Score identification method and device for test questions with scores, and storage medium

Abstract

The invention discloses a score recognition method and device for test questions with score strips and a storage medium, and the method comprises the steps: obtaining a to-be-recognized test paper image which at least comprises one test question with the score strips; the scoring strip is provided with a plurality of grids, and each grid is printed with a score value; when the test paper is corrected, marking the lattices of the corresponding scoring strips with specified colors according to the scores of the test questions so as to mark the scores of the corresponding test questions; positioning the position of a scoring strip in the test paper image to be identified; performing context interference removal processing on the positioned score bars; and finally, identifying the scores recorded in the score bar, and assigning the scores to the corresponding test questions according to the identified scores. According to the method, the score bar is positioned on the test paper image to be identified, reasonable interference removal processing is performed on the area outside the score bar, and then the correction trace and the corresponding score in the score bar are identified, so that the score of the test question can be obtained more accurately.

Description

Technical Field

[0001] The present invention relates to the field of intelligent education technology, and in particular to a score recognition method, device and storage medium for a test question with a scoring strip. Background Art

[0002] As education informatization continues to advance, question cards or teaching aids with scoring strips are increasingly being used by school students. Figure 1 As shown, the test questions and the score strip correspond one to one, and the score strip has multiple grids, each of which is pre-printed with multiple score values. When grading, the teacher directly marks the score value obtained by the test question with a red pen to record the score of the test question. The conventional score strip recognition method is to first identify the correction traces on the corrected test paper image, and then calculate the position of the coordinates of the midpoints of each correction trace on the score strip to obtain the score value marked by the teacher. However, the marking methods of teachers vary greatly, and this recognition method is prone to score recognition errors. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention aims to provide a method, device and storage medium for scoring recognition of test questions with scoring strips.

[0004] In order to achieve the above object, the present invention adopts the following technical solution:

[0005] A method for identifying scores of test questions with scoring strips, the specific process is as follows:

[0006] S1. Obtaining a test paper image to be identified, wherein the test paper image to be identified includes at least one test question with a scoring strip; the scoring strip has a plurality of grids, each of which is printed with a score value; when marking the test paper, marking the grid of the scoring strip with a designated color according to the score of the test question to mark the score of the corresponding test question;

[0007] S2, locating the scoring bar in the test paper image to be identified obtained in step S1;

[0008] S3. Perform context interference removal on the located scoring strips in any of the following ways:

[0009] (1) By presetting the threshold range, the area within the threshold range outside the scoring strip is de-interferenced;

[0010] (2) If the area where the test question is located is above or below the corresponding scoring bar, the heights of the areas corresponding to the two test questions above and below the scoring bar are dynamically obtained, and interference removal is performed on the range that does not exceed the heights of the areas corresponding to the two test questions above and below the scoring bar;

[0011] S4. Identify the designated color on the scoring bar, then detect the grid of the scoring bar where the designated color is located and further identify the score value in the grid, so as to obtain the score of the corresponding test question.

[0012] Furthermore, in step S1, after acquiring the test paper image to be identified, image correction processing is first performed on the test paper image to be identified.

[0013] As one of the solutions, in step S2, target detection is used to locate the scoring strips:

[0014] S2.1, collect image sample data of the scoring strips and construct a sample set;

[0015] S2.2, obtaining a scoring object detection model based on deep learning by training the sample set constructed in step S2.1;

[0016] S2.3. Input the test paper image to be identified obtained in step S1 into the scoring stripe target detection model, and the scoring stripe target detection model outputs the positioning result of the scoring stripe in the test paper image to be identified.

[0017] As another solution, in step S2, the scoring bar of the test paper image to be identified is located by obtaining preset first parameter information; the first parameter information includes position data of the scoring bar in the test paper template.

[0018] Furthermore, in step S3, the interference removal process is specifically to erase and / or block part or all of the data outside the scoring strip.

[0019] Furthermore, in step S3, for method (2), the horizontal widths of the areas corresponding to the two test questions above and below the scoring bar are set to fixed values.

[0020] Further, in step S4, if more than two positions with designated colors are identified on the scoring bar, and the designated colors of the two or more positions fall on different grids of the scoring bar respectively, then the score of the corresponding test question is determined to be a combination of the score values ​​in each grid from left to right in order.

[0021] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0022] The present invention also provides a computer device, comprising a processor and a memory, wherein the memory is used to store a computer program; and when the processor is used to execute the computer program, the above method is implemented.

[0023] The beneficial effect of the present invention is that the method of the present invention first locates the score strip on the test paper image to be identified, and reasonably removes interference from the area outside the score strip, and then identifies the correction marks in the score strip, and then obtains the score recorded on the grid where the correction marks are located, so as to more accurately obtain the score value of the test question on the score strip. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is an example of an existing test paper with grading strips. DETAILED DESCRIPTION

[0025] The present invention will be further described below in conjunction with the accompanying drawings. It should be noted that this embodiment is based on the technical solution and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to this embodiment.

[0026] This embodiment provides a method for identifying the score of a test question with a scoring bar, and the specific process is as follows:

[0027] S1. Obtain a test paper image to be identified, wherein the test paper image to be identified includes at least one test question with a scoring strip; the scoring strip has multiple grids, each of which is printed with a score value (generally 0-9, and the score values ​​in each grid may or may not be repeated). When marking the test paper, the teacher uses a pen of a specified color (such as red) to mark the corresponding grid of the scoring strip according to the score of the test question, such as drawing a slash or ticking.

[0028] S2. Locate the scoring bar in the test paper image to be identified obtained in step S1.

[0029] S3. Perform context interference removal on the located scoring strips.

[0030] S4, identifying the designated color on the scoring bar, then detecting the grid of the scoring bar where the designated color is located and further identifying the score value in the grid, thereby obtaining the score of the corresponding test question. Specifically, the range of the grid where the designated color is located is determined by identifying the smallest square box circumscribed by the designated color. In addition, if more than two positions on the scoring bar are identified to have the designated color, and the designated colors at more than two positions fall on different grids of the scoring bar, then the score of the corresponding test question is determined to be a combination of the score values ​​in each grid from left to right in order.

[0031] In this embodiment, in step S1, after acquiring the test paper image to be identified, image correction processing is first performed on the test paper image to be identified, and specifically, feature alignment image correction processing can be adopted.

[0032] As one implementation method, in step S2, target detection is used to locate the scoring strips:

[0033] S2.1, collect image sample data of the scoring strips and construct a sample set;

[0034] S2.2, training the sample set constructed in step S2.1 to obtain a deep learning-based scoring target detection model (such as a YOLOv target detection model);

[0035] S2.3. Input the test paper image to be identified obtained in step S1 into the scoring stripe target detection model, and the scoring stripe target detection model outputs the positioning result of the scoring stripe in the test paper image to be identified.

[0036] As another implementation, in step S2, the scoring strip of the test paper image to be identified is located by obtaining the preset first parameter information; the first parameter information includes the position data of the scoring strip in the test paper template. The test paper template can be an image or a file, and the position data of the scoring strip can be obtained by automatically or manually marking the scoring strip.

[0037] In this embodiment, in step S3, the interference removal process is specifically to erase and / or block part or all of the data outside the scoring strip.

[0038] As an implementation method, by presetting a threshold range, in the interference removal process, the area within the threshold range outside the scoring bar is erased and / or blocked, so that the interference removal range is not too large and the scoring bar is also erased or blocked. Specifically, a range of a set size in the horizontal direction and / or vertical direction outside the scoring bar can be set as the threshold range.

[0039] As another implementation, the area where the test questions are located (including the test question content and the answer area, and if it is an answer sheet, it only includes the answer area) is located above or below the corresponding scoring bar, and the heights of the areas corresponding to the two test questions above and below the scoring bar are dynamically obtained. In the de-interference process, the range that does not exceed the height of the areas corresponding to the two test questions above and below the scoring bar is blocked and / or erased. This is equivalent to dynamically determining the threshold range for de-interference processing, so as to adapt to more different test paper images to be identified. Furthermore, the horizontal width of the area corresponding to the two test questions above and below the scoring bar can be set to a fixed value, so that only the height of the area corresponding to the two test questions above and below the scoring bar needs to be dynamically obtained to determine the size of the area that needs to be de-interference processed.

[0040] For those skilled in the art, various corresponding changes and modifications can be made according to the above technical solutions and concepts, and all of these changes and modifications should be included in the protection scope of the claims of the present invention.

Claims

1. A scoring identification method for test questions with scoring strips, characterized in that The specific process is: S1. Obtaining a test paper image to be identified, wherein the test paper image to be identified includes at least one test question with a scoring strip; the scoring strip has a plurality of grids, each of which is printed with a score value; when marking the test paper, marking the grid of the scoring strip with a designated color according to the score of the test question to mark the score of the corresponding test question; S2, locating the scoring bar in the test paper image to be identified obtained in step S1; S3. Perform context interference removal on the located scoring strips in any of the following ways: (1) By presetting the threshold range, the area within the threshold range outside the scoring strip is de-interferenced; (2) If the area where the test question is located is above or below the corresponding scoring bar, the heights of the areas corresponding to the two test questions above and below the scoring bar are dynamically obtained, and interference removal is performed on the range that does not exceed the heights of the areas corresponding to the two test questions above and below the scoring bar; S4. Identify the designated color on the scoring bar, then detect the grid of the scoring bar where the designated color is located and further identify the score value in the grid, so as to obtain the score of the corresponding test question.

2. The method according to claim 1, characterized in that In step S1, after obtaining the test paper image to be identified, image correction processing is first performed on the test paper image to be identified.

3. The method according to claim 1, characterized in that In step S2, target detection is used to locate the scoring strips: S2.1, collect image sample data of the scoring strips and construct a sample set; S2.2, obtaining a scoring object detection model based on deep learning by training the sample set constructed in step S2.1; S2.

3. Input the test paper image to be identified obtained in step S1 into the scoring stripe target detection model, and the scoring stripe target detection model outputs the positioning result of the scoring stripe in the test paper image to be identified.

4. The method according to claim 1, characterized in that: In step S2, the scoring bar of the test paper image to be identified is located by obtaining the preset first parameter information; the first parameter information includes the position data of the scoring bar in the test paper template.

5. The method according to claim 1, characterized in that In step S3, the interference removal process specifically includes erasing and / or blocking part or all of the data outside the scoring strips.

6. The method according to claim 1, characterized in that In step S3, for method (2), the horizontal widths of the areas corresponding to the two test questions above and below the scoring bar are set to fixed values.

7. The method according to claim 1, characterized in that In step S4, if more than two positions with designated colors are identified on the scoring bar, and the designated colors of the two or more positions fall on different grids of the scoring bar respectively, then the score of the corresponding test question is determined to be a combination of the score values ​​in each grid from left to right in order.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

9. A computer device, characterized in that: The method comprises a processor and a memory, wherein the memory is used to store a computer program; when the processor is used to execute the computer program, the method described in any one of claims 1 to 7 is implemented.

Images