Question card information extraction method and device based on image recognition, computer equipment, medium and product

Abstract

The invention relates to a question card information extraction method and device based on image recognition, computer equipment, a medium and a product. The method comprises the steps that identification blocks in a question card image of a question card to be processed are recognized, the identification blocks comprise at least two positioning blocks and at least one verification block, the rotation angle of the question card image is determined according to the relative position relation between the positioning blocks and the verification block, the question card image is corrected according to the rotation angle, and a corrected image is obtained, adjusting the image size of the corrected image according to the coordinate information in the standard question card format to obtain a target image, positioning a target answer area in the target image, identifying a filling area in the target answer area based on the question information in the standard question card format, and obtaining answer content according to the filling area. By adopting the method, the deviation of the question card image can be automatically corrected, so that various information on the question card image can be accurately identified, and the information processing speed and precision are improved.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a method, device, computer equipment, medium and product for extracting question card information based on image recognition. Background Art

[0002] At present, in the field of information processing, the traditional way of processing question card information mainly relies on manual proofreading to complete the processing of question card information. In this process, errors are very likely to occur due to the inevitable human factors. Moreover, the efficiency of manual proofreading is very low, and it is difficult to meet the modern needs of fast and efficient processing of large amounts of question card information. With the development of computer vision technology, automatic processing of question card information through image recognition technology has become a feasible solution, which can overcome many shortcomings of manual processing methods and improve processing efficiency and accuracy.

[0003] However, in actual applications, the question card image will often be deformed or skewed due to some external factors. For example, during the printing process, the question card image may be deformed due to the printer settings, paper characteristics, etc.; in the scanning process, the accuracy and operation mode of the scanning equipment may also cause the question card image to be skewed, thereby affecting the accuracy of question card information extraction. Summary of the invention

[0004] Based on this, it is necessary to provide a question card information extraction method, device, computer equipment, medium and product based on image recognition that can accurately identify question card information in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for extracting question card information based on image recognition, comprising:

[0006] Identify identification blocks in the question card image of the question card to be processed, where the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify page information of the question card;

[0007] According to the relative position relationship between the positioning block and the verification block, the rotation angle of the question card image is determined, and the question card image is corrected according to the rotation angle to obtain a corrected image;

[0008] Adjust the image size of the correction image according to the coordinate information in the standard question card format to obtain the target image, and locate the target answer area in the target image;

[0009] Based on the question information in the standard question card format, the filling area in the target answer area is identified, and the answer content is obtained according to the filling area.

[0010] In one embodiment, the process of determining the standard question card format includes:

[0011] The coordinate system of the question card is established by taking the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and y-axis; the standard question card is a rectangular question card of a preset size;

[0012] Obtaining standard coordinate information of standard identification blocks in the standard question card in the question card coordinate system, wherein each standard identification block is distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block;

[0013] According to the standard coordinate information, the relative coordinate information of the standard answer area in the question card coordinate system is obtained, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block;

[0014] The standard coordinate information and the relative coordinate information are used as a standard question card format, and question information is added to the standard question card format, where the question information includes the question number, the number of questions, and question information in the standard answer area;

[0015] Among them, at least two standard positioning blocks are at the same horizontal position, and at least one standard positioning block and a corresponding standard checking block are at the same vertical position.

[0016] In one embodiment, the steps of determining the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain the corrected image include:

[0017] Determine the two target positioning blocks corresponding to the two standard positioning blocks at the same horizontal position in the standard question card in the question card image, and the target positioning block and target verification block corresponding to the standard positioning block and standard verification block at the same vertical position in the standard question card in the question card image;

[0018] Obtaining a first angle between a line connecting two target positioning blocks and a horizontal direction, and a second angle between a line connecting the target positioning block and a target verification block and a vertical direction;

[0019] When the first angle is used as the rotation angle of the question card image, adjusting the horizontal rotation angle of the question card image to eliminate the first angle;

[0020] When the second angle is used as the rotation angle of the question card image, the vertical rotation angle of the question card image is adjusted to eliminate the second angle.

[0021] In one embodiment, adjusting the image size of the correction image according to the coordinate information in the standard question card format to obtain the target image includes:

[0022] When the first angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks;

[0023] When the second angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

[0024] In one embodiment, the method further comprises:

[0025] When the target answer area is incompletely filled or completely covered by an unrecognized area, the filled area in the unrecognized area is located according to the filled position in the adjacent answer area around the unrecognized area.

[0026] In one embodiment, the size of the question card to be processed is determined;

[0027] Convert the standard coordinate information and relative coordinate information into target coordinate information matching the size of the question card;

[0028] The question card image of the question card to be processed is corrected according to the target coordinate information, and the answer area in the corrected question card image is located to obtain the answer content.

[0029] In a second aspect, the present application also provides a question card information extraction device based on image recognition, comprising:

[0030] An identification block recognition module is used to recognize identification blocks in the question card image of the question card to be processed, wherein the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify the page information of the question card;

[0031] The image correction module is used to determine the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correct the question card image according to the rotation angle to obtain a corrected image;

[0032] A size adjustment module is used to adjust the image size of the correction image according to the coordinate information in the standard question card format, obtain the target image, and locate the target answer area in the target image;

[0033] The information extraction module is used to identify the filling area in the target answer area based on the question information in the standard question card format, and obtain the answer content according to the filling area.

[0034] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method steps of any one of the first aspects are implemented.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements any one of the method steps in the first aspect when the computer program is executed by a processor.

[0036] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements any one of the method steps in the first aspect when executed by a processor.

[0037] The above-mentioned question card information extraction method, device, computer equipment, medium and product based on image recognition can automatically realize the correction and adjustment of the question card image by identifying the positioning block and the check block in the question card image of the question card to be processed, so that the question card format in the processed question card image matches the standard question card format, thereby accurately locating the answer area in the question card image, and then accurately identifying the question card information, reducing errors caused by manual operation. In addition, by correcting and adjusting the question card image, it can adapt to answer cards of different formats and types, and improve the speed and accuracy of information processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0039] FIG1 is a diagram showing an application environment of a method for extracting question card information based on image recognition in one embodiment;

[0040] FIG2 is a schematic diagram of a flow chart of a method for extracting question card information based on image recognition in one embodiment;

[0041] FIG3 is a schematic diagram of a standard question card in one embodiment;

[0042] FIG4 is a schematic diagram of a question card image in one embodiment;

[0043] FIG5 is a flow chart of a method for extracting question card information based on image recognition in another embodiment;

[0044] FIG6 is a structural block diagram of a device for extracting question card information based on image recognition in one embodiment;

[0045] FIG. 7 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] The method for extracting question card information based on image recognition provided in the embodiment of the present application can be applied in the application environment shown in FIG. 1 . The terminal 102 communicates with the acquisition device 104 through a network. The terminal 102 is used to obtain the question card image of the question card to be processed through the acquisition device 104, identify the identification block in the question card image of the question card to be processed, the identification block includes at least two positioning blocks and at least one check block, determine the rotation angle of the question card image according to the relative position relationship between the positioning block and the check block, and correct the question card image according to the rotation angle to obtain the corrected image, adjust the image size of the corrected image according to the coordinate information in the standard question card format, obtain the target image, locate the target answer area in the target image, identify the fill area in the target answer area based on the question information in the standard question card format, and obtain the answer content according to the fill area. The terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device may be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The acquisition device 104 may be a scanner, an imaging device, etc.

[0048] In an exemplary embodiment, as shown in FIG2 , a method for extracting question card information based on image recognition is provided, and the method is applied to the extraction of question card information based on image recognition in FIG1 as an example for description, including the following steps 202 to 208. Among them:

[0049] S202: Identify identification blocks in the question card image of the question card to be processed, where the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify page information of the question card.

[0050] Optionally, the question card image is a digital image of the question card to be processed obtained by a scanner or other equipment. According to the predefined standard question card format, there is a preset identification block on the surface of the question card. Through the image processing method, combined with the coordinate information of the identification block in the standard question card format, the corresponding area in the question card image is identified and the identification block in the question card image is located. Among them, the identification block includes at least two positioning blocks and at least one verification block. The positioning block is used to locate the position of the question card to identify the information of the specified area in the question card, and the verification block is used to verify the page information of the question card. For example, the question card image is an image of a page in the question card to be processed. It can be understood that the standard question card format is a pre-defined standard format, which is stored locally in the terminal in the form of a file. The question card to be processed is a real question card generated according to the standard question card format. In the process of printing the question card to be processed or scanning the question card image of the question card to be processed, it may be affected by other factors and may cause the final presented question card image to be deformed or skewed. Therefore, by comparing with the pre-defined standard question card format, the question card image can be corrected and adapted so that the format of the question card image is consistent with the standard question card format, thereby accurately extracting the question card information in the question card image.

[0051] S204: Determine the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correct the question card image according to the rotation angle to obtain a corrected image.

[0052] Optionally, in the standard question card format, since the coordinates of the positioning block and the verification block are known, their relative position relationship can be determined based on the coordinates. Therefore, for the question card image of the question card to be processed, based on the relative position relationship between the positioning block and the verification block in the question card image, the change difference in the format of the question card image relative to the standard question card format can be determined, such as the rotation angle of the question card image. Thereafter, the question card image is corrected according to the rotation angle so that the question card format in the obtained corrected image matches the standard question card format.

[0053] S206: Adjust the image size of the correction image according to the coordinate information in the standard question card format to obtain the target image, and locate the target answer area in the target image.

[0054] Optionally, since the question card image is collected by a scanner or other device, it may not be consistent with the question card size defined in the standard question card format. In this case, the image size of the corrected image can be adjusted according to the coordinate information in the standard question card format to make the size of the target image consistent with the question card size defined in the standard question card format, so that the corresponding target answer area in the target image can be quickly located according to the answer area defined in the standard question card format.

[0055] S208: Based on the question information in the standard question card format, identify the fill-in area in the target answer area, and obtain the answer content according to the fill-in area.

[0056] Optionally, the answer area defined in the standard question card format can be divided into a subjective question answer area and an objective question answer area, wherein, when extracting the answer information of the objective question, it is achieved by identifying the filled-in area in the corresponding answer area. Specifically, according to the information such as the question number and the number of questions given by the question information in the standard question card format, the specified position in the target answer area can be quickly located, and then the filled-in area can be identified to obtain the answer content corresponding to the corresponding question. It can be understood that for the extraction of the answer content of the subjective question, the answer range of the objective question can be located according to the question information, and then the complete answer content within the answer range can be directly extracted.

[0057] In the above-mentioned question card information extraction method based on image recognition, by identifying the positioning block and the check block in the question card image of the question card to be processed, the question card image can be automatically corrected and adjusted so that the question card format in the processed question card image matches the standard question card format, thereby accurately locating the answer area in the question card image, and then accurately identifying the question card information, reducing errors caused by manual operation. In addition, by correcting and adjusting the question card image, it is possible to adapt to answer cards of different formats and types, and improve the speed and accuracy of information processing.

[0058] In an exemplary embodiment, the process of determining the standard question card format includes: establishing a question card coordinate system with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and the y-axis; the standard question card is a rectangular question card of a preset size; obtaining standard coordinate information of a standard identification block in the standard question card in the question card coordinate system, wherein each standard identification block is distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; obtaining relative coordinate information of a standard answer area in the question card coordinate system according to the standard coordinate information, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block; using the standard coordinate information and the relative coordinate information as the standard question card format, and adding question information to the standard question card format, and the question information includes the question number, the number of questions and the question information in the standard answer area; wherein at least two standard positioning blocks are in the same horizontal position, and at least one standard positioning block is in the same vertical position as a corresponding standard verification block.

[0059] Optionally, during the question card design stage, the coordinates of each identification block and the position of the answer area are predefined, and corresponding question information is added to form a standard question card format. For ease of understanding, the front of the standard question card shown in FIG3 is used as an example for explanation. The standard question card shown in FIG3 includes a standard positioning block a distributed in the upper left corner of the question card, a standard positioning block b distributed in the upper right corner of the question card, and a standard check block c distributed in the lower left corner of the question card.

[0060] Exemplarily, a question card coordinate system is established with the upper left vertex O of the standard question card as the coordinate origin, and the two sides constituting the upper left vertex as the x-axis and y-axis. According to the established question card coordinate system, the corresponding standard coordinate information of the standard positioning block a, the standard positioning block b and the standard verification block c can be accurately located, wherein the standard coordinate information is the position coordinate of the upper left vertex of the corresponding identification block.

[0061] Optionally, in order to facilitate the determination of the horizontal direction of the standard question card, the standard positioning block a and the standard positioning block b can be arranged at the same horizontal position. In order to facilitate the determination of the vertical direction of the standard question card, the standard positioning block a and the standard verification block c can be arranged at the same vertical position. Correspondingly, on the back of the standard question card, the standard positioning block b and the standard verification block c can be arranged at the same vertical position. The front and back of the standard question card can be distinguished by the position of the standard verification block c.

[0062] Furthermore, based on the standard coordinate information, the relative coordinate information of the standard answer area in the question card coordinate system can be obtained, where the relative coordinate information includes the relative position between the standard answer area and each standard identification block, as shown in the standard answer area X1 in the figure. Based on the relative position between each vertex of the X1 area and each identification block, the unique position of the standard answer area X1 in the standard question card can be accurately located.

[0063] Furthermore, after defining the standard question card format, it is necessary to add the corresponding question information, including the question number, question quantity and specific question information in the standard answer area. By adding the question information, the answer content of different questions can be identified and processed. The final complete standard question card format is saved in the form of a file as a standard for subsequent question card processing.

[0064] In this embodiment, according to the predefined standard question card format, the deviation of the question card image of the question card to be processed can be automatically corrected, thereby ensuring the accurate extraction of the answer content and improving the speed and accuracy of information processing.

[0065] In an exemplary embodiment, the rotation angle of the question card image is determined according to the relative position relationship between the positioning block and the verification block, and the question card image is corrected according to the rotation angle to obtain a corrected image, including: determining two target positioning blocks corresponding to two standard positioning blocks at the same horizontal position in the standard question card, and a target positioning block and a target verification block corresponding to the standard positioning block and the standard verification block at the same vertical position in the standard question card; obtaining a first angle between a line connecting the two target positioning blocks and the horizontal direction, and a second angle between a line connecting the target positioning block and the target verification block and the vertical direction; when the first angle is used as the rotation angle of the question card image, adjusting the horizontal rotation angle of the question card image to eliminate the first angle; when the second angle is used as the rotation angle of the question card image, adjusting the vertical rotation angle of the question card image to eliminate the second angle.

[0066] Optionally, since a standard question card may have multiple standard positioning blocks and multiple standard verification blocks, when performing image correction, identification blocks that can determine the horizontal and vertical directions of the question card are usually selected as reference targets, that is, two standard positioning blocks at the same horizontal position in the standard question card, and standard positioning blocks and standard verification blocks at the same vertical position. For ease of understanding, the question card image of the question card to be processed shown in FIG4 is used as an example for explanation. For the standard question card shown in FIG3, the two standard positioning blocks a and b at the same horizontal position in the standard question card correspond to two target positioning blocks A and B in the question card image, and the standard positioning block a and standard verification block c at the same vertical position in the standard question card correspond to target positioning blocks and target verification blocks A and C in the question card image.

[0067] Furthermore, the first angle α between the line between the target positioning block A and the target positioning block B and the horizontal direction is, and the second angle β between the line between the target positioning block A and the target verification block C and the vertical direction is. Since the first angle α may not be equal to the second angle β, when correcting the question card image, it is impossible to ensure that the horizontal and vertical directions are restored to the standard position at the same time. Therefore, in actual processing, the angle in one of the directions can be selected as the correction target, and the rotation angle of the question card image in the corresponding direction can be adjusted to eliminate the angle so that the question card image maintains a horizontal or vertical position.

[0068] In this embodiment, by accurately determining the positional relationship between the positioning block and the verification block to calculate the rotation angle, the question card image can be accurately corrected, so that the subsequent recognition, analysis and processing of the question card content are more accurate and reliable, and the erroneous judgment caused by image tilt or offset is reduced.

[0069] In an exemplary embodiment, the image size of the correction image is adjusted according to the coordinate information in the standard question card format to obtain the target image, including: when the first angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; when the second angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

[0070] Optionally, after the question card image is corrected, the image size of the corrected image needs to be adjusted so that the processed image completely matches the standard question card format. Specifically, in combination with the contents of Figures 3 and 4, when the first angle α is used as the rotation angle of the question card image, by adjusting the image size of the corrected image, the relative distance between the target positioning block A and the target positioning block B can be made the same as the relative distance between the corresponding two standard positioning blocks (i.e., d1 in Figure 3), and when the second angle β is used as the rotation angle of the question card image, by adjusting the image size of the corrected image, the relative distance between the target positioning block A and the target verification block C can be made the same as the relative distance between the corresponding standard positioning block and the standard verification block (i.e., d2 in Figure 3).

[0071] In this embodiment, by adjusting the image size of the corrected image, it is possible to ensure that the processed image is completely consistent with the standard question card format in size, so that the target answer area in the target image can be accurately located based on the relative position relationship between the answer area and the identification block, thereby ensuring accurate identification of the question card information.

[0072] In an exemplary embodiment, the method further includes: when there is incomplete filling in the target answer area or it is completely covered by an unrecognized area, locating the filled area in the unrecognized area according to the filling position in the adjacent answer area around the unrecognized area.

[0073] Optionally, when identifying the target answer area in the target image, if there is an area that is incompletely filled or completely covered and cannot be directly identified, the adjacent answer areas around the unidentified area will be analyzed first. These adjacent answer areas have a certain spatial correlation with the unidentified area. Then, the filled area in the unidentified area is inferred based on the filled position in the adjacent answer area.

[0074] In this embodiment, by locating the filling area as accurately as possible based on the filling position in the adjacent answering area around the unrecognized area, the recognition errors caused by local problems can be reduced, and the recognition accuracy of the overall answering area can be improved.

[0075] In an exemplary embodiment, the method also includes: determining the question card size of the question card to be processed; converting the standard coordinate information and the relative coordinate information into target coordinate information that matches the question card size; correcting the question card image of the question card to be processed according to the target coordinate information, and locating the answer area in the corrected question card image to obtain the answer content.

[0076] Optionally, in actual applications, if the size of the question card to be processed is inconsistent with the size in the standard question card format, the coordinate information in the standard question card format can be converted first to obtain a standard question card format that is compatible with the question card to be processed, thereby adapting to question cards of different sizes. Specifically, the standard coordinate information and relative coordinate information in the standard question card format are converted into target coordinate information that matches the question card size. For example, according to the size ratio between the question card to be processed and the standard question card, the coordinate value is scaled accordingly so that the coordinate information can be applied to the current question card. After that, the question card image of the question card to be processed is corrected according to the converted target coordinate information, so that the answer area in the corrected question card image can be directly located to accurately obtain the answer content.

[0077] In this embodiment, by processing the standard question card format, it can adapt to question cards of different sizes, thereby ensuring that the positioning of the answer area is more accurate.

[0078] In an exemplary embodiment, as shown in FIG5 , a method for extracting question card information based on image recognition is provided, and the method comprises the following steps:

[0079] S502: Establish a question card coordinate system with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and y-axis; the standard question card is a rectangular question card of a preset size; obtain standard coordinate information of the standard identification blocks in the standard question card in the question card coordinate system, wherein each standard identification block is respectively distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; obtain relative coordinate information of the standard answer area in the question card coordinate system according to the standard coordinate information, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block; use the standard coordinate information and the relative coordinate information as the standard question card format, and add question information to the standard question card format, and the question information includes the question number, the number of questions and the question information in the standard answer area.

[0080] Among them, at least two standard positioning blocks are at the same horizontal position, and at least one standard positioning block and a corresponding standard checking block are at the same vertical position.

[0081] S504: Identify identification blocks in the question card image of the question card to be processed, the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify the page information of the question card; determine two target positioning blocks corresponding to two standard positioning blocks at the same horizontal position in the standard question card in the question card image, and a target positioning block and a target verification block corresponding to the standard positioning block and the standard verification block at the same vertical position in the standard question card in the question card image; obtain a first angle between a line connecting the two target positioning blocks and the horizontal direction, and a second angle between a line connecting the target positioning block and the target verification block and the vertical direction; when the first angle is used as the rotation angle of the question card image, adjust the horizontal rotation angle of the question card image to eliminate the first angle; when the second angle is used as the rotation angle of the question card image, adjust the vertical rotation angle of the question card image to eliminate the second angle.

[0082] S506: When the first angle is used as the rotation angle of the question card image, the image size of the corrected image is adjusted so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; when the second angle is used as the rotation angle of the question card image, the image size of the corrected image is adjusted so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block, and the target answer area in the target image is located; based on the question information in the standard question card format, the filling area in the target answer area is identified, and the answer content is obtained according to the filling area.

[0083] S508: When the target answer area is incompletely filled or completely covered by an unrecognized area, locate the filled area in the unrecognized area according to the filled position in the adjacent answer area around the unrecognized area.

[0084] S510: Determine the question card size of the question card to be processed; convert the standard coordinate information and the relative coordinate information into target coordinate information that matches the question card size; correct the question card image of the question card to be processed according to the target coordinate information, and locate the answer area in the corrected question card image to obtain the answer content.

[0085] In this embodiment, by identifying the positioning block and the check block in the question card image of the question card to be processed, the question card image can be automatically corrected and adjusted so that the question card format in the processed question card image matches the standard question card format, thereby accurately locating the answer area in the question card image, and then accurately identifying the question card information, reducing errors caused by manual operation. In addition, by correcting and adjusting the question card image, it is possible to adapt to answer cards of different formats and types, and improve the speed and accuracy of information processing.

[0086] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0087] Based on the same inventive concept, the embodiment of the present application also provides a question card information extraction device based on image recognition for implementing the question card information extraction method based on image recognition involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the question card information extraction device based on image recognition provided below can refer to the limitations of the question card information extraction method based on image recognition above, and will not be repeated here.

[0088] In an exemplary embodiment, as shown in FIG6 , a device for extracting question card information based on image recognition is provided, comprising: a marker block recognition module 10, an image correction module 20, a size adjustment module 30 and an information extraction module 40, wherein:

[0089] The identification block recognition module 10 is used to recognize identification blocks in the question card image of the question card to be processed, and the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify the page information of the question card.

[0090] The image correction module 20 is used to determine the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correct the question card image according to the rotation angle to obtain a corrected image.

[0091] The size adjustment module 30 is used to adjust the image size of the correction image according to the coordinate information in the standard question card format, obtain the target image, and locate the target answer area in the target image.

[0092] The information extraction module 40 is used to identify the filling area in the target answer area based on the question information in the standard question card format, and obtain the answer content according to the filling area.

[0093] In an exemplary embodiment, the process of determining the standard question card format involved in the size adjustment module 30 includes: establishing a question card coordinate system with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and the y-axis; the standard question card is a rectangular question card of a preset size; obtaining standard coordinate information of the standard identification blocks in the standard question card in the question card coordinate system, wherein each standard identification block is distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; obtaining relative coordinate information of the standard answer area in the question card coordinate system according to the standard coordinate information, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block; using the standard coordinate information and the relative coordinate information as the standard question card format, and adding question information to the standard question card format, and the question information includes the question number, the number of questions and the question information in the standard answer area; wherein at least two standard positioning blocks are in the same horizontal position, and at least one standard positioning block is in the same vertical position as a corresponding standard verification block.

[0094] In an exemplary embodiment, the image correction module 20 is also used to determine the two target positioning blocks corresponding to the two standard positioning blocks at the same horizontal position in the standard question card, and the target positioning block and target verification block corresponding to the standard positioning block and standard verification block at the same vertical position in the standard question card; obtain a first angle between the connecting line between the two target positioning blocks and the horizontal direction, and a second angle between the connecting line between the target positioning block and the target verification block and the vertical direction; when the first angle is used as the rotation angle of the question card image, adjust the horizontal rotation angle of the question card image to eliminate the first angle; when the second angle is used as the rotation angle of the question card image, adjust the vertical rotation angle of the question card image to eliminate the second angle.

[0095] In an exemplary embodiment, the size adjustment module 30 is also used to adjust the image size of the correction image when the first angle is used as the rotation angle of the question card image, so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; when the second angle is used as the rotation angle of the question card image, adjust the image size of the correction image when the second angle is used as the rotation angle of the question card image, so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

[0096] In an exemplary embodiment, the information extraction module 40 is also used to locate the filled area in the unrecognized area based on the filled position in the adjacent answer area around the unrecognized area when the target answer area is incompletely filled or completely covered by the unrecognized area.

[0097] In an exemplary embodiment, the information extraction module 40 is also used to determine the question card size of the question card to be processed; convert the standard coordinate information and the relative coordinate information into target coordinate information that matches the question card size; correct the question card image of the question card to be processed according to the target coordinate information, and locate the answer area in the corrected question card image to obtain the answer content.

[0098] Each module in the above-mentioned image recognition-based question card information extraction device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0099] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG7. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for extracting question card information based on image recognition is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0100] Those skilled in the art will understand that the structure shown in FIG. 7 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0101] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: identifying an identification block in a question card image of a question card to be processed, the identification block including at least two positioning blocks and at least one verification block, the verification block being used to verify the page information of the question card; determining a rotation angle of the question card image according to a relative position relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain a corrected image; adjusting the image size of the corrected image according to coordinate information in a standard question card format to obtain a target image, and locating a target answer area in the target image; identifying a fill-in area within the target answer area based on question information in the standard question card format, and obtaining answer content based on the fill-in area.

[0102] In one embodiment, a process for determining a standard question card format involved in a processor executing a computer program includes: establishing a question card coordinate system with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and the y-axis; the standard question card is a rectangular question card of a preset size; obtaining standard coordinate information of a standard identification block in the standard question card in the question card coordinate system, wherein each standard identification block is respectively distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; obtaining relative coordinate information of a standard answer area in the question card coordinate system according to the standard coordinate information, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block; using the standard coordinate information and the relative coordinate information as the standard question card format, and adding question information to the standard question card format, and the question information includes the question number, the number of questions and the question information in the standard answer area; wherein at least two standard positioning blocks are in the same horizontal position, and at least one standard positioning block is in the same vertical position as a corresponding standard verification block.

[0103] In one embodiment, when a processor executes a computer program, the steps of determining the rotation angle of a question card image according to the relative position relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain a corrected image include: determining two target positioning blocks corresponding to two standard positioning blocks at the same horizontal position in a standard question card, and a target positioning block and a target verification block corresponding to the standard positioning block and the standard verification block at the same vertical position in the standard question card; obtaining a first angle between a line connecting the two target positioning blocks and a horizontal direction, and a second angle between a line connecting the target positioning block and the target verification block and a vertical direction; when the first angle is used as the rotation angle of the question card image, adjusting the horizontal rotation angle of the question card image to eliminate the first angle; when the second angle is used as the rotation angle of the question card image, adjusting the vertical rotation angle of the question card image to eliminate the second angle.

[0104] In one embodiment, when the processor executes the computer program, the image size of the correction image is adjusted according to the coordinate information in the standard question card format to obtain the target image, including: when the first angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; when the second angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

[0105] In one embodiment, when the processor executes the computer program, the following steps are also implemented: when there is incomplete filling in the target answer area, or when there is an unrecognized area that is completely covered, the filled area in the unrecognized area is located according to the filling position in the adjacent answer area around the unrecognized area.

[0106] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining the question card size of the question card to be processed; converting the standard coordinate information and the relative coordinate information into target coordinate information that matches the question card size; correcting the question card image of the question card to be processed according to the target coordinate information, and locating the answer area in the corrected question card image to obtain the answer content.

[0107] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: identifying an identification block in a question card image of a question card to be processed, the identification block including at least two positioning blocks and at least one verification block, the verification block being used to verify the page information of the question card; determining a rotation angle of the question card image according to a relative positional relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain a corrected image; adjusting the image size of the corrected image according to coordinate information in a standard question card format to obtain a target image, and locating a target answer area in the target image; identifying a fill-in area within the target answer area based on question information in the standard question card format, and obtaining answer content based on the fill-in area.

[0108] In one embodiment, the process of determining the standard question card format involved in the execution of the computer program by the processor includes: establishing a question card coordinate system with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and the y-axis; the standard question card is a rectangular question card of a preset size; obtaining standard coordinate information of a standard identification block in the standard question card in the question card coordinate system, wherein each standard identification block is respectively distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; obtaining relative coordinate information of a standard answer area in the question card coordinate system according to the standard coordinate information, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block; using the standard coordinate information and the relative coordinate information as the standard question card format, and adding question information to the standard question card format, and the question information includes the question number, the number of questions and the question information in the standard answer area; wherein at least two standard positioning blocks are in the same horizontal position, and at least one standard positioning block is in the same vertical position as a corresponding standard verification block.

[0109] In one embodiment, when the computer program is executed by a processor, the steps of determining the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain a corrected image include: determining two target positioning blocks corresponding to two standard positioning blocks at the same horizontal position in the standard question card, and a target positioning block and a target verification block corresponding to the standard positioning block and the standard verification block at the same vertical position in the standard question card; obtaining a first angle between a line connecting the two target positioning blocks and the horizontal direction, and a second angle between a line connecting the target positioning block and the target verification block and the vertical direction; when the first angle is used as the rotation angle of the question card image, adjusting the horizontal rotation angle of the question card image to eliminate the first angle; when the second angle is used as the rotation angle of the question card image, adjusting the vertical rotation angle of the question card image to eliminate the second angle.

[0110] In one embodiment, when the computer program is executed by a processor, the steps involved in adjusting the image size of the correction image according to the coordinate information in the standard question card format to obtain the target image include: when a first angle is used as the rotation angle of the question card image, adjusting the image size of the correction image so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; when a second angle is used as the rotation angle of the question card image, adjusting the image size of the correction image so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

[0111] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when there is incomplete filling in the target answer area, or when there is an unrecognized area that is completely covered, the filled area in the unrecognized area is located according to the filling position in the adjacent answer area around the unrecognized area.

[0112] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the question card size of the question card to be processed; converting the standard coordinate information and the relative coordinate information into target coordinate information that matches the question card size; correcting the question card image of the question card to be processed according to the target coordinate information, and locating the answer area in the corrected question card image to obtain the answer content.

[0113] In one embodiment, a computer program product is provided, including a computer program, which implements the following steps when executed by a processor: identifying identification blocks in a question card image of a question card to be processed, the identification blocks including at least two positioning blocks and at least one verification block, the verification block being used to verify the page information of the question card; determining the rotation angle of the question card image according to the relative position relationship between the positioning blocks and the verification blocks, and correcting the question card image according to the rotation angle to obtain a corrected image; adjusting the image size of the corrected image according to the coordinate information in the standard question card format to obtain a target image, and locating a target answer area in the target image; identifying a fill-in area within the target answer area based on the question information in the standard question card format, and obtaining the answer content based on the fill-in area.

[0114] In one embodiment, the process of determining the standard question card format involved in the execution of the computer program by the processor includes: establishing a question card coordinate system with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and the y-axis; the standard question card is a rectangular question card of a preset size; obtaining standard coordinate information of a standard identification block in the standard question card in the question card coordinate system, wherein each standard identification block is respectively distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; obtaining relative coordinate information of a standard answer area in the question card coordinate system according to the standard coordinate information, and the relative coordinate information includes the relative position between the standard answer area and each standard identification block; using the standard coordinate information and the relative coordinate information as the standard question card format, and adding question information to the standard question card format, and the question information includes the question number, the number of questions and the question information in the standard answer area; wherein at least two standard positioning blocks are in the same horizontal position, and at least one standard positioning block is in the same vertical position as a corresponding standard verification block.

[0115] In one embodiment, when the computer program is executed by a processor, the steps of determining the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain a corrected image include: determining two target positioning blocks corresponding to two standard positioning blocks at the same horizontal position in the standard question card, and a target positioning block and a target verification block corresponding to the standard positioning block and the standard verification block at the same vertical position in the standard question card; obtaining a first angle between a line connecting the two target positioning blocks and the horizontal direction, and a second angle between a line connecting the target positioning block and the target verification block and the vertical direction; when the first angle is used as the rotation angle of the question card image, adjusting the horizontal rotation angle of the question card image to eliminate the first angle; when the second angle is used as the rotation angle of the question card image, adjusting the vertical rotation angle of the question card image to eliminate the second angle.

[0116] In one embodiment, when the computer program is executed by a processor, the steps involved in adjusting the image size of the correction image according to the coordinate information in the standard question card format to obtain the target image include: when a first angle is used as the rotation angle of the question card image, adjusting the image size of the correction image so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; when a second angle is used as the rotation angle of the question card image, adjusting the image size of the correction image so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

[0117] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: when there is incomplete filling in the target answer area, or when there is an unrecognized area that is completely covered, the filled area in the unrecognized area is located according to the filling position in the adjacent answer area around the unrecognized area.

[0118] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the question card size of the question card to be processed; converting the standard coordinate information and the relative coordinate information into target coordinate information that matches the question card size; correcting the question card image of the question card to be processed according to the target coordinate information, and locating the answer area in the corrected question card image to obtain the answer content.

[0119] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0120] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0121] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for extracting question card information based on image recognition, characterized in that: The method comprises: Identify identification blocks in the question card image of the question card to be processed, wherein the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify the page information of the question card; Determine the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correct the question card image according to the rotation angle to obtain a corrected image; Adjusting the image size of the correction image according to the coordinate information in the standard question card format to obtain a target image, and locating a target answer area in the target image; Based on the question information in the standard question card format, the filling area in the target answer area is identified, and the answer content is obtained according to the filling area.

2. The method according to claim 1, characterized in that The process of determining the standard question card format includes: A question card coordinate system is established with the upper left vertex of the standard question card as the coordinate origin and the two sides constituting the upper left vertex as the x-axis and the y-axis; the standard question card is a rectangular question card of a preset size; Obtaining standard coordinate information of a standard identification block in the standard question card in the question card coordinate system, wherein each of the standard identification blocks is respectively distributed in different areas of the edge of the standard question card, including at least two standard positioning blocks and at least one standard verification block, and the standard coordinate information includes the coordinates of the upper left vertex of the standard identification block; According to the standard coordinate information, relative coordinate information of the standard answer area in the question card coordinate system is obtained, wherein the relative coordinate information includes the relative position between the standard answer area and each standard identification block; The standard coordinate information and the relative coordinate information are used as a standard question card format, and question information is added to the standard question card format, wherein the question information includes the question number, the number of questions and the question information in the standard answer area; Among them, at least two standard positioning blocks are at the same horizontal position, and at least one standard positioning block and a corresponding standard checking block are at the same vertical position.

3. The method according to claim 2, characterized in that The step of determining the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correcting the question card image according to the rotation angle to obtain a corrected image includes: Determine two target positioning blocks corresponding to two standard positioning blocks at the same horizontal position in the standard question card in the question card image, and a target positioning block and a target verification block corresponding to a standard positioning block and a standard verification block at the same vertical position in the standard question card in the question card image; Acquire a first angle between a line connecting the two target positioning blocks and a horizontal direction, and a second angle between a line connecting the target positioning block and the target verification block and a vertical direction; When the first angle is used as the rotation angle of the question card image, adjusting the horizontal rotation angle of the question card image to eliminate the first angle; When the second angle is used as the rotation angle of the question card image, the vertical rotation angle of the question card image is adjusted to eliminate the second angle.

4. The method according to claim 3, characterized in that The step of adjusting the image size of the correction image according to the coordinate information in the standard question card format to obtain the target image includes: In the case where the first angle is used as the rotation angle of the question card image, the image size of the correction image is adjusted so that the relative distance between the two target positioning blocks is the same as the relative distance between the corresponding two standard positioning blocks; When the second angle is used as the rotation angle of the question card image, the image size of the corrected image is adjusted so that the relative distance between the target positioning block and the target verification block is the same as the relative distance between the corresponding standard positioning block and the standard verification block.

5. The method according to claim 1, characterized in that The method further comprises: In the case where the target answering area is incompletely filled or completely covered by an unrecognized area, the filled area in the unrecognized area is located according to the filled position in the adjacent answering area around the unrecognized area.

6. The method according to claim 1, characterized in that The method further comprises: Determining the size of the question card to be processed; Converting the standard coordinate information and the relative coordinate information into target coordinate information matching the size of the question card; The question card image of the question card to be processed is corrected according to the target coordinate information, and the answer area in the corrected question card image is located to obtain the answer content.

7. A device for extracting question card information based on image recognition, characterized in that: The device comprises: An identification block recognition module, used to recognize identification blocks in the question card image of the question card to be processed, wherein the identification blocks include at least two positioning blocks and at least one verification block, and the verification block is used to verify the page information of the question card; An image correction module, used to determine the rotation angle of the question card image according to the relative position relationship between the positioning block and the verification block, and correct the question card image according to the rotation angle to obtain a corrected image; A size adjustment module, used to adjust the image size of the correction image according to the coordinate information in the standard question card format, obtain the target image, and locate the target answer area in the target image; The information extraction module is used to identify the filling area in the target answer area based on the question information in the standard question card format, and obtain the answer content according to the filling area.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images