CN111860094A - Method, system and computer readable storage medium for verifying image information - Google Patents
Method, system and computer readable storage medium for verifying image information Download PDFInfo
- Publication number
- CN111860094A CN111860094A CN202010180923.5A CN202010180923A CN111860094A CN 111860094 A CN111860094 A CN 111860094A CN 202010180923 A CN202010180923 A CN 202010180923A CN 111860094 A CN111860094 A CN 111860094A
- Authority
- CN
- China
- Prior art keywords
- information
- building
- vehicle
- preset
- detected
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000003860 storage Methods 0.000 title claims abstract description 11
- 238000012795 verification Methods 0.000 claims abstract description 52
- 238000001514 detection method Methods 0.000 claims abstract description 43
- 239000013598 vector Substances 0.000 claims description 44
- 238000012512 characterization method Methods 0.000 claims description 42
- 238000004590 computer program Methods 0.000 claims description 11
- 230000001360 synchronised effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 11
- 238000012015 optical character recognition Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 1
- 238000011066 ex-situ storage Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method, a system and a computer readable storage medium for verifying image information, wherein the method for verifying the image information comprises the steps of receiving the image information, and acquiring preset vehicle information and preset building information; acquiring corresponding information of the vehicle to be detected according to the image information, and acquiring information of at least one building to be detected; and matching the information of the vehicle to be detected with the preset vehicle information, matching the information of any building to be detected with the preset building information, and determining that the image information passes the verification. By applying the technical scheme provided by the invention, when the network appointment vehicle identity is verified, the vehicle information and the building information in the picture information are verified simultaneously, the network appointment vehicle identity is ensured to be correct, and meanwhile, the condition of vehicle cheating verification in different places is avoided by introducing synchronous detection on random landmark buildings, and the accuracy and reliability of vehicle image detection are improved.
Description
Technical Field
The present invention relates to the field of image recognition technologies, and in particular, to an image information verification method, an image information verification system, and a computer-readable storage medium.
Background
In the related art, the network appointment platform has tens of thousands of vehicles operating every day, and a small part of vehicles which go out of the network appointment platform are not registered vehicles, so that whether the vehicles are consistent with the registered vehicles or not needs to be verified on line for the operating vehicles in order to supervise the situation that the vehicles do not accord with each other. However, in the existing verification method, only the vehicle image is verified, but the region where the vehicle is located is not verified, so that a vulnerability of 'ex-situ' verification exists.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art or the related art.
To this end, a first aspect of the present invention proposes a method of verifying image information.
A second aspect of the present invention proposes a verification system of image information.
A third aspect of the invention proposes a computer-readable storage medium.
In view of this, a first aspect of the present invention provides a method for verifying image information, including: receiving image information, and acquiring preset vehicle information and preset building information; acquiring corresponding information of the vehicle to be detected according to the image information, and acquiring information of at least one building to be detected; and matching the information of the vehicle to be detected with the preset vehicle information, matching the information of any building to be detected with the preset building information, and determining that the image information passes the verification.
In this technical solution, when verifying image information, the image information is received first. The image information is used for verifying the network car booking identity, and comprises information of a vehicle to be detected and information of a building to be detected. The information of the vehicle to be detected, namely the vehicle image of the online appointment, can comprise the vehicle appearance, the vehicle color and the license plate information. The information of the building to be detected, namely the network appointment registration place, namely landmark buildings at the location where the vehicle to be detected is registered, such as railway stations, monuments, skyscrapers and other buildings with 'unique' properties, can be used as landmark buildings.
When the image information is verified, the information of the vehicle to be detected and the information of the building to be detected, which is located in the same image as the vehicle to be detected, are obtained from the image information through an image recognition method, if and only if the vehicle to be detected is matched with the preset vehicle information recorded by the network appointment platform, and any building information to be detected included in the image information is matched with the preset building information, the image information is judged to pass the verification, otherwise, the image information does not pass the verification. The preset building information is building information corresponding to a randomly designated building in landmark buildings where the vehicle to be detected is registered.
By applying the technical scheme provided by the invention, when the network appointment vehicle identity is verified, the vehicle information and the building information in the picture information are verified simultaneously, the network appointment vehicle identity is ensured to be correct, and meanwhile, the condition of vehicle cheating verification in different places is avoided by introducing synchronous detection on random landmark buildings, and the accuracy and reliability of vehicle image detection are improved.
In addition, the method for verifying the image information in the above technical solution provided by the present invention may further have the following additional technical features:
in the above technical solution, the step of obtaining information of at least one building to be detected specifically includes: analyzing the image information through a building detection model to obtain a characteristic image containing one or more buildings; determining coordinate information corresponding to one or more buildings according to the characteristic images, and determining characteristic information corresponding to any building; and generating a characterization vector corresponding to any building according to the coordinate information and the characteristic information, and generating the information of the building to be detected according to the characterization vector.
In the technical scheme, original image information is analyzed through a building detection model, and a characteristic image containing one or more buildings is obtained. Specifically, the building detection model integrates a target detection model and a feature similarity matching model, and basic features are extracted through a common backbone network of the models. Specifically, a target detection branch is added on the backbone network, and the positions of different buildings are detected through the target detection branch to determine coordinate information. Next, the basic feature map extracted from the backbone network by the coordinate information of different buildings, i.e. the feature image, is used to extract the local feature information corresponding to each building, i.e. the feature information.
The feature information of different buildings is respectively accessed into the pixel row identification heads to output a characterization vector corresponding to any building, the building information to be detected is generated according to the characterization vector, whether the building is a target building or not is determined according to the building information to be detected, the similarity between the feature images of the buildings and the specified building does not need to be repeatedly matched and verified one by one, the calculation resources are effectively saved, and the detection efficiency is improved.
In any of the above technical solutions, the preset building information includes a preset characterization vector corresponding to a preset building; determining the similarity between any characterization vector in the building information to be detected and a preset characterization vector; and if the similarity between any characterization vector and the preset characterization vector is smaller than a preset threshold value, determining that the information of the building to be detected is matched with the preset building information.
According to the technical scheme, the similarity between the characterization vector of the building to be detected and the preset characterization vector can be determined by comparing the characterization vector in the building information to be detected with the preset characterization vector. And if the similarity is higher than the preset threshold value, the building to be detected and the preset building are determined to be the same building, and the building information in the image information is matched with the preset building information. When any building included in the image information is determined as a preset building, the shooting place of the current image information can be determined to be consistent with the registration place of the vehicle to be detected on the vehicle-ordering platform, so that the cheating means of shooting in different places is effectively prevented, and the image verification accuracy of the vehicle-ordering is improved.
In any of the above technical solutions, the step of obtaining the corresponding information of the vehicle to be detected according to the image information specifically includes: analyzing the image information to obtain a vehicle image; detecting a vehicle image through a license plate recognition model to determine license plate number information corresponding to the vehicle image; and determining the vehicle information according to the license plate number information.
In the technical scheme, when vehicle verification is carried out, an image of a vehicle to be detected in image information is detected and intercepted through a target recognition model, and the vehicle image should comprise an integral image of a certain side of the vehicle and a license plate image. After the vehicle image is intercepted, the corresponding license plate number information is determined through a license plate Recognition model (OCR, Optical Character Recognition), and the vehicle information is determined according to the license plate number information so as to ensure the integrity and the accuracy of the vehicle information.
In some embodiments, the vehicle information further includes vehicle appearance information, including in particular vehicle color, vehicle size, vehicle type, etc., and may also include vehicle brand information, etc.
In any of the above technical solutions, the preset vehicle information includes preset license plate number information; and if the license plate number information is matched with the preset license plate number information, determining that the information of the vehicle to be detected is matched with the preset vehicle information.
In the technical scheme, when determining whether the vehicle to be detected is a preset vehicle registered by the platform, the license plate number is matched with the lowest requirement. Therefore, when the detected license plate number information is matched with the preset license plate number information registered by the platform, the information of the vehicle to be detected can be determined to be matched with the preset vehicle information, namely, the vehicle in the image information is the platform registered vehicle.
In any of the above technical solutions, before the step of receiving the image information, the method for verifying the image information further includes: receiving a verification request, and acquiring first position information of a corresponding vehicle to be detected according to the verification request; acquiring a preset building set, determining a landmark building closest to the vehicle to be detected in the preset building set according to the first position information, and determining the landmark building as the preset building; and sending the second position information of the preset building to a terminal corresponding to the vehicle to be detected.
In the technical scheme, when the online car booking is subjected to image verification, a verification request can be initiated by the online car booking platform, and the verification request can also be initiated by a terminal held by a driver or a terminal held by a signed user of online car booking service. After receiving the verification request, first location information of the vehicle to be detected, namely, a target network appointment vehicle of the verification request is obtained, and the first location information can be obtained by a Global Positioning System (GPS) installed on the network appointment vehicle and uploaded by a vehicle side.
And after the first position information of the vehicle to be detected is acquired, acquiring a preset building set prestored by the platform. The preset building set comprises landmark buildings of all cities or regions with the network car booking service opened and characteristic information of the landmark buildings at all viewing angles. And selecting a landmark building closest to the vehicle to be detected from the preset building set, determining the landmark building as the preset building, and sending the second position information of the preset building to the terminal corresponding to the vehicle to be detected so that the driver can drive to the position near the landmark building for photographing.
When the preset building is determined, a plurality of landmark buildings within a certain range away from the vehicle to be detected can be determined, one landmark building is randomly selected from the landmark buildings with short distances to serve as the preset building, the randomness and the diversity of the preset building during selection are further improved, and cheating is avoided.
In any of the above technical solutions, the step of receiving the image information specifically includes: and receiving image information sent by the terminal, wherein the image information simultaneously comprises an image corresponding to the vehicle to be detected and an image corresponding to a preset building.
In the technical scheme, the image information is uploaded by the terminal corresponding to the vehicle to be detected, namely the terminal held by the current vehicle booking driver receiving verification for verification, so as to ensure the identity of the driver. Meanwhile, the image information needs to simultaneously contain the image corresponding to the vehicle to be detected, which is pushed to the preset building, so that the accuracy of the vehicle is ensured on one hand, and the accuracy of the position of the vehicle is ensured on the other hand, and the verification accuracy is improved.
In any of the above technical solutions, before the step of receiving the image information, the method for verifying the image information further includes: acquiring a preset detection model and building sample data, wherein the building sample data comprises coordinate information and characteristic information corresponding to a plurality of buildings; and generating a training set according to the building sample data, and training a preset detection model through the training set to obtain a building detection model.
In the technical scheme, pictures of each landmark building at east, south, west and north view angles are collected, characteristic information of each building is calibrated through an image recognition technology, and position information is calibrated. Building sample data are generated through the marked pictures of the buildings, and a training set used for training a preset detection model is finally generated. The preset detection model is trained through the training set, and a high-efficiency building detection model can be obtained.
A second aspect of the invention provides a system for verification of image information, comprising a processor configured to store a computer program; a processor configured to execute a computer program to implement the method for verifying the image information provided in any of the above technical solutions, and therefore, the system for verifying the image information includes all the beneficial effects of the method for verifying the image information provided in any of the above technical solutions, which are not described herein again.
A third aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when being executed by a processor, implements the method for verifying the image information provided in any one of the above technical solutions, and therefore, the computer-readable storage medium includes all the beneficial effects of the method for verifying the image information provided in any one of the above technical solutions, which are not described herein again.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 shows a flow diagram of a method of authentication of image information according to one embodiment of the invention;
FIG. 2 shows another flow diagram of a method of authentication of image information according to one embodiment of the invention;
FIG. 3 illustrates yet another flow diagram of a method of authentication of image information according to one embodiment of the invention;
FIG. 4 illustrates yet another flow diagram of a method of authentication of image information according to one embodiment of the present invention;
FIG. 5 illustrates yet another flow diagram of a method of authentication of image information according to one embodiment of the present invention;
FIG. 6 illustrates yet another flow diagram of a method of authentication of image information according to one embodiment of the present invention;
FIG. 7 shows an architectural diagram of a building detection model according to one embodiment of the invention;
FIG. 8 illustrates a network appointment verification flow diagram according to one embodiment of the present invention;
fig. 9 shows a block diagram of the structure of a verification system of image information according to an embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
A method of authenticating image information, an authentication system of image information, and a computer-readable storage medium according to some embodiments of the present invention are described below with reference to fig. 1 to 9.
Example one
As shown in fig. 1, in an embodiment of the first aspect of the present invention, there is provided a method for verifying image information, including:
Step S102, receiving image information, and acquiring preset vehicle information and preset building information;
step S104, acquiring corresponding information of the vehicle to be detected according to the image information, and acquiring information of at least one building to be detected;
and S106, matching the information of the vehicle to be detected with the preset vehicle information, matching any building information to be detected with the preset building information, and determining that the image information passes the verification.
In this embodiment, in verifying the image information, the image information is received first. The image information is used for verifying the network car booking identity, and comprises information of a vehicle to be detected and information of a building to be detected. The information of the vehicle to be detected, namely the vehicle image of the online appointment, can comprise the vehicle appearance, the vehicle color and the license plate information. The information of the building to be detected, namely the network appointment registration place, namely landmark buildings at the location where the vehicle to be detected is registered, such as railway stations, monuments, skyscrapers and other buildings with 'unique' properties, can be used as landmark buildings.
When the image information is verified, the information of the vehicle to be detected and the information of the building to be detected, which is located in the same image as the vehicle to be detected, are obtained from the image information through an image recognition method, if and only if the vehicle to be detected is matched with the preset vehicle information recorded by the network appointment platform, and any building information to be detected included in the image information is matched with the preset building information, the image information is judged to pass the verification, otherwise, the image information does not pass the verification. The preset building information is building information corresponding to a randomly designated building in landmark buildings where the vehicle to be detected is registered.
According to the embodiment provided by the invention, when the identity of the networked car appointment is verified, the vehicle information and the building information in the picture information are verified at the same time, the identity of the networked car appointment vehicle is ensured to be correct, and meanwhile, the condition of cheating of vehicles verified in different places is avoided by introducing synchronous detection on random landmark buildings, so that the accuracy and the reliability of vehicle image detection are improved.
In an embodiment of the present invention, as shown in fig. 2, the step of acquiring information of at least one building to be detected specifically includes:
step S202, analyzing the image information through a building detection model to obtain a characteristic image containing one or more buildings;
step S204, determining coordinate information corresponding to one or more buildings according to the characteristic images, and determining characteristic information corresponding to any building;
and S206, generating a characterization vector corresponding to any building according to the coordinate information and the characteristic information, and generating the information of the building to be detected according to the characterization vector.
In this embodiment, the original image information is parsed by a building detection model, and a feature image containing one or more buildings is obtained. Specifically, the building detection model integrates a target detection model and a feature similarity matching model, and basic features are extracted through a common backbone network of the models. Specifically, a target detection branch is added on the backbone network, and the positions of different buildings are detected through the target detection branch to determine coordinate information. Next, the basic feature map extracted from the backbone network by the coordinate information of different buildings, i.e. the feature image, is used to extract the local feature information corresponding to each building, i.e. the feature information.
The feature information of different buildings is respectively accessed into the pixel row identification heads to output a characterization vector corresponding to any building, the building information to be detected is generated according to the characterization vector, whether the building is a target building or not is determined according to the building information to be detected, the similarity between the feature images of the buildings and the specified building does not need to be repeatedly matched and verified one by one, the calculation resources are effectively saved, and the detection efficiency is improved.
In an embodiment of the present invention, as shown in fig. 3, the preset building information includes a preset token vector corresponding to the preset building, and the method for verifying the image information further includes:
step S302, determining the similarity between any characterization vector in the building information to be detected and a preset characterization vector;
step S304, if the similarity between any characterization vector and the preset characterization vector is smaller than a preset threshold value, determining that the building information to be detected is matched with the preset building information.
In this embodiment, by comparing the characterization vector in the building information to be detected with the preset characterization vector, the similarity between the characterization vector of the building to be detected and the preset characterization vector can be determined. And if the similarity is higher than the preset threshold value, the building to be detected and the preset building are determined to be the same building, and the building information in the image information is matched with the preset building information. When any building included in the image information is determined as a preset building, the shooting place of the current image information can be determined to be consistent with the registration place of the vehicle to be detected on the vehicle-ordering platform, so that the cheating means of shooting in different places is effectively prevented, and the image verification accuracy of the vehicle-ordering is improved.
In an embodiment of the present invention, as shown in fig. 4, the step of obtaining corresponding information of the vehicle to be detected according to the image information specifically includes:
step S402, analyzing the image information to obtain a vehicle image;
step S404, detecting a vehicle image through a license plate recognition model to determine license plate number information corresponding to the vehicle image;
and step S406, determining vehicle information according to the license plate number information.
In this embodiment, when vehicle verification is performed, an image of a vehicle to be detected in image information is detected and intercepted by a target recognition model, where the vehicle image should include an entire image of a certain side of the vehicle and include a license plate image. After the vehicle image is intercepted, the corresponding license plate number information is determined through a license plate Recognition model (OCR, Optical Character Recognition), and the vehicle information is determined according to the license plate number information so as to ensure the integrity and the accuracy of the vehicle information.
In some embodiments, the vehicle information further includes vehicle appearance information, including in particular vehicle color, vehicle size, vehicle type, etc., and may also include vehicle brand information, etc.
In one embodiment of the present invention, the preset vehicle information includes preset license plate number information; and if the license plate number information is matched with the preset license plate number information, determining that the information of the vehicle to be detected is matched with the preset vehicle information.
In this embodiment, when determining whether the vehicle to be detected is a preset vehicle registered by the platform, the license plate number is matched to be the minimum requirement. Therefore, when the detected license plate number information is matched with the preset license plate number information registered by the platform, the information of the vehicle to be detected can be determined to be matched with the preset vehicle information, namely, the vehicle in the image information is the platform registered vehicle.
In one embodiment of the present invention, as shown in fig. 5, before the step of receiving the image information, the method for verifying the image information further includes:
step S502, receiving a verification request, and acquiring first position information of a corresponding vehicle to be detected according to the verification request;
step S504, a preset building set is obtained, and a landmark building closest to the vehicle to be detected is determined in the preset building set according to the first position information and determined as the preset building;
and S506, sending the second position information of the preset building to a terminal corresponding to the vehicle to be detected.
In this embodiment, when the online car appointment is performed with image verification, a verification request may be initiated by the online car appointment platform, or the verification request may be initiated by a terminal held by a driver, or a terminal held by a subscriber of the online car appointment service. After receiving the verification request, first location information of the vehicle to be detected, namely, a target network appointment vehicle of the verification request is obtained, and the first location information can be obtained by a Global Positioning System (GPS) installed on the network appointment vehicle and uploaded by a vehicle side.
And after the first position information of the vehicle to be detected is acquired, acquiring a preset building set prestored by the platform. The preset building set comprises landmark buildings of all cities or regions with the network car booking service opened and characteristic information of the landmark buildings at all viewing angles. And selecting a landmark building closest to the vehicle to be detected from the preset building set, determining the landmark building as the preset building, and sending the second position information of the preset building to the terminal corresponding to the vehicle to be detected so that the driver can drive to the position near the landmark building for photographing.
When the preset building is determined, a plurality of landmark buildings within a certain range away from the vehicle to be detected can be determined, one landmark building is randomly selected from the landmark buildings with short distances to serve as the preset building, the randomness and the diversity of the preset building during selection are further improved, and cheating is avoided.
In an embodiment of the present invention, the step of receiving the image information specifically includes: and receiving image information sent by the terminal, wherein the image information simultaneously comprises an image corresponding to the vehicle to be detected and an image corresponding to a preset building.
In this embodiment, the image information is uploaded by the terminal corresponding to the vehicle to be detected, that is, the terminal held by the currently-received-verified online appointment driver, for verification, so as to ensure the identity of the driver. Meanwhile, the image information needs to simultaneously contain the image corresponding to the vehicle to be detected, which is pushed to the preset building, so that the accuracy of the vehicle is ensured on one hand, and the accuracy of the position of the vehicle is ensured on the other hand, and the verification accuracy is improved.
In one embodiment of the present invention, as shown in fig. 6, before the step of receiving the image information, the method for verifying the image information further includes:
step S602, acquiring a preset detection model and building sample data, wherein the building sample data comprises coordinate information and characteristic information corresponding to a plurality of buildings;
and step S604, generating a training set according to the building sample data, and training a preset detection model through the training set to obtain a building detection model.
In the embodiment, the pictures of each landmark building in east, south, west and north view angles are collected, the characteristic information of each building is calibrated through an image recognition technology, and the position information is calibrated. Building sample data are generated through the marked pictures of the buildings, and a training set used for training a preset detection model is finally generated. The preset detection model is trained through the training set, and a high-efficiency building detection model can be obtained.
Example two
In a complete embodiment of the present invention, a process of checking whether a vehicle of a network appointment is a registered vehicle in a network appointment service process is taken as an example, and the embodiment of the present invention is described in detail.
Specifically, a driver specifies a nearby landmark building to have the vehicle and the landmark group together to complete vehicle verification when the driver completes a call. The method comprises the steps of uploading pictures after group photo, checking whether a designated landmark building exists in the group photo, comparing each building with a target landmark building if a plurality of buildings exist in a vehicle group photo background generally, and if each comparison is independent from the beginning, comparing the time consumption is long, so that the invention provides a new target building matching model, and matching of a plurality of buildings is completed by only one-time detection.
The schematic architecture of the building detection model is shown in fig. 7, and has the following characteristics:
1. a target detection model and a feature similarity matching model are fused, and the basic features are extracted through a common trunk network of the models. Adding target detection branches on a backbone network to detect different building positions, extracting local features corresponding to the buildings on a basic feature map extracted by the backbone network by using the positions of the different buildings respectively, accessing the local features of the different buildings and then accessing a similarity recognition head to output characterization vectors of the buildings, comparing the characterization vectors with the characterization vectors of the target buildings respectively, and regarding the same building when the distance between the features is less than a threshold value.
2. After local features of different buildings are extracted, an identification head of the same building in the south-east, the west-north directions is connected, and the feature vectors of the same building in different directions are strengthened to be consistent.
When the driver finishes an order, whether a recognizable landmark building exists near the driver is automatically judged, if yes, the position of the landmark is sent to the driver, and the driver is required to go to the appointed landmark to shoot a vehicle and a co-shooting uploading system of the landmark for real-time authentication.
The implementation process comprises the following steps:
Step 1, collecting pictures of the southeast, northwest and viewing angles of landmark buildings and calibrating the positions.
And 2, training the model of the invention offline through the collected pictures until the characterization vectors output by different visual angles of the same landmark building are similar.
And 3, during online verification, issuing a landmark building position closest to the driver, driving the driver to a specified landmark, and shooting a vehicle and the landmark for combined photograph uploading verification.
And 4, recognizing the license plate number through the license plate OCR recognition model, comparing the license plate number with the registered license plate number, and successfully confirming the vehicle when the license plate numbers are consistent.
And 5, extracting a basic characteristic diagram by using the model of the invention.
And 6, detecting the coordinates of different buildings in the group photo on the basic feature map in the step 5.
And 7, extracting local features of different buildings on the basis of the steps 5 and 6.
And 8, outputting the characterization vectors of different buildings on the basis of the step 7.
And 9, comparing the characterization vectors of the specified landmark buildings with the building characterization vectors in the group photo one by one.
And step 10, confirming that the landmark succeeds when the similarity of a certain building characterization vector and the specified landmark characterization vector in the group photo is higher than a threshold value.
And 11, when the step 4 and the step 10 are successful at the same time, the verification is regarded as successful, otherwise, the verification is failed.
A specific network appointment verification process is shown in fig. 8, and includes:
starting verification;
step S802, issuing a landmark building position which is closest to a driver and has a contrast picture;
step S804, the driver shoots a vehicle and the co-photograph of the building position of the appointed landmark and uploads the co-photograph;
step S806, extracting a basic characteristic diagram through the building detection model;
step S808, detecting coordinates of different buildings by the target detection branch;
step S810, extracting local features according to coordinates of different buildings;
step S812, outputting the characterization vectors of different buildings;
step S814, identifying the license plate number;
step S816, judging whether the license plate number is consistent with the registration and a landmark building exists in the group photo; if yes, the verification is successful, otherwise, the verification fails.
EXAMPLE III
As shown in fig. 9, in one embodiment of the present invention, there is provided an authentication system 900 for image information, comprising a memory 902 configured to store a computer program; the processor 904 is configured to execute a computer program to implement the method for verifying image information provided in any of the above embodiments, and therefore, the system 900 for verifying image information includes all the advantages of the method for verifying image information provided in any of the above embodiments, which are not described herein again.
Example four
In an embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program is executed by a processor to implement the method for verifying the image information provided in any of the above embodiments, so that the computer-readable storage medium includes all the beneficial effects of the method for verifying the image information provided in any of the above embodiments, and details are not described herein.
In the description of the present invention, the terms "plurality" or "a plurality" refer to two or more, and unless otherwise specifically defined, the terms "upper", "lower", and the like indicate orientations or positional relationships based on the orientations or positional relationships illustrated in the drawings, and are only for convenience in describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention; the terms "connected," "mounted," "secured," and the like are to be construed broadly and include, for example, fixed connections, removable connections, or integral connections; may be directly connected or indirectly connected through an intermediate. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the description of the present invention, the description of the terms "one embodiment," "some embodiments," "specific embodiments," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In the present invention, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or 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 verifying image information, comprising:
receiving image information, and acquiring preset vehicle information and preset building information;
acquiring corresponding information of the vehicle to be detected according to the image information, and acquiring information of at least one building to be detected;
And matching the information of the vehicles to be detected with the preset vehicle information, matching any one of the information of the buildings to be detected with the preset building information, and determining that the image information passes the verification.
2. The method for verifying image information according to claim 1, wherein the step of obtaining at least one piece of building information to be detected specifically comprises:
analyzing the image information through a building detection model to obtain a characteristic image containing one or more buildings;
determining coordinate information corresponding to one or more buildings according to the characteristic images, and determining characteristic information corresponding to any building;
and generating a characterization vector corresponding to any one building according to the coordinate information and the characteristic information, and generating the information of the building to be detected according to the characterization vector.
3. The method for verifying image information according to claim 2, wherein the preset building information includes a preset characterization vector corresponding to a preset building;
determining the similarity between any one of the characterization vectors in the building information to be detected and the preset characterization vector;
and if the similarity between any one of the characterization vectors and the preset characterization vector is smaller than a preset threshold value, determining that the building information to be detected is matched with the preset building information.
4. The method for verifying the image information according to claim 1, wherein the step of obtaining the corresponding vehicle information to be detected according to the image information specifically includes:
analyzing the image information to obtain a vehicle image;
detecting the vehicle image through a license plate recognition model to determine license plate number information corresponding to the vehicle image;
and determining the vehicle information according to the license plate number information.
5. The method for verifying image information according to claim 4, wherein the preset vehicle information includes preset license plate number information;
and if the license plate number information is matched with the preset license plate number information, determining that the information of the vehicle to be detected is matched with the preset vehicle information.
6. The method for authenticating image information according to any one of claims 3 to 5, wherein before the step of receiving image information, the method for authenticating image information further comprises:
receiving a verification request, and acquiring first position information of a corresponding vehicle to be detected according to the verification request;
acquiring a preset building set, determining a landmark building closest to the vehicle to be detected in the preset building set according to the first position information, and determining the landmark building as the preset building;
And sending the second position information of the preset building to a terminal corresponding to the vehicle to be detected.
7. The method for verifying image information according to claim 6, wherein the step of receiving image information specifically includes:
and receiving the image information sent by the terminal, wherein the image information simultaneously comprises an image corresponding to the vehicle to be detected and an image corresponding to the preset building.
8. The method for authenticating image information according to any one of claims 2 to 5, wherein before the step of receiving image information, the method for authenticating image information further comprises:
acquiring a preset detection model and building sample data, wherein the building sample data comprises coordinate information and characteristic information corresponding to a plurality of buildings;
and generating a training set according to the building sample data, and training the preset detection model through the training set to obtain the building detection model.
9. A system for authenticating image information, comprising:
a memory configured to store a computer program;
a processor configured to execute the computer program to implement the method of authentication of image information according to any one of claims 1 to 8.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a method of authenticating image information according to any one of claims 1 to 8.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010180923.5A CN111860094A (en) | 2020-03-16 | 2020-03-16 | Method, system and computer readable storage medium for verifying image information |
| PCT/CN2021/072977 WO2021147927A1 (en) | 2020-01-21 | 2021-01-21 | Method and system for verifying vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010180923.5A CN111860094A (en) | 2020-03-16 | 2020-03-16 | Method, system and computer readable storage medium for verifying image information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111860094A true CN111860094A (en) | 2020-10-30 |
Family
ID=72985963
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010180923.5A Pending CN111860094A (en) | 2020-01-21 | 2020-03-16 | Method, system and computer readable storage medium for verifying image information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111860094A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021147927A1 (en) * | 2020-01-21 | 2021-07-29 | 北京嘀嘀无限科技发展有限公司 | Method and system for verifying vehicle |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160327A (en) * | 2015-09-16 | 2015-12-16 | 小米科技有限责任公司 | Building identification method and device |
| CN106529452A (en) * | 2016-11-04 | 2017-03-22 | 重庆市勘测院 | Mobile intelligent terminal building rapid identification method based on building three-dimensional model |
| CN108507541A (en) * | 2018-03-01 | 2018-09-07 | 广东欧珀移动通信有限公司 | Building recognition method and system and mobile terminal |
| CN108596277A (en) * | 2018-05-10 | 2018-09-28 | 腾讯科技（深圳）有限公司 | A kind of testing vehicle register identification method, apparatus and storage medium |
| CN109242018A (en) * | 2018-08-31 | 2019-01-18 | 平安科技（深圳）有限公司 | Image authentication method, device, computer equipment and storage medium |
| CN109974733A (en) * | 2019-04-02 | 2019-07-05 | 百度在线网络技术（北京）有限公司 | POI display methods, device, terminal and medium for AR navigation |
| CN110458168A (en) * | 2019-08-29 | 2019-11-15 | 上海眼控科技股份有限公司 | Processing method, device, computer equipment and the storage medium of vehicle detection report |
| CN110715668A (en) * | 2019-10-18 | 2020-01-21 | 南京速立世信息技术有限公司 | Device and method for positioning automobile based on street view map |
| CN110826627A (en) * | 2019-11-06 | 2020-02-21 | 广东三维家信息科技有限公司 | Image similarity measuring method and device and electronic equipment |
-
2020
- 2020-03-16 CN CN202010180923.5A patent/CN111860094A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160327A (en) * | 2015-09-16 | 2015-12-16 | 小米科技有限责任公司 | Building identification method and device |
| CN106529452A (en) * | 2016-11-04 | 2017-03-22 | 重庆市勘测院 | Mobile intelligent terminal building rapid identification method based on building three-dimensional model |
| CN108507541A (en) * | 2018-03-01 | 2018-09-07 | 广东欧珀移动通信有限公司 | Building recognition method and system and mobile terminal |
| CN108596277A (en) * | 2018-05-10 | 2018-09-28 | 腾讯科技（深圳）有限公司 | A kind of testing vehicle register identification method, apparatus and storage medium |
| CN109242018A (en) * | 2018-08-31 | 2019-01-18 | 平安科技（深圳）有限公司 | Image authentication method, device, computer equipment and storage medium |
| CN109974733A (en) * | 2019-04-02 | 2019-07-05 | 百度在线网络技术（北京）有限公司 | POI display methods, device, terminal and medium for AR navigation |
| CN110458168A (en) * | 2019-08-29 | 2019-11-15 | 上海眼控科技股份有限公司 | Processing method, device, computer equipment and the storage medium of vehicle detection report |
| CN110715668A (en) * | 2019-10-18 | 2020-01-21 | 南京速立世信息技术有限公司 | Device and method for positioning automobile based on street view map |
| CN110826627A (en) * | 2019-11-06 | 2020-02-21 | 广东三维家信息科技有限公司 | Image similarity measuring method and device and electronic equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021147927A1 (en) * | 2020-01-21 | 2021-07-29 | 北京嘀嘀无限科技发展有限公司 | Method and system for verifying vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108009528B (en) | Triple Loss-based face authentication method and device, computer equipment and storage medium | |
| CN113538714B (en) | Parking lot control method, system and computer readable storage medium | |
| CN105260779B (en) | Personalized method and device based on label | |
| CN109685956B (en) | Method, device, equipment and storage medium for scanning and unlocking shared bicycle license plate | |
| KR102085489B1 (en) | Hybrid breath alcohol analyzing management system | |
| KR20120024073A (en) | Apparatus and method for providing augmented reality using object list | |
| CN105354773A (en) | System for evidence preservation and verification of traffic accident scene | |
| CN110826484A (en) | Vehicle weight recognition method and device, computer equipment and model training method | |
| CN111444798A (en) | Method and device for identifying driving behavior of electric bicycle and computer equipment | |
| CN111444808A (en) | Image-based accident liability assignment method and device, computer equipment and storage medium | |
| CN111860477A (en) | Vehicle information verification method and system, terminal, and computer-readable storage medium | |
| CN111415428A (en) | Vehicle self-service payment realization method and device based on ETC (electronic toll Collection) equipment | |
| CN113677409A (en) | Treasure hunting game guiding technology | |
| US20150281673A1 (en) | Method for authenticating an image capture of a three-dimensional entity | |
| CN111860094A (en) | Method, system and computer readable storage medium for verifying image information | |
| CN105913264A (en) | Face payment device based on fingerprint auxiliary identity identification | |
| CN111401287A (en) | Vehicle authentication method and device in parking charging system | |
| CN113315639A (en) | Identity authentication system and method | |
| CN110532910B (en) | Unmanned aerial vehicle RFID auxiliary scanning control method and device | |
| CN111401258B (en) | Living body detection method and device based on artificial intelligence | |
| CN109102581B (en) | Card punching method, device, system and storage medium | |
| CN116452006A (en) | Wind control method, device, computer equipment and medium for new activity of driver | |
| CN209471482U (en) | A kind of shared bicycle auto-unlocking control system based on recognition of face | |
| CN111599208A (en) | Parking lot access management method, device and equipment and readable storage medium | |
| CN114821845B (en) | Card punching method and device |
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 |