CN111414866A

CN111414866A - Vehicle application form detection method and device, computer equipment and storage medium

Info

Publication number: CN111414866A
Application number: CN202010213677.9A
Authority: CN
Inventors: 周康明; 党银强
Original assignee: Shanghai Eye Control Technology Co Ltd
Current assignee: Shanghai Eye Control Technology Co Ltd
Priority date: 2020-03-24
Filing date: 2020-03-24
Publication date: 2020-07-14

Abstract

The application relates to a method and a device for detecting a vehicle application form, computer equipment and a storage medium. The method comprises the following steps: inputting an image to be detected for displaying application form information of a vehicle into a preset text detection model to obtain a plurality of corresponding text strips; determining a target text strip, and judging whether a frame to be selected in the target text strip is selected or not to obtain a verification result; obtaining a signature subimage in the image to be detected according to each text strip, and detecting whether the signature subimage contains a user signature to obtain a signature detection result; and detecting the image to be detected according to the verification result and the signature detection result to obtain a detection result. By adopting the method, the efficiency of obtaining the detection result of the image to be detected can be improved.

Description

Vehicle application form detection method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of image processing technologies, and in particular, to a method and an apparatus for detecting a vehicle application form, a computer device, and a storage medium.

Background

With the continuous development of economy, the keeping quantity of motor vehicles is continuously increased, and further, the annual inspection workload of the motor vehicles is increased.

The registration, transfer, deregistration, transfer to application form identification of the motor vehicle is an important detection in the annual inspection work of the motor vehicle. In the traditional technology, the method for identifying the motor vehicle application form in the annual inspection work of the motor vehicle is mainly carried out in a manual mode.

However, the conventional method for identifying the motor vehicle application form has the problem of low identification efficiency.

Disclosure of Invention

In view of the above, it is necessary to provide a method, an apparatus, a computer device and a storage medium for detecting a vehicle application form, which can improve the efficiency of identifying a vehicle application form.

A method of detecting a vehicle application form, the method comprising:

inputting an image to be detected into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected; the image to be detected is used for displaying application form information of the vehicle;

determining a target text bar from text information corresponding to each text bar according to preset standard target text information, and judging whether a frame to be selected in the target text bar is selected or not to obtain a verification result;

according to each text bar, obtaining a signature subimage in the image to be detected, and detecting whether the signature subimage contains a user signature to obtain a signature detection result;

and detecting the image to be detected according to the verification result and the signature detection result to obtain a detection result.

In one embodiment, the determining whether the frame to be selected in the target text strip is selected to obtain a check result includes:

carrying out binarization processing on the target text strip to obtain a processed text strip;

performing connected domain segmentation on the processed text strip, and determining a frame to be selected in the target text strip from the segmented connected domain;

and judging whether the frame to be selected is selected according to the target pixel of the frame to be selected in the target text strip to obtain the verification result.

In one embodiment, the determining, according to a target pixel of a frame to be selected in the target text strip, whether the frame to be selected is selected to obtain the verification result includes:

acquiring the number of the target pixels in a preset area in the frame to be selected;

and judging whether the number of the target pixels is larger than a preset pixel threshold value, if so, determining that the check result is that the frame to be selected is selected.

In one embodiment, the obtaining a signature sub-image in the image to be detected according to each text entry includes:

determining a first coordinate and a second coordinate of a signature area in the image to be detected according to a text bar containing the signature of a principal of a motor vehicle owner and a text bar containing a date in each text bar;

and obtaining a signature sub-image in the image to be detected according to the first coordinate and the second coordinate.

In one embodiment, before the image to be detected is input into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected, the method further includes:

acquiring an included angle value between a boundary line in the vertical direction of the image to be detected and a horizontal line;

according to the included angle value, performing first rotation processing on the image to be detected, and rotating the image to be detected until the included angle value is 90 degrees to obtain a first rotation image;

judging whether the first rotation image comprises an inverted text or not, if so, performing second rotation processing on the first rotation image to obtain a second rotation image;

in the text detection model that the image input of will waiting to examine is predetermine, obtain a plurality of text strips that the image of waiting to examine corresponds include:

and inputting the second rotating image into the preset text detection model to obtain a plurality of text strips corresponding to the image to be detected.

In one embodiment, the determining whether the first rotated image includes an inverted text, and if so, performing a second rotation process on the first rotated image to obtain a second rotated image includes:

inputting the first rotating image into a preset text direction classification model, judging whether the text in the first rotating image comprises an inverted text or not, and if so, determining the number of the inverted text;

and if the number of the inverted texts is larger than a preset threshold value, performing second rotation processing on the first rotation image according to a preset rotation angle to obtain a second rotation image.

In one embodiment, the detecting the image to be detected according to the verification result and the signature detection result to obtain a detection result includes:

and if the verification result and the signature detection result both pass, determining that the detection result of the image to be detected passes.

A vehicle application form detection apparatus, the apparatus comprising:

the first acquisition module is used for inputting an image to be detected into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected; the image to be detected is used for displaying application form information of the vehicle;

the determining module is used for determining a target text bar from text information corresponding to each text bar according to preset standard target text information, judging whether a frame to be selected in the target text bar is selected or not, and obtaining a checking result;

the signature detection module is used for obtaining a signature sub-image in the image to be detected according to each text bar, and detecting whether the signature sub-image contains a user signature or not to obtain a signature detection result;

and the image detection module is used for detecting the image to be detected according to the verification result, the text information verification result and the signature detection result to obtain a detection result.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

The detection method, the device, the computer equipment and the storage medium for the vehicle application form input the image to be detected for displaying the application form information of the vehicle into a preset text detection model, can quickly obtain a plurality of text strips corresponding to the image to be detected, improve the efficiency of obtaining the plurality of text strips corresponding to the image to be detected, further improve the efficiency of determining the target text strip from the text information corresponding to each text strip according to the preset standard target text information, further improve the efficiency of judging the check result of whether the frame to be selected in the target text strip is selected, simultaneously, can quickly obtain the signature subimage in the image to be detected according to the plurality of text strips corresponding to the image to be detected, further improve the detection efficiency of whether the signature subimage contains the user signature, namely improve the efficiency of obtaining the signature detection result, and then, the image to be detected can be quickly detected according to the verification result and the signature detection result, so that the efficiency of obtaining the detection result of the image to be detected is improved.

Drawings

FIG. 1 is a schematic diagram of an internal structure of a computer device according to an embodiment;

FIG. 2 is a schematic flow chart diagram illustrating a method for detecting a vehicle application form, according to an embodiment;

FIG. 3 is a schematic flow chart diagram illustrating a method for detecting a vehicle application form according to another embodiment;

FIG. 4 is a schematic flow chart diagram illustrating a method for detecting a vehicle application form according to another embodiment;

FIG. 5 is a schematic flow chart diagram illustrating a method for detecting a vehicle application form according to another embodiment;

FIG. 6 is a schematic flow chart diagram illustrating a method for detecting a vehicle application form according to another embodiment;

FIG. 7 is a flowchart illustrating a method for detecting a vehicle application form according to another embodiment;

FIG. 8 is a flowchart illustrating a method for detecting a vehicle application form according to one embodiment;

fig. 9 is a schematic structural diagram of a vehicle certification verification device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The detection method of the vehicle application form provided by the embodiment of the application can be applied to computer equipment shown in fig. 1. The computer device comprises a processor and a memory connected by a system bus, wherein a computer program is stored in the memory, and the steps of the method embodiments described below can be executed when the processor executes the computer program. Optionally, the computer device may further comprise a network interface, a display screen and an input device. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a nonvolatile storage medium storing an operating system and a computer program, and an internal memory. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. Optionally, the computer device may be a server, a personal computer, a personal digital assistant, other terminal devices such as a tablet computer, a mobile phone, and the like, or a cloud or a remote server, and the specific form of the computer device is not limited in the embodiment of the present application.

In the method for detecting a vehicle application form provided in the embodiment of the present application, an execution subject may be a device for detecting a vehicle application form, and the device for detecting a vehicle application form may be implemented as part or all of a computer device by software, hardware, or a combination of software and hardware. In the following method embodiments, the execution subject is a computer device as an example.

The following describes the technical solution of the present invention and how to solve the above technical problems with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.

FIG. 2 is a flowchart illustrating a method for detecting a vehicle application form according to an embodiment. The embodiment relates to a specific implementation process for detecting an image to be detected by computer equipment according to a verification result and a signature detection result of the image to be detected to obtain a detection result. As shown in fig. 2, the method may include:

s201, inputting an image to be detected into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected; the image to be detected is used for displaying the application form information of the vehicle.

The text bar refers to a continuous text belonging to the same line, and no space is left in the middle of the text. The image to be detected is used for displaying the application form information of the vehicle.

Specifically, the computer device inputs the acquired image to be detected into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected. It should be noted that the text information of the image to be detected is stored in the plurality of text pieces corresponding to the image to be detected. Optionally, the computer device may obtain the image to be detected from a server storing the application form information image of the vehicle, or may acquire the image to be detected in real time through a shooting device connected to the computer device.

Optionally, the preset text detection model may be a Progressive extension Network (PSEnet) model, where the PSEnet Network has a simple Network structure and runs at a speed that can meet the requirement of real-time use, and the text detection precision of the PSEnet Network is high (for example, the PSEnet Network can accurately detect a too-long text string, a too-short text string, or two character strings close to each other), and in addition, the PSEnet Network can also accurately detect a text of any Shape.

It can be understood that the text detection model is a model that has been trained in advance, optionally, the computer device may obtain sample images of different sizes that show application form information of the vehicle and are acquired under different shooting conditions (such as illumination, rotation angle, and text direction), for example, the number of the sample images of different sizes may be 20000, a polygon frame is used to mark the text in the sample images according to text bars, so as to obtain labeled images corresponding to the sample images, i.e., a sample set, where the labeled information is a one-dimensional 10-length array point set, the point set is derived from a circumscribed polygon of the text bars, the circumscribed polygon of the text bars can be drawn by the points, the sample set is divided into a training set and a testing set, where the training set accounts for 95%, the testing set accounts for 5%, each sample is randomly allocated, and the sample set is converted into standard data required by the preset initial text detection model, setting hyper-parameters in an initial text detection model, such as basic learning rate, weight iteration rate, change degree value of learning rate each time, maximum network iteration times, model overfitting prevention strategy, weight initialization mode, model saving for each iteration time and the like, inputting sample images in a sample set into the initial text detection model, training the initial text detection model until the initial text detection model reaches preset iteration times, stopping training, testing all initial text detection models saved in the training process in a test set, selecting an optimal text detection model according to a test result of each initial text detection model and a labeled image corresponding to the sample image, and determining the selected optimal text detection model as the text detection model.

S202, determining a target text bar from the text information corresponding to each text bar according to preset standard target text information, and judging whether a frame to be selected in the target text bar is selected or not to obtain a verification result.

Specifically, the computer device can input each text bar corresponding to the image to be detected into a preset text recognition model, recognize text information (including characters, numbers, characters and other information) in each text bar, determine a target text bar from the text information corresponding to a plurality of text bars corresponding to the image to be detected according to preset standard target text information, and judge whether a frame to be selected in the determined target text bar is selected to obtain a verification result. The standard target text information comprises standard application item text information and standard vehicle use property text information. Optionally, the computer device may extract text information in each text strip, compare the extracted text information with preset standard target text information, find text information that is the same as the standard target text information from the text information corresponding to each text strip of the image to be detected, and determine the corresponding text strip as the target text strip. For example, the computer device may compare each extracted text message with the standard application item text message, and determine a text strip that is the same as the standard application item text message as the target text strip. Optionally, the frame to be selected may be a rectangular frame, a circular frame, or a triangular frame. Optionally, the to-be-selected frame may be a frame with a hook, a circle black dot, or a black solid frame filled with the to-be-selected frame. Optionally, the computer device may determine whether the frame to be selected in the target text strip is selected according to the number of the target pixels in the frame to be selected, so as to obtain the verification result. Optionally, the check result may be passed or not passed. Exemplarily, if the computer device detects that the to-be-selected box in the text bar corresponding to the determined application item text information is selected, the verification result of the application item may be marked as 1, otherwise, the verification result is marked as 0; if the computer device detects that the frame to be selected in the text bar corresponding to the text information of the determined vehicle use property is selected, the check result of the vehicle use property can be marked as 1, otherwise, the check result is marked as 0; if the verification result of the application item and the verification result of the vehicle use property are marked as 1, the computer equipment determines that the verification result is passed, and otherwise, determines that the verification result is not passed.

Optionally, the preset text recognition model may be a long-Short Term Memory network (L ong-Term Memory, L STM) model, and the computer device may send each text bar corresponding to the image to be detected into the L STM model to recognize, and recognize information such as characters, numbers, and characters in each text bar corresponding to the image to be detected.

S203, obtaining a signature sub-image in the image to be detected according to the text pieces, and detecting whether the signature sub-image contains the user signature to obtain a signature detection result.

Specifically, the computer device obtains a signature sub-image in the image to be detected according to each text strip corresponding to the image to be detected, and detects whether the signature sub-image contains a user signature to obtain a signature detection result. Optionally, the computer device may obtain a signature sub-image in the image to be detected according to a text strip that includes a signature of a principal of the owner of the motor vehicle in each text strip corresponding to the image to be detected. Optionally, the computer device may input the obtained signature sub-image into the signature detection model to obtain a signature detection result of the image to be detected. It can be understood that the signature detection model is a model that has been trained in advance, optionally, the computer device may acquire sample images with different sizes that show application form information of the vehicle and are acquired under different shooting conditions (such as illumination, rotation angle, and text direction), for example, 20000 sample images are acquired, a rectangular frame is used to mark a signature region in the sample image, and a signature labeled image corresponding to the sample image, that is, a sample set, is obtained, wherein the labeled information is a one-dimensional array [ class, x, y, width, height ], where class in the array represents a target class (that is, a class of a signature), x and y represent horizontal and vertical coordinates, width, and height of an upper left corner of the rectangular frame in the sample image, respectively represent width and height of the rectangular frame in the sample image, and the sample set is divided into a training set and a test set, 95% of training set and 5% of testing set, randomly distributing each sample, converting the sample set into standard data required by a preset initial signature detection model, setting hyper-parameters in the initial signature detection model, such as basic learning rate, weight iteration rate, change degree value of learning rate each time, maximum network iteration times, model overfitting prevention strategy, weight initialization mode, model saving for each iteration time, inputting sample images in the sample set into the initial signature detection model to obtain a sample detection result, wherein the sample detection result is used for indicating whether the sample images include signatures, training the initial signature detection model until the initial signature detection model reaches the preset iteration times, stopping training, and testing all initial signature detection models saved in the training process in the testing set, and selecting an optimal signature detection model according to the test result of each initial signature detection model and the signature marking image corresponding to the sample image, and determining the selected optimal signature detection model as the signature detection model.

And S204, detecting the image to be detected according to the verification result and the signature detection result to obtain a detection result.

Specifically, the computer device detects the image to be detected according to the obtained verification result and the signature detection result to obtain the detection result of the image to be detected. Optionally, the detection result obtained by the computer device may be that the detection is passed or that the detection is not passed. Further, if the detection result obtained by the computer device is a failure of detection, optionally, the reason for the failure of detection may be at least one of failure of verification result and failure of signature detection result.

In this embodiment, the computer device inputs the image to be detected for displaying the application form information of the vehicle into the preset text detection model, so as to rapidly obtain a plurality of text strips corresponding to the image to be detected, thereby improving the efficiency of obtaining the plurality of text strips corresponding to the image to be detected, further improving the efficiency of determining the target text strip from the text information corresponding to each text strip according to the preset standard target text information, further improving the efficiency of determining the check result of whether the frame to be selected in the target text strip is selected, meanwhile, according to the plurality of text strips corresponding to the image to be detected, the signature in the image to be detected can be rapidly obtained, further improving the detection efficiency of whether the signature sub-image contains the user signature, namely improving the efficiency of obtaining the signature detection result, further according to the check result and the signature detection result, and the image to be detected is quickly detected, so that the efficiency of obtaining the detection result of the image to be detected is improved.

Fig. 3 is a schematic flowchart of a method for detecting a vehicle application form according to another embodiment. The embodiment relates to a specific implementation process of judging whether a frame to be selected in a target text strip is selected by computer equipment to obtain a verification result. As shown in fig. 3, on the basis of the foregoing embodiment, as an optional implementation manner, the determining in S202 whether the frame to be selected in the target text entry is selected, and obtaining the verification result includes:

s301, performing binarization processing on the target text bar to obtain a processed text bar.

Specifically, the computer device performs binarization processing on the obtained target text bar to obtain a processed text bar. Optionally, the computer device may perform binarization processing on the target text strip according to pixels in the target text strip, and exemplarily, the computer device may mark an area of black pixels in the target text strip as 1, and mark other areas as 0, to obtain a processed text strip.

S302, performing connected domain segmentation on the processed text strip, and determining a frame to be selected in the target text strip from the segmented connected domain.

It can be understood that, after the computer device performs binarization processing on the target text strip, the obtained processed text strip includes a plurality of connected domains, that is, the colors of pixels in the same connected domain are the same. Specifically, the computer device performs connected domain segmentation on the processed text strip, and determines a frame to be selected in the target text strip from the segmented connected domain. Optionally, the computer device may determine the first separated connected domain as a candidate box in the target text strip. It can be understood that the target text bar is also a continuous text belonging to the same line, and there is no space in the middle of the text, the first text in the target text bar is the frame to be selected in the target text bar, and the first connected domain thus divided is the frame to be selected in the target text bar.

S303, judging whether the frame to be selected is selected according to the target pixel of the frame to be selected in the target text strip, and obtaining a verification result.

Specifically, the computer device judges whether the frame to be selected is selected according to the target pixel of the frame to be selected in the target text strip, and obtains the check result. Alternatively, the target pixel may be a black pixel, or may be an ink pixel or a pixel of another color. Optionally, the computer device may determine whether the frame to be selected is selected according to the number of target pixels of the frame to be selected in the target text strip, so as to obtain a verification result of the image to be detected.

In this embodiment, the computer device performs binarization processing on the target text strip, can quickly obtain the processed text strip, improves the efficiency of the computer device in segmenting the processed text strip by the connected domain, and can quickly determine the frame to be selected in the target text strip from the segmented connected domain, so that the computer device can quickly judge whether the frame to be selected is selected according to the target pixel of the frame to be selected in the target text strip, and improves the efficiency of obtaining the check result of the image to be detected.

Fig. 4 is a flowchart illustrating a method for detecting a vehicle application form according to another embodiment. The embodiment relates to a specific implementation process of judging whether a frame to be selected in a target text strip is selected or not by computer equipment according to target pixels of the frame to be selected in the target text strip to obtain a verification result. As shown in fig. 3, on the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S303 includes:

s401, the number of target pixels in a preset area in the frame to be selected is obtained.

Specifically, the computer device obtains the number of target pixels in a preset area in the frame to be selected. Optionally, the preset area may be one tenth of the area of the center of the frame to be selected. Optionally, the computer device may count the target pixels in the preset area to obtain the number of the target pixels in the preset area.

S402, judging whether the number of the target pixels is larger than a preset pixel threshold value or not, and if yes, determining that the check result is that the frame to be selected is selected.

Specifically, the computer device judges whether the number of target pixels in a preset area in the frame to be selected is larger than a preset pixel threshold value, if so, the verification result of the image to be detected is determined that the frame to be selected is selected, otherwise, the verification result of the image to be detected is determined that the frame to be selected is not selected. Optionally, the preset pixel threshold may be 30, and for example, if the number of the target pixels in the preset area in the frame to be selected is 40 and is greater than the preset pixel threshold, the computer device determines that the check result of the image to be detected is that the frame to be selected is selected.

In this embodiment, the computer device can quickly and accurately obtain the number of the target pixels in the preset area in the frame to be selected, so that the efficiency and accuracy for obtaining the number of the target pixels in the preset area in the frame to be selected are improved, and then whether the number of the target pixels is greater than the preset pixel threshold value can be quickly and accurately judged, so that the computer device can quickly and accurately obtain the verification result of the image to be detected, namely, the efficiency and accuracy for obtaining the verification result are improved.

Fig. 5 is a flowchart illustrating a method for detecting a vehicle application form according to another embodiment. The embodiment relates to a specific implementation process for obtaining a signature sub-image in an image to be detected by computer equipment according to a plurality of text pieces corresponding to the image to be detected. As shown in fig. 5, on the basis of the foregoing embodiment, as an optional implementation manner, the obtaining, according to each text entry, a signature sub-image in the image to be detected in step S203 includes:

s501, determining a first coordinate and a second coordinate of a signature area in the image to be detected according to the text bar containing the signature of the principal of the owner of the motor vehicle and the text bar containing the date in each text bar.

Specifically, the computer device determines a first coordinate and a second coordinate of a signature area in the image to be detected according to a text strip containing a signature of a principal of a motor vehicle owner and a text strip containing a date in a plurality of text strips corresponding to the image to be detected. Optionally, the computer device searches text information corresponding to a plurality of text entries of the image to be detected in a traversal order from top to bottom and from left to right, finds a text entry containing a signature of a principal of the owner of the motor vehicle, determines a coordinate of an upper left corner point of the text entry as a first coordinate p1(x, y) of a signature region in the image to be detected, finds a text entry containing a year, month and day and located at a lower right side of the text entry containing the signature of the principal of the owner of the motor vehicle in a traversal order from top to bottom and from left to right, and determines a coordinate of a lower right corner point of the text entry as a second coordinate p2(x, y) of the signature region in the image to be detected.

And S502, obtaining a signature sub-image in the image to be detected according to the first coordinate and the second coordinate.

Specifically, the computer device obtains the signature subimage in the image to be detected according to the obtained first coordinate and the second coordinate. Optionally, the computer device may determine the obtained first coordinate as an upper left corner point coordinate of the signature sub-image in the image to be detected, and determine the obtained second coordinate as a lower right corner point coordinate of the signature sub-image in the image to be detected, so as to obtain the signature sub-image in the image to be detected.

In this embodiment, the computer device can accurately determine the first coordinate and the second coordinate of the signature area in the image to be detected according to the text bar containing the signature of the principal of the motor vehicle owner and the text bar containing the date in each text bar of the image to be detected, and further can accurately determine the signature subimage in the image to be detected according to the determined first coordinate and second coordinate, so that the accuracy of the obtained signature subimage in the image to be detected is improved.

In some scenes, the direction of the obtained image to be detected and the direction of characters in the image to be detected deflect due to the influence of the shooting angle, the shooting illumination condition and the like of the obtained image to be detected, so that the image to be detected needs to be corrected before the application form information of the vehicle displayed by the image to be detected is detected. Fig. 6 is a flowchart illustrating a method for detecting a vehicle application form according to another embodiment. The embodiment relates to a specific implementation process for correcting an image to be detected by computer equipment. As shown in fig. 6, on the basis of the foregoing embodiment, as an optional implementation manner, before S201, the method further includes:

s601, an included angle value between a boundary line in the vertical direction of the image to be detected and a horizontal line is obtained.

Specifically, the computer device acquires an included angle value between a boundary line in the vertical direction and a horizontal line of an image to be detected. Optionally, the computer device may detect a straight line in the image to be detected, obtain a longest straight line from the detected straight lines, where the straight line is a left boundary line or a right boundary line of the image to be detected, and then calculate an included angle value between the straight line and the horizontal line to obtain an included angle value between the vertical boundary line and the horizontal line of the image to be detected.

S602, according to the included angle value, carrying out first rotation processing on the image to be detected, and rotating the image to be detected until the included angle value is 90 degrees to obtain a first rotation image.

Specifically, the computer device performs a first rotation process on the image to be detected according to the obtained included angle value, and rotates the image to be detected until the included angle value is 90 degrees, so as to obtain a first rotation image. Optionally, the first rotation processing may be counterclockwise rotation processing or clockwise rotation processing, as long as an included angle between a vertical boundary line and a horizontal line of the image to be detected is rotated to 90 degrees.

S603, judging whether the first rotation image comprises an inverted text, if so, performing second rotation processing on the first rotation image to obtain a second rotation image.

Specifically, the computer device determines whether the first rotated image includes an inverted text, and if it is determined that the first rotated image includes the inverted text, performs a second rotation process on the first rotated image to obtain a second rotated image. Optionally, the computer device may input the first rotated image into a preset text direction classification model, determine whether the text in the first rotated image includes an inverted text, determine the number of the inverted text if the text in the first rotated image includes the inverted text, and perform second rotation processing on the first rotated image according to a preset rotation angle if the number of the inverted text in the first rotated image is greater than a preset threshold value, to obtain a second rotated image. Alternatively, the preset threshold may be 60%, and the preset rotation angle may be 180 °.

The above S201 includes: and inputting the second rotating image into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected.

Specifically, the computer device inputs the obtained second rotation image into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected. It can be understood that the second rotated image is an image that has been corrected, the directions of the text strips corresponding to the second rotated image are all horizontal directions, and there is no inclination, and the directions of the text in the second rotated image are all positive text directions, that is, the directions of the text strips corresponding to the image to be detected are all horizontal directions, and the directions of the text strips are all positive text directions.

In this embodiment, the direction of the text strip corresponding to the second rotated image obtained by the computer device is the horizontal direction, and the text direction in the second rotated image is the positive text direction, so that the obtained second rotated image is input into the preset text detection model, the accuracy of the obtained plurality of text strips corresponding to the image to be detected is improved, and then the plurality of text strips corresponding to the image to be detected can be accurately detected, and the accuracy of the obtained detection result of the image to be detected is improved.

On the basis of the foregoing embodiment, as an optional implementation manner, the foregoing S204 includes: and if the verification result and the signature detection result both pass, determining that the detection result of the image to be detected is pass.

Specifically, if the check result and the signature detection result obtained by the computer equipment both pass, the detection result of the image to be detected is determined to be passing. That is, if the frame to be selected in the target text strip in the image to be detected, which is obtained by the computer device, is selected, and the signature sub-image in the image to be detected contains the user signature, the detection result of the image to be detected is determined to be passed. Optionally, if the detection result of the image to be detected obtained by the computer device is that the image does not pass, the reason that the detection does not pass is returned. Optionally, the reason why the detection result of the image to be detected is not passed may be that at least one of a frame to be selected in the target text strip in the image to be detected is not selected, and a signature sub-image in the image to be detected does not include a user signature.

In this embodiment, if the obtained verification result and the signature detection result both pass, the computer device determines that the detection result of the image to be detected passes, so that the text information in the image to be detected can be accurately detected, and the accuracy of the obtained detection result of the image to be detected is further improved.

Optionally, on the basis of the above embodiment, if the vehicle application form is to be completely detected, the computer device may further check text information in each text strip except for the target text strip to obtain a text information check result of the image to be detected. And the text information checking result is used for indicating whether the text information corresponding to each text strip except the target text strip is accurate or not. Optionally, the computer device may respectively determine whether the text information in each text strip except the target text strip is consistent with the corresponding standard text information, so as to obtain a text information check result of the image to be detected. Exemplarily, if the computer device obtains that the text information in each text strip except the target text strip is consistent with the corresponding standard text information, the verification result of the text information in each text strip except the target text strip is marked as a first value, and if the verification result of the text information in each text strip except the target text strip is the first value, the verification result of the text information is determined to be passed. Exemplarily, the computer device may traverse each text strip except for the target text strip from top to bottom and from left to right, find a text strip in which the text information in each text strip except for the target text strip contains several words, such as "motor vehicle registration transfer logout registration transfer into application form", and mark the verification result of the header name comparison as a first value if found, that is, mark the verification result of the header name comparison as 1, otherwise, mark the verification result as 0; the computer equipment can traverse each text strip from top to bottom and from left to right, search for text strips of which the text information contains a plurality of characters of 'number plate type', select the first text information on the right side of the text strip after finding the text strip, wherein the text information is number plate type information, compare the number plate type information with standard number plate type information acquired from a server, mark the verification result of the number plate type as a first value if the comparison result is consistent, namely mark the verification result of the number plate type as 1, otherwise mark the verification result as 0; the computer equipment can traverse each text strip from top to bottom and from left to right, search for text strips of which the text information contains a plurality of words of 'number plate number', select the first text information on the right side of the text strips, wherein the text information is the number plate number information, compare the number plate number information with the standard number plate number information acquired from the server, mark the verification result of the number plate number as a first value if the comparison result is consistent, namely mark the verification result of the number plate number as 1, otherwise mark the verification result as 0; the computer equipment can traverse all text strips from top to bottom and from left to right, search for text strips of which the text information contains a plurality of characters of 'brand models', select the first text information on the right side of the text strips, wherein the text information is brand model information, compare the brand model information with standard brand model information acquired from a server, mark the verification result of the brand model as a first value if the comparison result is consistent, namely mark the verification result of the brand model as 1, otherwise mark the verification result as 0; the computer equipment can traverse all text strips from top to bottom and from left to right, search text strips of which the text information contains a plurality of characters of 'vehicle identification codes' in all the text strips, select the first text information on the right side of the text strips, wherein the text information is the vehicle identification code information, compare the vehicle identification code information with the standard vehicle identification code information acquired from the server, mark the verification result of the vehicle identification code as a first value if the comparison result is consistent, namely mark the verification result of the vehicle identification code as 1, otherwise mark the verification result as 0; after traversing all text strips except the target text strip by the computer equipment, if the verification results of the header name, the number plate type, the number plate number, the brand model and the vehicle identification code obtained are all first values, namely the verification results of the header name, the number plate type, the number plate number, the brand model and the vehicle identification code are all marked as 1, determining that the text information verification result of the image to be detected is passed, and otherwise, determining that the text information verification result of the image to be detected is not passed; and if the verification result, the text information verification result and the signature detection result obtained by the computer equipment are all passed, determining that the detection result of the image to be detected is passed. That is, if the frame to be selected in the target text strip in the image to be detected, which is obtained by the computer device, is selected, and the text information corresponding to each text strip except the target text strip is accurate, and the signature sub-image in the image to be detected contains the user signature, it is determined that the detection result of the image to be detected is passed. Optionally, if the detection result of the image to be detected obtained by the computer device is that the image does not pass, the reason that the detection does not pass is returned. Optionally, the reason why the detection result of the image to be detected is not passed may be at least one of that a to-be-selected frame in a target text bar in the image to be detected is not selected, text information corresponding to each text bar other than the target text bar is inaccurate, and a signature sub-image in the image to be detected does not include a user signature.

Fig. 7 is a flowchart illustrating a method for detecting a vehicle application form according to another embodiment. As shown in fig. 7, the detection method for performing complete detection on a vehicle application form provided by the present application may include: obtaining answers of an image to be detected and standard data to be compared; the image to be detected is used for displaying application form information of the vehicle; the method comprises the steps of performing rotation correction on an image to be detected according to the direction of a text in the image to be detected, detecting and identifying the text in the image to be detected (including header name comparison, application item comparison, number plate type comparison, number plate number comparison, brand model comparison, vehicle identification code comparison, use property comparison, signature detection and the like), and detecting vehicle application form information displayed by the image to be detected according to a detection comparison result of the image to be detected to obtain a detection result.

It should be noted that, for the description of the detection method of the vehicle application table in this embodiment, reference may be made to the description related to the foregoing embodiment, and the effect is similar, and no further description is given here in this embodiment.

To facilitate understanding by those skilled in the art, the following detailed description is provided for a detection method for performing a complete detection on a vehicle application form, which may include:

s801, acquiring an included angle value between a boundary line in the vertical direction of an image to be detected and a horizontal line, performing first rotation processing on the image to be detected according to the included angle value, and rotating the image to be detected until the included angle value is 90 degrees to obtain a first rotation image; the image to be detected is used for displaying application form information of the vehicle;

s802, inputting the first rotating image into a preset text direction classification model, judging whether the text in the first rotating image comprises an inverted text or not, if so, determining the number of the inverted text, and if the number of the inverted text is greater than a preset threshold value, performing second rotating processing on the first rotating image according to a preset rotating angle to obtain a second rotating image;

s803, inputting the second rotation image into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected;

s804, determining a target text bar from the text information corresponding to each text bar according to preset standard target text information;

s805, performing binarization processing on the target text strip to obtain a processed text strip, performing connected domain segmentation on the processed text strip, and determining a frame to be selected in the target text strip from the segmented connected domain;

s806, acquiring the number of target pixels in a preset area in the frame to be selected; judging whether the number of the target pixels is larger than a preset pixel threshold value or not, and if so, determining that the check result of the image to be detected is that a frame to be selected is selected;

s807, verifying the text information in each text strip except the target text strip to obtain a text information verification result of the image to be detected; the text information checking result is used for indicating whether the text information corresponding to each text strip except the target text strip is accurate or not;

s808, determining a first coordinate and a second coordinate of the signature area in the image to be detected according to the text strip containing the signature of the principal of the motor vehicle owner and the text strip containing the date in each text strip;

s809, obtaining a signature subimage in the image to be detected according to the first coordinate and the second coordinate, and detecting whether the signature subimage contains the user signature to obtain a signature detection result;

and S810, if the verification result, the text information verification result and the signature detection result are all passed, determining that the detection result of the image to be detected is passed.

It should be noted that, for the descriptions in S801 to S810, reference may be made to the descriptions related to the foregoing embodiments, and the effects are similar, and this embodiment is not described again here.

It should be understood that although the various steps in the flow charts of fig. 2-8 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-8 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.

Fig. 9 is a schematic structural diagram of a vehicle certification verification device according to an embodiment. As shown in fig. 6, the apparatus may include: a first acquisition module 10, a determination module 11, a signature detection module 12 and an image detection module 13.

Specifically, the first obtaining module 10 is configured to input an image to be detected into a preset text detection model, so as to obtain a plurality of text strips corresponding to the image to be detected; the image to be detected is used for displaying application form information of the vehicle;

the determining module 11 is configured to determine a target text bar from text information corresponding to each text bar according to preset standard target text information, and determine whether a frame to be selected in the target text bar is selected, so as to obtain a verification result;

the signature detection module 12 is configured to obtain a signature sub-image in the image to be detected according to each text entry, and detect whether the signature sub-image includes a user signature, so as to obtain a signature detection result;

and the image detection module 13 is configured to detect the image to be detected according to the verification result, the text information verification result and the signature detection result, so as to obtain a detection result.

The detection apparatus for a vehicle application form provided in this embodiment may implement the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.

On the basis of the foregoing embodiment, optionally, the determining module 11 includes: the device comprises a first processing unit, a first determining unit and a first judging unit.

Specifically, the first processing unit is configured to perform binarization processing on a target text bar to obtain a processed text bar;

the first determining unit is used for carrying out connected domain segmentation on the processed text strip and determining a frame to be selected in the target text strip from the segmented connected domain;

and the first judgment unit is used for judging whether the frame to be selected is selected according to the target pixel of the frame to be selected in the target text strip to obtain a verification result.

On the basis of the foregoing embodiment, optionally, the first determining unit is specifically configured to obtain the number of target pixels in a preset area in the frame to be selected; and judging whether the number of the target pixels is larger than a preset pixel threshold value or not, and if so, determining that the check result is that the frame to be selected is selected.

On the basis of the foregoing embodiment, optionally, the signature detection module 12 includes: a second determining unit and an obtaining unit.

The second determining unit is used for determining a first coordinate and a second coordinate of the signature area in the image to be detected according to the text strip containing the signature of the principal of the motor vehicle owner and the text strip containing the date in each text strip;

and the acquisition unit is used for obtaining the signature subimage in the image to be detected according to the first coordinate and the second coordinate.

On the basis of the foregoing embodiment, optionally, the apparatus further includes: the device comprises a second acquisition module, a first processing module and a second processing module.

Specifically, the second obtaining module is used for obtaining an included angle value between a boundary line in the vertical direction of the image to be detected and a horizontal line;

the first processing module is used for performing first rotation processing on an image to be detected according to the included angle value, and rotating the image to be detected until the included angle value is 90 degrees to obtain a first rotation image;

and the second processing module is used for judging whether the first rotation image comprises an inverted text or not, and if so, performing second rotation processing on the first rotation image to obtain a second rotation image.

On the basis of the foregoing embodiment, the first obtaining module 10 is specifically configured to input the second rotated image into a preset text detection model, so as to obtain a plurality of text strips corresponding to an image to be detected.

On the basis of the foregoing embodiment, optionally, the second processing module includes: a second judging unit and a second processing unit.

Specifically, the second judging unit is configured to input the first rotated image into a preset text direction classification model, judge whether the text in the first rotated image includes an inverted text, and determine the number of the inverted text if the text in the first rotated image includes the inverted text;

and the second processing unit is used for carrying out second rotation processing on the first rotation image according to a preset rotation angle to obtain a second rotation image if the number of the inverted texts is greater than a preset threshold value.

On the basis of the above embodiment, optionally, the image detection module 13 includes a detection unit.

Specifically, the detection unit is configured to determine that the detection result of the image to be detected is passed if both the verification result and the signature detection result are passed.

For specific limitations of the detection device of the vehicle application form, reference may be made to the above limitations of the detection method of the vehicle application form, and details are not repeated here. The modules in the detection device of the vehicle application form can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

obtaining a signature subimage in the image to be detected according to each text strip, and detecting whether the signature subimage contains a user signature to obtain a signature detection result;

and detecting the image to be detected according to the verification result, the text information verification result and the signature detection result to obtain a detection result.

The implementation principle and technical effect of the computer device provided by the above embodiment are similar to those of the above method embodiment, and are not described herein again.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

The implementation principle and technical effect of the computer-readable storage medium provided by the above embodiments are similar to those of the above method embodiments, and are not described herein again.

It will be understood by those of ordinary skill in the art that all or a portion of the processes of the methods of the embodiments described above may be implemented by a computer program that may be stored on a non-volatile computer-readable storage medium, which when executed, may include the processes of the embodiments of the methods described above, wherein any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of detecting a vehicle application form, the method comprising:

2. The method according to claim 1, wherein the determining whether the frame to be selected in the target text strip is selected to obtain the check result includes:

3. The method according to claim 2, wherein the determining whether the frame to be selected is selected according to the target pixel of the frame to be selected in the target text strip to obtain the verification result includes:

4. The method of claim 1, wherein obtaining the signature sub-image in the image to be detected according to each text entry comprises:

5. The method according to claim 1, wherein before the image to be detected is input into a preset text detection model to obtain a plurality of text strips corresponding to the image to be detected, the method further comprises:

6. The method according to claim 5, wherein the determining whether the first rotated image includes an inverted text, and if so, performing a second rotation process on the first rotated image to obtain a second rotated image includes:

7. The method according to claim 1, wherein the detecting the image to be detected according to the verification result and the signature detection result to obtain a detection result comprises:

8. A device for detecting a vehicle application form, the device comprising:

the signature detection module is used for obtaining a signature subimage in the image to be detected according to each text bar, and detecting whether the signature subimage contains a user signature to obtain a signature detection result;

and the image detection module is used for detecting the image to be detected according to the verification result and the signature detection result to obtain a detection result.

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.