information infrastructure and surveillance services in the United Kingdom
US8725526B2 (en)Methods, systems, and apparatus for providing real time query support and graphical views of patient care information
Tavares et al.Hefestos: an intelligent system applied to ubiquitous accessibility
US20120112883A1 (en)Disease Mapping and Infection Control System and Method
CN113435833B (en)City three-dimensional model collaborative management method and system for smart community
CN107710330A (en)Context aware looks after flow engine, platform, device, system, method and computer-readable medium
MuradCreating a GIS application for local health care planning in Saudi Arabia
Bitala et al.Predictive Analysis Towards Integration of Urban Planning and GIS to Manage Health Care Organization
Mériade et al.Integrated care pathway for breast cancer: a relational and geographical approach
McLaffertyPlacing substance abuse: Geographical perspectives on substance use and addiction
Nik Khah et al.Route analysis in the architecture of museums and tomb buildings through space syntax case study:(Tomb of Nader Shah in Mashhad, Avicenna Mausoleum in Hamadan, and Mausoleum of poets in Tabriz)
Searcy et al.Reaching Latinx communities with algorithmic optimization for SARS-CoV-2 testing locations
CN113435834B (en)City three-dimensional model collaborative management method and system for community medical emergency organization
Zhang et al.Walking in the cities without ground, how 3d complex network volumetrics improve analysis
LyuExploring the influence of dynamic indicators in urban spaces on residents’ environmental behavior: A case study in Shanghai utilizing mixed-methods approach and artificial neural network (ANN) modeling
Bilal et al.CitySEIRCast: an agent-based city digital twin for pandemic analysis and simulation
GaoAdvanced health information sharing with web-based GIS
AlattasThe Integration of LADM and IndoorGML to Support the Indoor Navigation Based on the User Access Rights
Samuel et al.A composed index of critical accessibility (CICA) to healthcare services in a traditional African City
Jayaraman et al.An ontology-based framework for real-time collection and visualization of mobile field triage data in mass gatherings
Smith et al.Can you model growth of trust? A study of the sustainability of a rural community health centre in North India
Hirshorn et al.Geographic information systems in community-based gerontological research and practice
Hooda et al.Advancement of Geospatial Technology in Healthcare Systems
Santosa et al.Value of smart city services in improving the quality of life: A literature review
Al-DouriImpact of utilizing 3D digital urban models on the design content of urban design plans in US cities

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant
CP03Change of name, title or address

Address after:22 / F, building 683, zone 2, No. 5, Zhongguancun South Street, Haidian District, Beijing 100086

Patentee after:Terry digital technology (Beijing) Co.,Ltd.

Address before:100190 22 / F, building 683, zone 2, 5 Zhongguancun South Street, Haidian District, Beijing

Patentee before:Terra-IT Technology (Beijing) Co.,Ltd.

CP03Change of name, title or address

[8]ページ先頭

©2009-2025 Movatter.jp