CN112733593A - Method and system for realizing image information identification based on image position - Google Patents

Abstract

The invention relates to a method and a system for realizing image information identification based on image positions, which are characterized in that a mark building is identified based on a machine learning image training technology, picture information of different angles of the mark building is collected, mark building characteristic points are identified, and a three-dimensional characteristic point model base of the mark building is constructed; establishing a model and a scenic spot relation graph of a basic resource scenic spot; uploading the picture by a user, extracting the mark buildings in the picture, and comparing the mark buildings with the model library to obtain corresponding mark building identification results; and sending corresponding tourist strategy recommendations to the user according to the identified mark building and the mark building and basic resource scenic spot relation graph.

Description

Method and system for realizing image information identification based on image position

Technical Field

The invention relates to the field of image information identification, in particular to a method and a system for realizing image information identification based on image positions.

Background

The image identification is to identify key feature points in an image based on an AI technology, so as to obtain a corresponding identification result. Currently, in the tourism industry, image recognition is mainly used for video surveillance and character behavior feature analysis in scenic spots. Through long-term research, tourists can know the characteristics of local scenic spots by searching when entering a scene in the tourism process, at present, the operation is generally realized based on character search, but for some scenic spots, the tourists cannot know the names of the scenic spots, and under the condition, effective information can not be provided for searching or consulting obviously.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides a method and a system for realizing image information identification based on an image position.

The purpose of the invention is realized by the following technical scheme:

a method for realizing image information identification based on image position comprises the following steps:

1) identifying a sign building based on a machine learning image training technology, collecting picture information of the sign building at different angles, identifying the feature points of the sign building, and constructing a three-dimensional feature point model library of the sign building;

2) establishing a model and a scenic spot relation graph of a basic resource scenic spot;

3) uploading the picture by a user, extracting the mark buildings in the picture, and comparing the mark buildings with the model library to obtain corresponding mark building identification results;

4) and sending corresponding travel strategy recommendations to the user according to the identified mark buildings and the relation graph of the mark buildings and the scenic spots of the basic resource scenic spot.

Further, in the step 1), the identification of the marker buildings is carried out based on a big data base, and the marker buildings are updated in real time.

Further, the model and the basic resource scenic spot and scenic spot relation map comprise scenic spot introductions corresponding to tourist routes, food recommendations, travel strategies, lodging recommendations, schedule arrangement, scenic spot and scenic spot resource lists and sign buildings.

Further, the extracting of the landmark buildings in the picture in 3) includes:

11) identifying and extracting a mark building in the picture;

12) focusing adjustment is carried out on the mark building so as to extract a clear mark building pattern;

13) dynamically correcting the picture angle, dynamically rotating the picture angle in the picture comparison process until a model with a consistent angle is matched, if the corresponding model is not matched all the time, manually identifying the mark building by a background, and adding the corresponding information into a model library after the corresponding information is obtained;

or the mark building information close to the mark building pattern is returned to the user, the user confirms for the second time, and the confirmed mark building information is returned after the user confirms.

Further, in the step 3), the user uploads the picture and obtains the current geographical position information of the user.

Further, in the step 3), a model of an area where the geographical location of the user is located is preferentially matched.

A system for implementing a method for implementing image information recognition based on image location, the system comprising:

the data acquisition module is used for collecting picture information of different angles of the mark building;

the image feature extraction module is used for extracting the mark building feature points in the picture;

the machine learning image training module is used for training based on the mark building feature points to obtain a three-dimensional feature point model base of the corresponding mark building and establishing a model and basic resource scenic spot relation graph;

a database for storing the model;

the data receiving module is used for receiving the pictures uploaded by the user;

the server compares the model with the mark building characteristics in the picture uploaded by the user and extracted by the image characteristic extraction module, and obtains an identification result;

and the signal sending module is used for sending the identification result to the user.

Furthermore, a user uploads the picture through the mobile terminal, and the data receiving module simultaneously acquires the positioning data of the user in the uploading process.

The invention has the beneficial effects that: according to the method, the image shot by the user is utilized, then the features in the image are extracted, so that the corresponding mark building information is obtained, the mark building information is compared with the model in the database, an accurate identification result is obtained, the identification result and the corresponding travel recommendation are returned to the user, and therefore the tourists can play more conveniently.

Drawings

FIG. 1 is a block diagram of the system of the present invention.

Detailed Description

The technical solution of the present invention is further described in detail with reference to the following specific examples, but the scope of the present invention is not limited to the following.

A method for realizing image information identification based on image position comprises the following steps:

1) identifying a sign building based on a machine learning image training technology, collecting picture information of the sign building at different angles, identifying the feature points of the sign building, and constructing a three-dimensional feature point model library of the sign building;

2) establishing a model and a scenic spot relation graph of a basic resource scenic spot;

3) uploading the picture by a user, extracting the mark buildings in the picture, and comparing the mark buildings with the model library to obtain corresponding mark building identification results;

4) and sending corresponding travel strategy recommendations to the user according to the identified mark buildings and the relation graph of the mark buildings and the scenic spots of the basic resource scenic spot.

Optionally, in the identification of the marker buildings, the marker buildings are identified based on a big data base and updated in real time, and specifically, a crawler technology may be used to obtain corresponding big data in the internet as base data to identify the corresponding marker buildings.

Further, the method for realizing image information identification based on the image position comprises the steps that the model and the basic resource scenic spot relation graph comprise tourist routes, food recommendation, travel strategy, accommodation recommendation, schedule arrangement, scenic spot resource list and scenic spot introduction corresponding to the landmark building.

Further, a method for realizing image information identification based on image position, wherein the step of extracting the mark building in the picture comprises the following steps:

11) identifying and extracting a mark building in the picture;

12) focusing adjustment is carried out on the mark building so as to extract a clear mark building pattern;

13) dynamically correcting the picture angle, dynamically rotating the picture angle in the picture comparison process until a model with a consistent angle is matched, if the corresponding model is not matched all the time, manually identifying the mark building by a background, and adding the corresponding information into a model library after the corresponding information is obtained;

or the mark building information close to the mark building pattern is returned to the user, the user confirms for the second time, and the confirmed mark building information is returned after the user confirms.

Further, a method for realizing image information identification based on image position, wherein a user uploads a picture and obtains the current geographical position information of the user, and the method specifically determines the geographical position according to the positioning information when the user sends the picture.

Furthermore, the method for realizing image information identification based on the image position carries out the matching of the mark buildings in the image identification process and preferentially matches the model of the area where the user geographic position is located.

Referring to fig. 1, the present embodiment also provides a system for implementing a method for implementing image information recognition based on image location, the system comprising:

the data acquisition module is used for collecting picture information of different angles of the mark building;

the image feature extraction module is used for extracting the mark building feature points in the picture;

the machine learning image training module is used for training based on the mark building feature points to obtain a three-dimensional feature point model base of the corresponding mark building and establishing a model and basic resource scenic spot relation graph;

a database for storing models;

the data receiving module is used for receiving the pictures uploaded by the user;

the server compares the model with the mark building characteristics in the picture uploaded by the user and extracted by the image characteristic extraction module, and obtains an identification result;

and the signal sending module is used for sending the identification result to the user.

Optionally, the user uploads the picture through the mobile terminal, and the data receiving module simultaneously obtains the positioning data of the user during the uploading process, where the mobile terminal is generally used by the user.

The foregoing is illustrative of the preferred embodiments of this invention, and it is to be understood that the invention is not limited to the precise form disclosed herein and that various other combinations, modifications, and environments may be resorted to, falling within the scope of the concept as disclosed herein, either as described above or as apparent to those skilled in the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (8)

1. A method for realizing image information identification based on image position is characterized by comprising the following steps:

1) identifying a sign building based on a machine learning image training technology, collecting picture information of the sign building at different angles, identifying the feature points of the sign building, and constructing a three-dimensional feature point model library of the sign building;

2) establishing a model and a scenic spot relation graph of a basic resource scenic spot;

3) uploading the picture by a user, extracting the mark buildings in the picture, and comparing the mark buildings with the model library to obtain corresponding mark building identification results;

4) and sending corresponding travel strategy recommendations to the user according to the identified mark buildings and the relation graph of the mark buildings and the scenic spots of the basic resource scenic spot.

2. The method for realizing image information identification based on image position according to claim 1, characterized in that in 1), the identification of the landmark buildings is carried out based on big data base and is updated in real time.

3. The method of claim 2, wherein the model and the map of the scenic spot relationships of the basic resource comprise a tour route, a food recommendation, a travel strategy, an accommodation recommendation, a schedule, a scenic spot resource list, and a scenic spot introduction corresponding to a landmark building.

4. The method of claim 3, wherein the extracting of the landmark buildings in the picture in 3) comprises:

11) identifying and extracting a mark building in the picture;

12) focusing adjustment is carried out on the mark building so as to extract a clear mark building pattern;

13) dynamically correcting the picture angle, dynamically rotating the picture angle in the picture comparison process until a model with a consistent angle is matched, if the corresponding model is not matched all the time, manually identifying the mark building by a background, and adding the corresponding information into a model library after the corresponding information is obtained;

or the mark building information close to the mark building pattern is returned to the user, the user confirms for the second time, and the confirmed mark building information is returned after the user confirms.

5. The method of claim 4, wherein the user uploads the picture and obtains the current geographical location information of the user in 3).

6. The method of claim 5, wherein the model of the area where the geographic location of the user is located in 3) is preferentially matched.

7. A system for implementing the method for realizing image information identification based on image position according to any one of claims 1-6, the system is characterized by comprising:

the data acquisition module is used for collecting picture information of different angles of the mark building;

the image feature extraction module is used for extracting the mark building feature points in the picture;

the machine learning image training module is used for training based on the mark building feature points to obtain a three-dimensional feature point model base of the corresponding mark building and establishing a model and basic resource scenic spot relation graph;

a database for storing the model;

the data receiving module is used for receiving the pictures uploaded by the user;

the server compares the model with the mark building characteristics in the picture uploaded by the user and extracted by the image characteristic extraction module, and obtains an identification result;

and the signal sending module is used for sending the identification result to the user.

8. The system according to claim 7, wherein the user uploads the picture through the mobile terminal, and the data receiving module simultaneously obtains the positioning data of the user during the uploading process.

Images