Fill-in card image recognition method and device based on point cloud algorithm

Abstract

The present invention relates to a method and device for image recognition of a fill-in card based on a point cloud algorithm, and also discloses an electronic device and a non-transitory computer-readable storage medium, belonging to the field of image recognition technology. The method comprises: evenly tiling two-dimensional point cloud data on a binary image of a fill-in card; gridding the binary image with the tiled two-dimensional point cloud data; adjusting the weight of each point in the two-dimensional point cloud data of each area of the divided grid; obtaining the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binary image in each area of the divided grid, and comparing the obtained total weight value with a set total weight value threshold to determine the selected option corresponding to each question on the fill-in card. The method provided in the present application can accurately identify fill-in cards, overcoming the problem of low recognition accuracy for wrinkled or irregularly filled fill-in cards in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a method and device for image recognition of a fill-in card based on a point cloud algorithm. Background Art

[0002] Currently, in some standardized answer sheets, users fill in their chosen answers on the sheet, which is then scanned by an optical reader using the principle of light reflection. The optical reader primarily utilizes carbon's electrical sensitivity to infrared light to identify the filled-in areas on the sheet. This approach has greatly facilitated the recognition of answer sheets. In recent years, with the advancement of computer technology, some answer sheet recognition tasks have shifted from optical readers to computer image recognition algorithms, typically using algorithms from libraries such as OpenCV. A common approach involves first preprocessing the image to generate a binary image. OpenCV's contour search function is then used to locate a specific answer sheet box within the binary image. Once the box is located, the color of the box is determined to complete the answer sheet recognition. However, this method places high demands on the quality of the answer sheet and the preservation of the answer sheet, and has low accuracy for wrinkled or irregularly filled answer sheets. Summary of the Invention

[0003] The present invention intends to provide a method, device, electronic device and non-transitory computer-readable storage medium for filling-in card image recognition based on point cloud algorithm to solve the deficiencies in the prior art. The technical problems to be solved by the present invention are achieved through the following technical solutions.

[0004] The present invention provides a method for image recognition of a filled-in card based on a point cloud algorithm, the method comprising:

[0005] Obtain the two-dimensional point cloud data of the filled card to be identified;

[0006] Obtaining a filled-in card image to be identified, and performing binarization processing on the filled-in card image to be identified to obtain a binarized image of the filled-in card;

[0007] Evenly tiling the two-dimensional point cloud data on the binary image of the fill-in card;

[0008] According to the number of questions on the fill-in card and the number of options corresponding to each question, the binary image of the tiled two-dimensional point cloud data is grid-divided to obtain the corresponding grid;

[0009] Adjusting the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binary image in each area of the divided grid;

[0010] According to the weight of each point in the obtained two-dimensional point cloud data, the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binary image in each area of the divided grid is obtained;

[0011] The obtained total weight value is compared with the set total weight value threshold to determine the selected option corresponding to each question on the fill-in card.

[0012] In the above solution, the divided grid is a grid of M rows and N columns, where M is the number of questions on the fill-in-the-blank card, and N is the number of options corresponding to each question.

[0013] In the above solution, adjusting the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binary image in each area of the divided grid includes:

[0014] Increase the weight of each point in the two-dimensional point cloud data in the white pixel area in each area of the divided grid by a specific value;

[0015] The weight of each point in the two-dimensional point cloud data in the black pixel area in each area of the divided grid is reduced by a specific value.

[0016] The filled-in card image recognition device based on point cloud algorithm provided by the present invention includes: a data acquisition module, a data processing module and a data output module. The data acquisition module is used to collect relevant data of the filled-in card to be identified, the data processing module is used to perform relevant processing on the data collected by the data acquisition module, and the data output module is used to output the processing results of the data processing module.

[0017] In the above solution, the data acquisition module includes a point cloud data acquisition unit and an image acquisition unit. The point cloud data acquisition unit is used to acquire two-dimensional point cloud data of the filled-in card to be identified, and the image acquisition unit is used to acquire the image of the filled-in card to be identified.

[0018] In the above scheme, the data processing module includes a binarization processing unit, a point cloud homogenization unit and a grid division unit. The binarization processing unit is used to perform binarization processing on the filled-in card image to be identified to obtain a binarized image of the filled-in card; the point cloud homogenization unit is used to evenly tile the two-dimensional point cloud data on the binarized image of the filled-in card; the grid division unit is used to grid the binary image with tiled two-dimensional point cloud data according to the number of questions on the filled-in card and the number of options corresponding to each question, to obtain a corresponding divided grid.

[0019] In the above scheme, the data processing module also includes a weight adjustment unit, a total weight value acquisition unit and a judgment unit. The weight adjustment unit is used to adjust the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binarized image in each area of the divided grid; the total weight value acquisition unit obtains the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binarized image in each area of the divided grid according to the weight of each point in the obtained two-dimensional point cloud data; the judgment unit is used to compare the obtained total weight value with the set total weight value threshold to determine the selected option corresponding to each question on the fill-in-sheet.

[0020] The electronic device provided by the present invention includes a memory, a processor, and a computer program stored in the memory and runnable on the processor, and is characterized in that when the processor executes the program, the steps of the filled-in card image recognition method based on the point cloud algorithm as described above are implemented.

[0021] The non-transitory computer-readable storage medium provided by the present invention stores a computer program thereon, and is characterized in that when the computer program is executed by a processor, the steps of the filled-in card image recognition method based on the point cloud algorithm as described above are implemented.

[0022] The embodiments of the present invention include the following advantages:

[0023] The present invention evenly tiles the two-dimensional point cloud data on the binary image of the filling-in card, grids the binary image with the tiled two-dimensional point cloud data, adjusts the weight of each point in the two-dimensional point cloud data of each area of the divided grid, obtains the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binary image in each area of the divided grid, and compares the obtained total weight value with the set total weight value threshold to determine the selected option corresponding to each question on the filling-in card. The filling-in card can be accurately identified, which overcomes the problem of low recognition accuracy for wrinkled and irregularly filled filling-in cards in the prior art of finding a specific shape of answer sheet selection box through the contour search function of OpenCV, and then determining the color of the selection box after finding the selection box to complete the recognition of the filling-in card. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a flowchart of the steps of an embodiment of a method for image recognition of a filled-in card based on a point cloud algorithm of the present invention.

[0025] Figure 2 Schematic diagram of a binarized image according to the present invention.

[0026] Figure 3 Schematic diagram of point cloud homogenization according to the present invention.

[0027] Figure 4Schematic diagram of image grid division according to the present invention.

[0028] Figure 5 This is a schematic diagram of the composition of a filling-in card image recognition device based on a point cloud algorithm of the present invention. DETAILED DESCRIPTION

[0029] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0030] like Figure 1 As shown, the present invention provides a method for image recognition of a filled-in card based on a point cloud algorithm, comprising:

[0031] Step S1: Obtaining the two-dimensional point cloud data of the filled card to be identified;

[0032] Step S2: obtaining a filled-in card image to be identified, and performing binarization processing on the filled-in card image to be identified to obtain a binarized image of the filled-in card;

[0033] Step S3: evenly tiling the two-dimensional point cloud data on the binary image of the fill-in card;

[0034] Step S4: dividing the binary image of the tiled two-dimensional point cloud data into grids according to the number of questions on the filling-in card and the number of options corresponding to each question to obtain corresponding divided grids;

[0035] Step S5: adjusting the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binary image in each area of the divided grid;

[0036] Step S6: according to the weight of each point in the obtained two-dimensional point cloud data, obtain the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binary image in each area of the divided grid;

[0037] Step S7: Compare the obtained total weight value with the set total weight value threshold to determine the selected option corresponding to each question on the filling-in card.

[0038] like Figure 2 As shown, in the present invention, the filled-in card image to be identified is binarized by the adaptiveThreshold() function in the opencv library to obtain a binary image of the filled-in card, wherein the question number, the options corresponding to each question, and the filled-in area are displayed in white, and other areas are displayed in black.

[0039] Figure 3 The image obtained by evenly tiling the two-dimensional point cloud data on the binarized image of the filled card in the present invention is shown.

[0040] Figure 4 This figure shows the result of meshing a binary image of tiled 2D point cloud data based on the number of questions on the sheet and the number of options for each question.

[0041] In this embodiment, the divided grid is a grid with M rows and N columns, where M is the number of questions on the filling-in card, and N is the number of options corresponding to each question.

[0042] In this embodiment, adjusting the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binarized image in each area of the divided grid includes:

[0043] Increase the weight of each point in the two-dimensional point cloud data in the white pixel area in each area of the divided grid by a specific value;

[0044] The weight of each point in the two-dimensional point cloud data in the black pixel area in each area of the divided grid is reduced by a specific value.

[0045] In this embodiment, the weight of each point in the two-dimensional point cloud data in the white pixel area in each area of the divided grid is increased by a specific value, so that the weight of each point in the two-dimensional point cloud data in the white pixel area in each area is greater than the weight of each point in the two-dimensional point cloud data in the black pixel area in each area, so that the total weight value of the point cloud data corresponding to the filled area is greater than the total weight value of the point cloud data corresponding to other areas, so that the filled options can be accurately and quickly identified.

[0046] like Figure 5 As shown, the present invention provides a filled-in card image recognition device based on a point cloud algorithm, comprising: a data acquisition module, a data processing module and a data output module, wherein the data acquisition module is used to collect relevant data of the filled-in card to be identified, the data processing module is used to perform relevant processing on the data collected by the data acquisition module, and the data output module is used to output the processing results of the data processing module.

[0047] Furthermore, the data acquisition module includes a point cloud data acquisition unit and an image acquisition unit. The point cloud data acquisition unit is used to acquire two-dimensional point cloud data of the filled-in card to be identified, and the image acquisition unit is used to acquire an image of the filled-in card to be identified.

[0048] Furthermore, the data processing module includes a binarization processing unit, a point cloud homogenization unit and a grid division unit. The binarization processing unit is used to perform binarization processing on the filled-in card image to be identified to obtain a binarized image of the filled-in card; the point cloud homogenization unit is used to evenly tile the two-dimensional point cloud data on the binarized image of the filled-in card; the grid division unit is used to grid the binary image with tiled two-dimensional point cloud data according to the number of questions on the filled-in card and the number of options corresponding to each question to obtain a corresponding divided grid.

[0049] Furthermore, the data processing module also includes a weight adjustment unit, a total weight value acquisition unit and a judgment unit. The weight adjustment unit is used to adjust the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binarized image in each area of the divided grid; the total weight value acquisition unit obtains the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binarized image in each area of the divided grid according to the weight of each point in the obtained two-dimensional point cloud data; the judgment unit is used to compare the obtained total weight value with the set total weight value threshold to determine the selected option corresponding to each question on the fill-in-sheet.

[0050] The present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for image recognition of a filled-in card based on a point cloud algorithm when executing the program.

[0051] The present invention provides a non-transitory 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 above-mentioned method for image recognition of a filled-in card based on a point cloud algorithm are implemented.

[0052] It should be noted that the above detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs.

[0053] It should be noted that the terms used herein are intended only to describe specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0054] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0055] In addition, the terms "comprises" and "comprising" and any variations thereof are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, product, or apparatus.

[0056] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used herein to describe the spatial positional relationship of a device or feature to other devices or features as shown in the figures. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figures. For example, if the device in the drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be subsequently positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways, such as rotated 90 degrees or in other orientations, and the spatially relative descriptions used herein are interpreted accordingly.

[0057] In the above detailed description, reference is made to the accompanying drawings, which form a part hereof. In the drawings, similar symbols typically identify similar components, unless the context dictates otherwise. The illustrated embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be used, and other changes may be made, without departing from the spirit or scope of the subject matter presented herein.

[0058] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for image recognition of a filled-in card based on a point cloud algorithm, characterized in that: The method comprises: Obtain the two-dimensional point cloud data of the filled card to be identified; Obtaining a filled-in card image to be identified, and performing binarization processing on the filled-in card image to be identified to obtain a binarized image of the filled-in card; Evenly tiling the two-dimensional point cloud data on the binary image of the fill-in card; According to the number of questions on the fill-in card and the number of options corresponding to each question, the binary image of the tiled two-dimensional point cloud data is grid-divided to obtain the corresponding grid; Adjusting the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binary image in each area of the divided grid, including: increasing the weight of each point in the two-dimensional point cloud data in the white pixel area of each area of the divided grid by a specific value; and decreasing the weight of each point in the two-dimensional point cloud data in the black pixel area of each area of the divided grid by a specific value; According to the weight of each point in the obtained two-dimensional point cloud data, the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binary image in each area of the divided grid is obtained; The obtained total weight value is compared with the set total weight value threshold to determine the selected option corresponding to each question on the fill-in card.

2. The method for image recognition of a filled-in card based on a point cloud algorithm according to claim 1, characterized in that: The divided grid is a grid with M rows and N columns, where M is the number of questions on the filling-in card and N is the number of options corresponding to each question.

3. A filling card image recognition device based on point cloud algorithm, characterized in that: The device includes: a data acquisition module, a data processing module and a data output module, the data acquisition module is used to collect relevant data of the filled-in card to be identified, the data acquisition module includes a point cloud data acquisition unit and an image acquisition unit, the point cloud data acquisition unit is used to collect two-dimensional point cloud data of the filled-in card to be identified, and the image acquisition unit is used to collect images of the filled-in card to be identified; The data processing module is used to perform relevant processing on the data collected by the data acquisition module, and the data processing module includes a binarization processing unit, a point cloud homogenization unit, a grid division unit, a weight adjustment unit, a weight total value acquisition unit and a judgment unit. The binarization processing unit is used to perform binarization processing on the fill-in card image to be identified to obtain a binarized image of the fill-in card; the point cloud homogenization unit is used to evenly tile the two-dimensional point cloud data on the binary image of the fill-in card; the grid division unit is used to tile the tiled two-dimensional point cloud data according to the number of questions on the fill-in card and the number of options corresponding to each question. The binarized image is grid-divided to obtain corresponding divided grids; the weight adjustment unit is used to adjust the weight of each point in the two-dimensional point cloud data of each area of the divided grid according to the pixel areas of different colors of the binarized image in each area of the divided grid; the total weight value acquisition unit acquires the total weight value of the point cloud data corresponding to the pixel areas of different colors of the binarized image in each area of the divided grid according to the weight of each point in the acquired two-dimensional point cloud data; the judgment unit is used to compare the acquired total weight value with a set total weight value threshold to determine the selected option corresponding to each question on the fill-in-the-blank card; The adjusting the weight of each point in the two-dimensional point cloud data of each divided grid area according to the pixel areas of different colors of the binarized image in each divided grid area includes: increasing the weight of each point in the two-dimensional point cloud data in the white pixel area in each divided grid area by a specific value; and decreasing the weight of each point in the two-dimensional point cloud data in the black pixel area in each divided grid area by a specific value; The data output module is used to output the processing results of the data processing module.

4. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the filled-in card image recognition method based on the point cloud algorithm as described in any one of claims 1 to 2 are implemented.

5. A non-transitory 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 filled-in card image recognition method based on the point cloud algorithm as described in any one of claims 1 to 2 are implemented.

Images