CN112381014A - Illegal parking vehicle detection and management method and system based on urban road - Google Patents
Illegal parking vehicle detection and management method and system based on urban roadDownload PDFInfo
- Publication number
- CN112381014A CN112381014ACN202011298904.9ACN202011298904ACN112381014ACN 112381014 ACN112381014 ACN 112381014ACN 202011298904 ACN202011298904 ACN 202011298904ACN 112381014 ACN112381014 ACN 112381014A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- parking
- vehicles
- illegal parking
- illegal
- 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.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
Abstract
Description
Technical Field
The invention relates to the field of image processing, in particular to a method and a system for detecting and managing illegal parking vehicles based on urban roads.
Background
Along with the economic development and the improvement of the living standard of people, the number of motor vehicles is increased sharply, and the phenomenon of illegal parking on the road surface is more and more. Illegal parking refers to that vehicles stay in road sections or parking prohibited areas provided with parking prohibited signs and marked lines, and the behaviors not only can seriously interfere normal traffic operation, but also can possibly cause damage to lives and properties of people, so that the illegal parking phenomenon needs to be managed and controlled.
The current common mode of managing and controlling illegal parking is that a camera is arranged in an illegal parking area, the camera acquires images of the illegal parking area and transmits the images to a background for manual processing or computer processing, the mode of computer processing can greatly save labor cost compared with the mode of manual processing, illegal vehicles can be found more efficiently, and therefore the method can be more widely applied to the field of traffic control.
However, the existing illegal parking detection technology of the auxiliary image processing technology has some defects: the illegal parking detection is not comprehensive and accurate enough, and whether the illegal parking is judged by matching the vehicle detection technology with the parking time statistics is judged only, so that the temporary parking is easily judged as the illegal parking by mistake; in addition, the existing technology only records the illegal parking vehicles and carries out electronic ticket processing, illegal parking data are not analyzed, and the actual illegal parking phenomenon is not relieved.
Disclosure of Invention
The invention aims to provide a method and a system for detecting and managing illegal parking vehicles based on urban roads, which can distinguish and detect illegal parking vehicles in an illegal parking area and a temporary parking area, and when the illegal parking vehicles in the temporary parking area are detected, a face detection technology is adopted to be matched with a parking duration judgment technology for detection, so that the phenomenon that the temporary parking is wrongly judged as an illegal parking situation is reduced; in addition, the management method can perform systematic management and control on the illegal parking vehicles, accurately identify vehicles and places with high illegal parking times, and can perform management and control in a targeted manner.
In order to achieve the above object, a first aspect of the present disclosure provides a method for detecting illegal parking vehicles based on urban roads, including the following steps: acquiring monitoring video images of continuous frames and selecting a monitoring area; detecting vehicle coordinate data and license plate numbers of vehicles in the monitoring area in the monitoring video images of the continuous frames, wherein if the deviation value of the vehicle coordinate data of the vehicles corresponding to the same license plate number is smaller than a position threshold value and a no-parking area corresponding to the monitoring area, the vehicles are illegal parking vehicles; if the temporary parking area corresponding to the monitoring area is used, carrying out face detection and parking time counting on the vehicle, if the conditions that the face is detected and the parking time is smaller than a time threshold value are met, determining that the vehicle is a temporary parking vehicle, and if not, determining that the vehicle is an illegal parking vehicle.
A second aspect of the present disclosure provides a method for managing illegal vehicles on an urban road, including the following steps: acquiring monitoring video images of continuous frames, and detecting whether illegal vehicles exist by using the illegal vehicle detection method based on the urban road; acquiring illegal parking data of illegal parking vehicles, analyzing the illegal parking data, and acquiring the vehicles with the most illegal parking times and the areas with the most illegal parking vehicles, wherein the illegal parking data comprises dates, illegal parking addresses, parking time of the illegal parking vehicles, and license plate numbers
The third aspect of the scheme provides a system for detecting illegal vehicles based on the urban road, which is used as a carrier for operating a method for detecting illegal vehicles based on the urban road.
The fourth aspect of the scheme provides an urban road-based illegal parking vehicle management system, which is used as a carrier for operating an urban road-based illegal parking vehicle management method.
Compared with the prior art, the scheme has the following characteristics and beneficial effects: 1. distinguishing a forbidden parking area and a temporary parking area, acquiring vehicle central point coordinate data and vehicle size data by using a vehicle detection model aiming at vehicles in the forbidden parking area, identifying the vehicles by using a license plate identification technology, judging and identifying illegal parking vehicles in the forbidden parking area based on judging whether an offset value of the vehicle central point coordinate data of the vehicles with the same license plate number is greater than a threshold value, and timely discovering illegal parking behaviors in the forbidden parking area in this way; and aiming at the temporary parking area, face detection is carried out through a face detection model, whether a person exists on the vehicle in the vehicle stopping state is judged, so that whether the vehicle is temporarily parked or not is judged, and the phenomenon that the temporary parking is mistakenly judged as an illegal parking condition is reduced. 2. And acquiring the date of the illegal parking vehicle, the illegal parking address, the parking time and the license plate number, and managing big data and the like so as to acquire regions and vehicles with high illegal parking times and serious conditions.
Drawings
Fig. 1 is a schematic flow chart of a method for urban road-based illegal vehicle detection according to an embodiment of the present invention.
Fig. 2 is a flowchart illustrating a method for urban road-based illegal vehicle management according to an embodiment of the present invention.
Fig. 3 is a schematic operation logic diagram of the urban road-based illegal parking vehicle detection method according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments that can be derived by one of ordinary skill in the art from the embodiments given herein are intended to be within the scope of the present invention.
The scheme provides a method for detecting and managing illegal parking vehicles based on urban roads, which can be used for distinguishing a forbidden parking area and a temporary parking area in a targeted manner so as to obtain more prepared illegal parking vehicle data, and can be used for carrying out big data analysis on the illegal parking vehicle data so as to carry out subsequent management and control on the illegal parking vehicles.
In a first aspect of the present disclosure, a method for detecting illegal parking vehicles based on urban roads is provided, which includes the following steps
Acquiring monitoring video images of continuous frames and selecting a monitoring area;
if the monitoring area is a no-parking area, detecting vehicle coordinate data and license plate numbers of vehicles in the monitoring area in the continuous frame monitoring video images, and if the deviation value of the vehicle coordinate data of the vehicles corresponding to the same license plate number is smaller than a position threshold value, determining that the vehicles are illegal-parking vehicles;
if the monitoring area is a temporary parking area, detecting the vehicle coordinate data and the license plate number of the vehicle in the monitoring area in the continuous frame of monitoring video images, if the deviation value of the vehicle coordinate data of the vehicle corresponding to the same license plate number is smaller than a position threshold value, carrying out face detection and parking duration statistics on the vehicle, and if the conditions that the face is detected and the parking duration is smaller than a duration threshold value are met simultaneously, the vehicle is a temporary parking vehicle.
In the scheme, the monitoring video images are selected from monitoring cameras installed on urban roads, the urban roads are divided into no-parking areas and temporary parking areas according to traffic demands, each monitoring camera monitors a specific area, and the monitoring video images of continuous frames of the same monitoring camera can be selected as judgment data.
And inputting the monitoring video images of the continuous frames into a illegal parking judgment module to judge illegal parking, and firstly selecting the monitoring area in the monitoring video images of the continuous frames. It is worth mentioning that the selection of the monitoring area may be manually selected, or may be automatically selected by a computer through deep learning. The no-parking area refers to an area where parking is forbidden, such as a pedestrian crossing, a yellow grid area, a bus stop, a bus lane or a fire fighting channel; the temporary parking area refers to an area where parking is possible, such as a taxi boarding and disembarking point.
Whether the temporary parking area or the no-parking area is aimed at, the vehicle coordinate data and the license plate number of the vehicle in the monitoring area in the monitoring video images of the continuous frames are detected, and the difference value between the offset value and the position threshold value of the vehicle coordinate data of the vehicle corresponding to the same license plate number is judged.
The method comprises the following specific steps:
inputting the monitoring video images of the continuous frames into a vehicle detection model for vehicle detection, and acquiring vehicle coordinate data of the vehicle, wherein the vehicle coordinate data at least comprises a vehicle center point coordinate and a vehicle width and height; and intercepting a vehicle area from the corresponding monitoring video image based on the vehicle coordinate data, carrying out license plate recognition on the vehicle area, acquiring the vehicle number, selecting the vehicle center point coordinates of the vehicles corresponding to the same license plate number, calculating whether the difference value between the vehicle center point coordinates is smaller than a position threshold value, and if the difference value is smaller than the position threshold value, identifying that the vehicle is in a stop state.
For the no-parking area, if the same vehicle stops in the no-parking area, the vehicle can be determined as a no-parking vehicle, that is, the vehicle is determined as the no-parking vehicle only when the offset value of the vehicle coordinate data is smaller than the position threshold value.
In contrast, in the temporary parking area, if a driver only stops the temporary parking area for a predetermined time, the driver does not belong to the parking violation vehicle. In particular, a temporary parking area for facilitating passengers to get on and off is provided at a location such as a station entrance of a railway station. For the temporary parking area, firstly, whether the vehicle is in a parking state is judged based on the preamble, and if the vehicle is in the parking state, subsequent detection is carried out, wherein the specific detection process is as follows:
inputting the vehicle area corresponding to the vehicle into a face detection model for face detection, and if no face is detected on the driver seat, directly judging that the vehicle is an illegal parking vehicle; if the driver seat detects a face, parking time statistics is carried out, if the parking time is not less than a parking threshold value, the vehicle is an illegal parking vehicle, and if not, the vehicle is a temporary parking vehicle.
It is worth mentioning that the method for counting the parking time length comprises the following steps: and selecting the time corresponding to the monitoring video image of the vehicle appearing earliest and the time corresponding to the monitoring video image of the vehicle appearing latest, and solving the difference between the two times as the parking time.
The vehicle detection model adopts a trained deep learning network framework, and specifically, the construction process of the vehicle detection model is as follows:
labeling training samples: preprocessing the collected monitoring video image, and converting the image size of the monitoring video image into W1×H1And labeling the monitoring video image, wherein the labeling information comprises the position coordinates of the center point of the vehicle and the width and height of the vehicle.
Building and training a vehicle detection network structure: the vehicle detection network structure adopts an hourglass network to extract features of input data to obtain a feature map, the feature map is connected with 3 convolutional layers, the 3 convolutional layers are respectively connected with a focal loss function, a central offset loss function and a frame size loss function, the vehicle detection network is trained by adopting the focal loss function, the central offset loss function and the frame size loss function until the loss functions are not reduced, the training is stopped, the focal loss function carries out a target central point and target category prediction task, the central offset loss function carries out a central point offset prediction task, and the frame size loss function carries out a target width and height prediction task.
And inputting the monitoring video images of the continuous frames into the vehicle detection model obtained based on the training, so that at least the coordinates of the center point of the vehicle and the width and the height of the vehicle can be obtained.
The license plate recognition process comprises the following steps: intercepting a vehicle region from a corresponding monitoring video image based on the vehicle coordinate data, inputting the vehicle region into an open-source license plate detection model for license plate detection, acquiring a license plate region, and transforming the size of the license plate region into W2×H2And inputting the license plate number into an open-source license plate recognition model for license plate recognition to obtain a license plate number.
The face detection model adopts a trained deep learning network framework, and specifically, the construction process of the face detection model is as follows:
step 1: preparing a training sample: randomly cropping the public face data set WIDER FACE to generate a positive sample, a negative sample and an intermediate sample, wherein the negative sample is a sample which intersects the annotated face region at a ratio less than a threshold U1, the positive sample is a sample which intersects the annotated face region at a ratio greater than a threshold U2, and the intermediate sample is a sample which intersects the annotated face region at a ratio between the threshold U3 and a threshold U4;
step 2: and (3) building a network structure of the face detection model: the network structure of the face detection model comprises two networks, namely a candidate frame network and an output network, wherein the candidate frame network comprises p1 convolutional layers and p2 pooling layers, and the output network comprises p3 convolutional layers, p4 pooling layers and p5 full-connection layers;
and step 3: training: transforming the positive, negative and intermediate sample sizes into W3×H3Inputting the candidate frame network, and training the candidate frame network by adopting a SoftMax loss function and an Euclidean loss function, wherein the SoftMax loss function carries out a face two-classification task, and the Euclidean loss function carries out a frame regression task, and the training is stopped until the loss value is not reduced any more;
and 4, step 4: inputting the original image corresponding to the public face data set into the candidate frame network trained in the step 3 to obtain a candidate area, dividing the intersection ratio of the candidate area and the labeled face area into a positive sample, a negative sample and an intermediate sample, wherein the negative sample is a sample which is smaller than a threshold value U1 in the intersection ratio with the labeled face area, the positive sample is a sample which is larger than a threshold value U2 in the intersection ratio with the labeled face area, and the intermediate sample is a sample which is between a threshold value U3 and a threshold value U4 in the intersection ratio with the labeled face area;
and 5: converting the positive, negative and intermediate sample sizes obtained in step 4 into W4×H4Input to the output network, and train the output network by using SoftMax loss function and Euclidean loss function until the loss valueStopping training when the speed is not reduced any more;
step 6: and (3) testing: zooming the image to be tested for multiple times according to a certain proportion to form an image pyramid;
and 7: inputting the pyramid into the trained candidate frame network to obtain candidate frames, and deleting the candidate frames with excessively high overlap by adopting a non-maximum suppression algorithm;
and 8: intercepting an interested area from an original image to be tested according to the position of the candidate frame and transforming the size of the interested area into W4×H4Inputting the candidate frames into the trained output network, deleting some overlapped candidate frames by adopting a non-maximum suppression algorithm, and taking the remaining candidate frames as final face frames.
The second aspect of the scheme provides a method for managing illegal vehicles based on urban roads, which comprises the following steps:
acquiring monitoring video images of continuous frames, and detecting whether illegal vehicles exist by using the illegal vehicle detection method based on the urban road;
the method comprises the steps of obtaining illegal parking data of illegal parking vehicles, analyzing the illegal parking data, and obtaining the vehicles with the largest number of illegal parking times and the areas with the largest number of illegal parking vehicles, wherein the illegal parking data comprise dates, illegal parking addresses, parking time of the illegal parking vehicles and license plate numbers.
Correspondingly, in an embodiment of the scheme, manual management is strengthened for the address with the most illegal parking vehicles, and a blacklist library of the illegal parking vehicles is added for the vehicle with the most illegal parking times, so that the illegal parking phenomenon can be substantially controlled.
Specifically, the process of analyzing the violation data is as follows:
classifying the illegal parking addresses based on the illegal parking addresses of the illegal parking data, counting the quantity of illegal parking vehicles at different addresses, and sequentially acquiring the areas with the most illegal parking vehicles;
and counting the number of the same license plate numbers based on the license plate numbers of the illegal parking data, and sequentially acquiring the vehicles with the largest number of illegal parking times.
It should be noted that in some embodiments, a control time period may be defined, and the violation data in the control time period may be selected for control and analysis.
Through the scheme, the place, the time, the license plate number and other data of the vehicle parking against the regulations can be stored and subjected to data analysis, the place where the vehicle parking against the regulations often occurs is managed in an enhanced mode, the vehicle parking against the regulations often adds a black list library, and the vehicle parking against the regulations can be effectively managed.
A third aspect of the present invention provides an urban road-based illegal parking vehicle detection system, which is used as a carrier to operate the above urban road-based illegal parking vehicle detection method, and includes:
the monitoring area acquisition module is used for acquiring a monitoring area of a monitoring video image with continuous frames, wherein the monitoring area is a no-parking area or a temporary parking area;
the vehicle detection module is internally loaded with a trained vehicle detection model and used for acquiring vehicle coordinate data;
the license plate recognition module is used for recognizing the license plate number of the vehicle;
the stop state judgment module is used for judging whether the deviation value of the vehicle coordinate data of the vehicles corresponding to the same license plate number is smaller than a position threshold value or not;
the face recognition module is internally loaded with a trained face detection model and used for detecting whether a driver seat is occupied or not;
and the parking duration judging module is used for calculating the parking duration of the vehicle and judging whether the parking duration is less than a duration threshold value.
It should be noted that the steps involved in the operation of the urban road-based illegal vehicle detection system are as described above, and are not described herein in a cumbersome manner.
A fourth aspect of the present invention provides an urban road-based parking violation management system, which is used as a carrier to operate the above urban road-based parking violation management method, and additionally includes, on the basis of an urban road-based parking violation detection system:
and the illegal parking data analysis module is used for analyzing illegal parking data of illegal parking vehicles and analyzing the illegal parking data to obtain the vehicles with the most illegal parking times and the areas with the most illegal parking vehicles, wherein the illegal parking data comprises dates, illegal parking addresses, parking time of the illegal parking vehicles and license plate numbers.
The computer system of the server for implementing the urban road-based illegal vehicle detection and management method according to the present embodiment includes a central processing unit CPU) that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) or a program loaded from a storage section into a Random Access Memory (RAM). In the RAM, various programs and data necessary for system operation are also stored. The CPU, ROM, and RAM are connected to each other via a bus. An input/output (I/O) interface is also connected to the bus.
In particular, according to an embodiment of the present disclosure, the process of the urban road-based illegal vehicle detection and management method described above with reference to the flowchart may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method of urban road-based illegal vehicle detection and management method illustrated in the flowchart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication section, and/or installed from a removable medium. The computer program performs the above-described functions defined in the system of the present invention when executed by a Central Processing Unit (CPU).
It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing.
The flowchart and block diagrams of a method for urban road-based illegal parking vehicle detection and management in the figures illustrate the architecture, functionality, and operation of possible implementations of an urban road-based illegal parking vehicle detection and management system, an urban road-based illegal parking vehicle detection and management method, and a computer program product according to various embodiments of the present invention.
In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. Each block of the block diagrams or flowchart illustrations, and combinations of blocks in the block diagrams or flowchart illustrations, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present invention may be implemented by software, or may be implemented by hardware, and the described modules may also be disposed in a processor.
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by the apparatus, cause the apparatus to perform process steps corresponding to a method for urban road-based illegal vehicle detection and management.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for detecting illegal parking vehicles based on urban roads is characterized by comprising the following steps:
acquiring monitoring video images of continuous frames and selecting a monitoring area;
detecting vehicle coordinate data and license plate numbers of vehicles in the monitoring area in the monitoring video images of the continuous frames, wherein if the deviation value of the vehicle coordinate data of the vehicles corresponding to the same license plate number is smaller than a position threshold value and the monitoring area corresponds to a no-parking area, the vehicles are illegal parking vehicles; if the monitoring area corresponds to a temporary parking area, carrying out face detection and parking time counting on the vehicle, if the conditions that the face is detected and the parking time is less than a time threshold value are met, determining that the vehicle is a temporary parking vehicle, and if not, determining that the vehicle is an illegal parking vehicle.
2. The urban road-based illegal parking vehicle detection method according to claim 1, wherein the vehicle coordinate data at least comprises vehicle center point coordinates and vehicle width and height, vehicle areas are intercepted from corresponding monitoring video images based on the vehicle coordinate data, license plate recognition is performed on the vehicle areas, the vehicle numbers are obtained, the vehicle center point coordinates of vehicles corresponding to the same license plate numbers are selected, and whether the difference value between the vehicle center point coordinates is smaller than a position threshold value is calculated.
3. The urban road-based illegal parking vehicle detection method according to claim 1, wherein a vehicle region corresponding to the vehicle is input into a face detection model for face detection, and if no face is detected on a driver seat, the vehicle is an illegal parking vehicle; and if the driver seat detects a human face, counting the parking time, and if the parking time is not less than a parking threshold value, determining that the vehicle is an illegal parking vehicle.
4. The urban road-based illegal parking vehicle detection method according to claim 1, wherein the continuous frames of surveillance video images are input into a vehicle detection model to obtain vehicle coordinate data of the vehicle, a vehicle detection network structure corresponding to the vehicle detection model adopts an hourglass network to perform feature extraction on the input data to obtain a feature map, the feature map is followed by 3 convolutional layers, the 3 convolutional layers are respectively connected with a focal loss function, a center offset loss function and a frame size loss function, and the vehicle detection network is trained by adopting the focal loss function, the center offset loss function and the frame size loss function until the loss functions are not reduced.
5. The urban road-based illegal parking vehicle detection method according to claim 3, characterized in that the construction process of the face detection model is as follows:
step 1: preparing a training sample: randomly cropping the public face data set to generate a positive sample, a negative sample and an intermediate sample, wherein the negative sample is a sample which is intersected with the annotated face region and has a ratio smaller than a threshold value U1, the positive sample is a sample which is intersected with the annotated face region and has a ratio larger than a threshold value U2, and the intermediate sample is a sample which is intersected with the annotated face region and has a ratio between the threshold value U3 and a threshold value U4;
step 2: and (3) building a network structure of the face detection model: the network structure of the face detection model comprises two networks, namely a candidate frame network and an output network, wherein the candidate frame network comprises p1 convolutional layers and p2 pooling layers, and the output network comprises p3 convolutional layers, p4 pooling layers and p5 full-connection layers;
and step 3: training: transforming the positive, negative and intermediate sample sizes into W3×H3Inputting the frame candidate network, training the frame candidate network by adopting a SoftMax loss function and a Euclidean loss function, and stopping training until the loss value is not reduced any more;
and 4, step 4: inputting an original image corresponding to the public face data set into the candidate frame network trained in the step 3 to obtain a candidate area, and dividing a positive sample, a negative sample and an intermediate sample by the intersection ratio of the candidate area and the labeled face area;
and 5: converting the positive, negative and intermediate sample sizes obtained in step 4 into W4×H4And inputting the loss value into the output network, training the output network by adopting a SoftMax loss function and a Euclidean loss function, and stopping training until the loss value is not reduced any more.
6. The urban road-based illegal parking vehicle detection method according to claim 3, characterized in that a surveillance video image corresponding to a vehicle area corresponding to the vehicle is input to the face detection model for detection, the surveillance video image is scaled multiple times to form an image pyramid, the pyramid is input to a candidate frame network of the trained face detection model to obtain candidate frames, and the candidate frames with too high overlap are deleted by adopting a non-maximum suppression algorithm; and intercepting an interested area from the original monitoring video image to be tested according to the position of the candidate frame, converting the size of the interested area, inputting the interested area into a candidate frame network of the face detection model, deleting the overlapped candidate frame by adopting a non-maximum suppression algorithm, and taking the rest candidate frames as final face frames.
7. A method for managing illegal parking vehicles based on urban roads is characterized in that monitoring video images of continuous frames are obtained, and whether illegal parking vehicles exist is detected by the method for detecting illegal parking vehicles based on urban roads as claimed in any one of claims 1 to 6;
the method comprises the steps of obtaining illegal parking data of illegal parking vehicles, analyzing the illegal parking data, and obtaining the vehicles with the largest number of illegal parking times and the areas with the largest number of illegal parking vehicles, wherein the illegal parking data comprise dates, illegal parking addresses, parking time of the illegal parking vehicles and license plate numbers.
8. The urban road-based illegal parking vehicle management method according to claim 7, wherein manual management is enhanced for the address with the most illegal vehicles, and a blacklist of illegal parking vehicles is added to the vehicle with the most illegal parking times.
9. An urban road-based illegal parking vehicle detection system for operating the urban road-based illegal parking vehicle detection method according to any one of claims 1 to 7, comprising:
the monitoring area acquisition module is used for acquiring a monitoring area of a monitoring video image with continuous frames, wherein the monitoring area is a no-parking area or a temporary parking area;
the vehicle detection module is internally loaded with a trained vehicle detection model and used for acquiring vehicle coordinate data;
the license plate recognition module is used for recognizing the license plate number of the vehicle;
the stop state judgment module is used for judging whether the deviation value of the vehicle coordinate data of the vehicles corresponding to the same license plate number is smaller than a position threshold value or not;
the face recognition module is internally loaded with a trained face detection model and used for detecting whether a driver seat is occupied or not;
and the parking duration judging module is used for calculating the parking duration of the vehicle and judging whether the parking duration is less than a duration threshold value.
10. An urban road-based parking violation management system for operating the urban road-based parking violation management method of claim 8, comprising:
the monitoring area acquisition module is used for acquiring a monitoring area of a monitoring video image with continuous frames, wherein the monitoring area is a no-parking area or a temporary parking area;
the vehicle detection module is internally loaded with a trained vehicle detection model and used for acquiring vehicle coordinate data;
the license plate recognition module is used for recognizing the license plate number of the vehicle;
the stop state judgment module is used for judging whether the deviation value of the vehicle coordinate data of the vehicles corresponding to the same license plate number is smaller than a position threshold value or not;
the face recognition module is internally loaded with a trained face detection model and used for detecting whether a driver seat is occupied or not;
the parking duration judging module is used for calculating the parking duration of the vehicle and judging whether the parking duration is smaller than a duration threshold value;
and the illegal parking data analysis module is used for analyzing illegal parking data of illegal parking vehicles and analyzing the illegal parking data to obtain the vehicles with the most illegal parking times and the areas with the most illegal parking vehicles, wherein the illegal parking data comprises dates, illegal parking addresses, parking time of the illegal parking vehicles and license plate numbers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011298904.9ACN112381014A (en) | 2020-11-19 | 2020-11-19 | Illegal parking vehicle detection and management method and system based on urban road |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011298904.9ACN112381014A (en) | 2020-11-19 | 2020-11-19 | Illegal parking vehicle detection and management method and system based on urban road |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112381014Atrue CN112381014A (en) | 2021-02-19 |
Family
ID=74585153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011298904.9APendingCN112381014A (en) | 2020-11-19 | 2020-11-19 | Illegal parking vehicle detection and management method and system based on urban road |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112381014A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113473077A (en)* | 2021-05-31 | 2021-10-01 | 超级视线科技有限公司 | Multi-dimension-based roadside reverse parking judgment method and system |
| CN113920482A (en)* | 2021-12-13 | 2022-01-11 | 江西科技学院 | Vehicle illegal parking detection method and system |
| CN113990101A (en)* | 2021-11-19 | 2022-01-28 | 深圳市捷顺科技实业股份有限公司 | Method, system and processing device for detecting vehicles in no-parking area |
| CN114120659A (en)* | 2021-10-28 | 2022-03-01 | 北京中拓新源科技有限公司 | Illegal parking detection system based on video recognition |
| CN114529875A (en)* | 2022-04-24 | 2022-05-24 | 浙江这里飞科技有限公司 | Method and device for detecting illegal parking vehicle, electronic equipment and storage medium |
| CN114708425A (en)* | 2022-03-17 | 2022-07-05 | 深圳力维智联技术有限公司 | Method, device and computer-readable storage medium for identifying illegal parking of vehicles |
| US20220351622A1 (en)* | 2021-04-28 | 2022-11-03 | GM Global Technology Operations LLC | Intelligent park assist system to reduce parking violations |
| CN116052031A (en)* | 2021-10-28 | 2023-05-02 | 中国电信股份有限公司 | Vehicle parking violation detection method and device, storage medium and electronic equipment |
| CN116564085A (en)* | 2023-05-09 | 2023-08-08 | 智道网联科技(北京)有限公司 | Vehicle detection method, device, electronic device, and storage medium |
| CN116645646A (en)* | 2023-05-29 | 2023-08-25 | 智慧互通科技股份有限公司 | A method and system for abnormal parking detection and early warning |
| CN116758736A (en)* | 2023-05-25 | 2023-09-15 | 蘑菇车联信息科技有限公司 | Method and device for detecting illegal stop and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5450937A (en)* | 1992-12-10 | 1995-09-19 | Nippon Conlux Co., Ltd. | Paper currency discriminating device |
| CN102568206A (en)* | 2012-01-13 | 2012-07-11 | 大连民族学院 | Video monitoring-based method for detecting cars parking against regulations |
| CN107330373A (en)* | 2017-06-02 | 2017-11-07 | 重庆大学 | A kind of parking offense monitoring system based on video |
| CN109190442A (en)* | 2018-06-26 | 2019-01-11 | 杭州雄迈集成电路技术有限公司 | A kind of fast face detecting method based on depth cascade convolutional neural networks |
| JP2019018852A (en)* | 2018-10-30 | 2019-02-07 | みこらった株式会社 | Automatic driving vehicle and program therefor |
| CN110136449A (en)* | 2019-06-17 | 2019-08-16 | 珠海华园信息技术有限公司 | Traffic video frequency vehicle based on deep learning disobeys the method for stopping automatic identification candid photograph |
- 2020
- 2020-11-19CNCN202011298904.9Apatent/CN112381014A/enactivePending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5450937A (en)* | 1992-12-10 | 1995-09-19 | Nippon Conlux Co., Ltd. | Paper currency discriminating device |
| CN102568206A (en)* | 2012-01-13 | 2012-07-11 | 大连民族学院 | Video monitoring-based method for detecting cars parking against regulations |
| CN107330373A (en)* | 2017-06-02 | 2017-11-07 | 重庆大学 | A kind of parking offense monitoring system based on video |
| CN109190442A (en)* | 2018-06-26 | 2019-01-11 | 杭州雄迈集成电路技术有限公司 | A kind of fast face detecting method based on depth cascade convolutional neural networks |
| JP2019018852A (en)* | 2018-10-30 | 2019-02-07 | みこらった株式会社 | Automatic driving vehicle and program therefor |
| CN110136449A (en)* | 2019-06-17 | 2019-08-16 | 珠海华园信息技术有限公司 | Traffic video frequency vehicle based on deep learning disobeys the method for stopping automatic identification candid photograph |
Non-Patent Citations (3)
| Title |
|---|
| XINGYI ZHOU 等: "Objects as Points", 《ARXIV》* |
| ZHEMIN LIU 等: "Automatic Detection of Parking Violation and Capture of License Plate", 《2019 IEEE 10TH ANNUAL INFORMATION TECHNOLOGY, ELECTRONICS AND MOBILE COMMUNICATION CONFERENCE (IEMCON)》* |
| 梁书浩: "小区车辆违章停车监测系统设计", 《中国优秀硕士学位论文全文数据库 (信息科技辑)》* |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220351622A1 (en)* | 2021-04-28 | 2022-11-03 | GM Global Technology Operations LLC | Intelligent park assist system to reduce parking violations |
| CN115346387A (en)* | 2021-04-28 | 2022-11-15 | 通用汽车环球科技运作有限责任公司 | Intelligent parking auxiliary system for reducing illegal parking |
| CN113473077A (en)* | 2021-05-31 | 2021-10-01 | 超级视线科技有限公司 | Multi-dimension-based roadside reverse parking judgment method and system |
| CN114120659A (en)* | 2021-10-28 | 2022-03-01 | 北京中拓新源科技有限公司 | Illegal parking detection system based on video recognition |
| CN116052031A (en)* | 2021-10-28 | 2023-05-02 | 中国电信股份有限公司 | Vehicle parking violation detection method and device, storage medium and electronic equipment |
| CN113990101A (en)* | 2021-11-19 | 2022-01-28 | 深圳市捷顺科技实业股份有限公司 | Method, system and processing device for detecting vehicles in no-parking area |
| CN113920482A (en)* | 2021-12-13 | 2022-01-11 | 江西科技学院 | Vehicle illegal parking detection method and system |
| CN114708425A (en)* | 2022-03-17 | 2022-07-05 | 深圳力维智联技术有限公司 | Method, device and computer-readable storage medium for identifying illegal parking of vehicles |
| CN114529875A (en)* | 2022-04-24 | 2022-05-24 | 浙江这里飞科技有限公司 | Method and device for detecting illegal parking vehicle, electronic equipment and storage medium |
| CN116564085A (en)* | 2023-05-09 | 2023-08-08 | 智道网联科技(北京)有限公司 | Vehicle detection method, device, electronic device, and storage medium |
| CN116758736A (en)* | 2023-05-25 | 2023-09-15 | 蘑菇车联信息科技有限公司 | Method and device for detecting illegal stop and electronic equipment |
| CN116645646A (en)* | 2023-05-29 | 2023-08-25 | 智慧互通科技股份有限公司 | A method and system for abnormal parking detection and early warning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112381014A (en) | Illegal parking vehicle detection and management method and system based on urban road | |
| CN109637151B (en) | Method for identifying illegal driving of emergency lane on highway | |
| US11361556B2 (en) | Deterioration diagnosis device, deterioration diagnosis system, deterioration diagnosis method, and storage medium for storing program | |
| CN111429726B (en) | Monitoring video illegal parking vehicle detection and management method and corresponding system | |
| KR102122859B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN103279756B (en) | Vehicle detection based on integrated classifier analyzes system and determination method thereof | |
| CN110738857B (en) | Vehicle violation evidence obtaining method, device and equipment | |
| US6442474B1 (en) | Vision-based method and apparatus for monitoring vehicular traffic events | |
| CN114898297B (en) | Target detection and target tracking-based non-motor vehicle illegal behavior judgment method | |
| KR101908611B1 (en) | Parking control method for vehicle | |
| CN106373426A (en) | Computer vision-based parking space and illegal lane occupying parking monitoring method | |
| CN104282154B (en) | A kind of overload of vehicle monitoring system and method | |
| KR102174556B1 (en) | Apparatus for monitoring image to control traffic information employing Artificial Intelligence and vehicle number | |
| KR102122850B1 (en) | Solution for analysis road and recognition vehicle license plate employing deep-learning | |
| CN107067730B (en) | Network appointment vehicle-man-vehicle inconsistency monitoring method based on bayonet equipment | |
| CN109615864A (en) | Vehicle congestion analysis method, system, terminal and storage medium based on video structural | |
| CN108932849B (en) | Method and device for recording low-speed running illegal behaviors of multiple motor vehicles | |
| CN112380892B (en) | Image recognition method, device, equipment and medium | |
| CN113822285A (en) | Vehicle illegal parking identification method for complex application scene | |
| CN103914682A (en) | Vehicle license plate recognition method and system | |
| CN105259304A (en) | On-line monitoring system and method for pollutants in vehicle tail gas | |
| CN113851017A (en) | Pedestrian and vehicle identification and early warning multifunctional system based on road side RSU | |
| CN113723273A (en) | Vehicle track information determination method and device and computer equipment | |
| CN114387785A (en) | Safety management and control method and system based on intelligent highway and storable medium | |
| CN114693722A (en) | Vehicle driving behavior detection method, detection device and detection equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20210219 |