A method and device for detecting rotating Braille squares based on block detection preprocessing

Through the rotating Braille detection method based on block detection preprocessing, the M1 and M2 networks trained by the Rotated-RetinaNet network are solved by using the problem that traditional technology cannot effectively detect rotating Braille, which significantly improves the detection accuracy, especially in handwritten Braille scenes Down.

CN118429978BActive Publication Date: 2024-11-15HANGZHOU HUIGUANG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410529923.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-29
Publication Date
2024-11-15
Estimated Expiration
2044-04-29

AI Technical Summary

Technical Problem

Traditional object detection technology cannot effectively detect the rotating Braille square, resulting in limited detection effect of rotating targets. Especially in handwritten Braille scenes, the characteristics of Braille squares with smaller rotation angles and larger rotation angles are easily confused and affect the detection efficiency.

Method used

Using the rotating Braille square detection method based on block detection preprocessing, the M1 and M2 networks obtained through the Rotated-RetinaNet network training obtain the Braille block information and Braille square information in the Braille pictures, and crop and rotate the Braille block pictures, and detect it. The Braille information maps back the Braille pictures to be detected and integrates the final detection results.

Benefits of technology

It significantly improves the detection effect of the rotating Braille square, avoids the confusion of the characteristics of the rotating Braille square by rotating target detection, improves the existing Braille square detection method, and improves the detection accuracy in handwritten Braille scenes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118429978B_ABST
    Figure CN118429978B_ABST
Patent Text Reader

Abstract

The present invention provides a method and device for detecting a rotated Braille square based on block detection preprocessing, the method comprising: S110: obtaining a Braille picture to be detected; S120: obtaining all Braille block information in the Braille picture to be detected based on an M1 network trained by a Rotated‑RetinaNet network; S130: cropping all Braille blocks from the Braille picture to be detected based on the Braille block information to obtain a Braille block picture corresponding thereto; S140: rotating the Braille block picture to a horizontal direction; S150: detecting Braille square information existing in all Braille block pictures based on an M2 network trained by a Rotated‑RetinaNet network; S160: mapping the detected Braille square back to the Braille picture to be detected based on the Braille square information, and integrating all Braille squares to obtain a final detection result. The existing Braille square detection method is improved, and a Braille block detection preprocessing is introduced, which solves the problem that the detection effect of a rotated Braille square by a rotated target detection is limited, and the detection effect of a rotated Braille square in a handwritten Braille scene is greatly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of Braille square detection, and in particular to a rotation Braille square detection method and device based on block detection preprocessing. Background Art

[0002] Sighted people have not received special training and are inefficient in reading Braille. Errors will inevitably occur in handwritten Braille, which further affects the efficiency of sighted people in reading Braille. To convert Braille into Chinese documents that are easy for sighted people to read, or to process it into speech, it is usually necessary to use target detection technology to detect the Braille squares in Braille. Traditional technology directly applies target detection technology to Braille square detection. Braille handwritten with the help of Braille writing tools will have rotated Braille squares, and target detection technology cannot effectively detect rotated Braille squares. Because the features of the rotated Braille squares are complex, the features of the Braille squares with smaller rotation angles and the Braille squares with larger rotation angles are easily confused, so the detection effect of the rotation target detection technology is also limited.

[0003] The above problems need to be solved urgently. Summary of the invention

[0004] The present invention aims to overcome at least one of the above-mentioned disadvantages of the prior art. On the one hand, a rotated Braille square detection method based on block detection preprocessing is provided, and the method comprises: S110: obtaining a Braille picture to be detected; S120: obtaining all Braille block information in the Braille picture to be detected based on an M1 network obtained by training a Rotated-RetinaNet network; S130: cropping all Braille blocks from the Braille picture to be detected based on the Braille block information to obtain a Braille block picture corresponding thereto; S140: rotating the Braille block picture to a horizontal direction; S150: detecting Braille square information existing in all Braille block pictures based on an M2 network obtained by training a Rotated-RetinaNet network; S160: mapping the detected Braille squares back to the Braille picture to be detected based on the Braille square information, and integrating all Braille squares to obtain a final detection result.

[0005] Furthermore, the M1 network is trained by combining a Rotated-RetinaNet network with a Braille image and corresponding Braille block annotation information.

[0006] Furthermore, the obtaining of all braille block information in the braille image to be detected based on the M1 network trained by the Rotated-RetinaNet network includes: preprocessing the braille image to be detected to adjust the format and size of the braille image to be detected; inputting the preprocessed braille image to be detected into the M1 network to obtain the braille block information in the braille image to be detected; the braille block information includes the center coordinates of the braille block, the width of the braille block, the height of the braille block and the rotation angle of the braille block.

[0007] Furthermore, cropping all braille blocks from the braille image to be detected based on the braille block information to obtain a corresponding braille block image includes: cropping the braille blocks from the braille image to be detected based on a cv2.getRotationMatrix2D function, a cv2.warpAffine function, and a cv2.bitwise_and function.

[0008] Furthermore, rotating the braille block image to a horizontal direction includes: sequentially rotating the braille block image with the center coordinates of the braille block as the rotation center, and rotating the cropped braille block image to a horizontal direction based on the cv2.getRotationMatrix2D function and the cv2.warpAffine function.

[0009] Furthermore, rotating the braille block image to a horizontal direction includes: sequentially rotating the braille block image with the central coordinates of the braille block as the rotation center and at a rotation angle of the braille block, so that the braille block image is rotated to a horizontal direction.

[0010] Furthermore, the M2 network is trained by combining a Rotated-RetinaNet network with a Braille image and corresponding Braille annotation information.

[0011] Furthermore, the detecting of the Braille square information in all the Braille blocks based on the M2 network obtained through training of the Rotated-RetinaNet network includes: preprocessing the Braille block image; inputting the preprocessed Braille block image into the M2 network to obtain the Braille square information in the Braille block; the Braille square information includes the center coordinates of the Braille square, the width of the Braille square, the height of the Braille square, the rotation angle of the Braille square and the Braille square category.

[0012] Furthermore, mapping the detected Braille squares back to the Braille image to be detected based on the Braille square information, and integrating all Braille squares to obtain the final detection result includes: mapping the Braille squares of each Braille block back to the Braille image to be detected in turn; rotating the Braille squares by a corresponding angle in the opposite direction to step S140; and merging all the rotated Braille squares to generate the final detection result.

[0013] In a second aspect, the present invention provides a rotated Braille square detection device based on block detection preprocessing, the device comprising: an acquisition unit for obtaining a Braille picture to be detected, suitable for obtaining a Braille picture to be detected; an acquisition unit for obtaining Braille block information, suitable for obtaining all Braille block information in the Braille picture to be detected based on an M1 network trained by a Rotated-RetinaNet network; an acquisition unit for Braille block pictures, suitable for cropping all Braille blocks from the Braille picture to be detected based on the Braille block information to obtain a Braille block picture corresponding thereto; a rotation unit, suitable for rotating the Braille block picture to a horizontal direction; a detection unit for detecting Braille square information, suitable for detecting Braille square information existing in all Braille block pictures based on an M2 network trained by a Rotated-RetinaNet network; and a detection result generation unit, suitable for mapping the detected Braille squares back to the Braille picture to be detected based on the Braille square information, and integrating all Braille squares to obtain a final detection result.

[0014] In a third aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more instructions, and the computer instructions are used to enable the computer to execute the above-mentioned rotated Braille square detection method based on block detection preprocessing.

[0015] On the other hand, the present invention provides an electronic device, comprising: a memory and a processor; the memory stores at least one program instruction; the processor implements the above-mentioned rotated Braille square detection method based on block detection preprocessing by loading and executing the at least one program instruction.

[0016] The beneficial effects of the present invention are as follows: the present invention provides a rotated Braille square detection method based on block detection preprocessing, characterized in that the method comprises: S110: obtaining a Braille picture to be detected; S120: obtaining all Braille block information in the Braille picture to be detected based on an M1 network obtained through Rotated-RetinaNet network training; S130: cropping all Braille blocks from the Braille picture to be detected based on the Braille block information to obtain a Braille block picture corresponding thereto; S140: rotating the Braille block picture to a horizontal direction; S150: detecting Braille square information existing in all Braille block pictures based on an M2 network obtained through Rotated-RetinaNet network training; S160: mapping the detected Braille squares back to the Braille picture to be detected based on the Braille square information, and integrating all Braille squares to obtain a final detection result. The method detects rotated Braille squares in Braille images through a block detection preprocessing method, improves the existing Braille square detection method, introduces Braille block detection preprocessing, solves the problem that the detection effect of rotated Braille squares by rotated target detection is limited, and greatly improves the detection effect of rotated Braille squares in handwritten Braille scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0018] Figure 1 It is a flow chart of a method for detecting a rotated Braille square based on block detection preprocessing provided by an embodiment of the present invention.

[0019] Figure 2 It is a schematic diagram of Braille squares and Braille blocks contained in a Braille picture to be detected provided by an embodiment of the present invention.

[0020] Figure 3 It is a structural schematic diagram of a rotated Braille square detection device based on block detection preprocessing provided by an embodiment of the present invention.

[0021] Figure 4 It is a partial block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] It should be mentioned before discussing the exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations as sequential processes, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0023] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0024] The present invention will now be described in detail with reference to the accompanying drawings. This figure is a simplified schematic diagram, which only illustrates the basic structure of the present invention in a schematic manner, and therefore only shows the components related to the present invention.

[0025] Example 1

[0026] like Figure 1 As shown, it is a flow chart of a rotation Braille square detection method based on block detection preprocessing provided by the present invention.

[0027] As an example, combining Figure 2 , the method comprising:

[0028] S110: Obtain a Braille image to be detected.

[0029] S120: Acquire all Braille block information in the Braille image to be detected based on the M1 network trained by the Rotated-RetinaNet network.

[0030] Preferably, the M1 network is trained by combining a Rotated-RetinaNet network with a braille image and corresponding braille block annotation information. Specifically, the M1 network is trained by combining a Rotated-RetinaNet network (which is a rotated target detection version of RetinaNet, which adds the angle of the prediction box on the basis of horizontal detection) with a braille image and corresponding braille block annotation information (including the position of the braille block) to detect the braille blocks in the image.

[0031] Preferably, the method of obtaining all the braille block information in the braille image to be detected based on the M1 network obtained through training of the Rotated-RetinaNet network includes: preprocessing the braille image to be detected to adjust the format and size of the braille image to be detected; inputting the preprocessed braille image to be detected into the M1 network to obtain the braille block information in the braille image to be detected; the braille block information includes the center coordinates of the braille block, the width of the braille block, the height of the braille block, and the rotation angle of the braille block. Specifically, the input braille image to be detected is adjusted to the same size input and format, and then all the braille block information {b1, b2, ..., bk} contained in the image is obtained through the M1 network, wherein each braille block bi is composed of {(xi, yi), wi, hi, anglei}, (xi, yi) represents the center coordinates of the braille block, wi represents the width of the braille block, hi represents the height of the braille block, and anglei represents the rotation angle of the braille block.

[0032] S130: based on the braille block information, cropping all braille blocks from the braille image to be detected to obtain a corresponding braille block image.

[0033] Preferably, the step of cropping all the braille blocks from the braille picture to be detected based on the braille block information to obtain the corresponding braille block picture comprises: cropping the braille blocks from the braille picture to be detected based on the cv2.getRotationMatrix2D function, the cv2.warpAffine function and the cv2.bitwise_and function.

[0034] S140: Rotate the braille block image to a horizontal direction.

[0035] Preferably, rotating the braille block image to the horizontal direction comprises: sequentially rotating the braille block image to the horizontal direction with the center coordinates of the braille block as the rotation center based on the cv2.getRotationMatrix2D function and the cv2.warpAffine function. Specifically, according to all the detected braille blocks {b1, b2, ..., bk}, all the braille blocks are cropped from the original braille image to be detected, and then the cropped braille blocks are rotated to the horizontal direction with the center coordinates (xi, yi) as the rotation center according to their own rotation angle anglei.

[0036] S150: Detecting the Braille information in all the Braille block images based on the M2 network trained by the Rotated-RetinaNet network.

[0037] Preferably, the M2 network is trained by combining a Rotated-RetinaNet network with a braille image and corresponding braille square annotation information. Specifically, the M2 network is trained by combining a Rotated-RetinaNet network with a braille image and corresponding braille square annotation information (including the position and category of the braille square) to detect the braille squares in the image.

[0038] Preferably, the detecting the Braille square information existing in all Braille blocks based on the M2 network obtained through Rotated-RetinaNet network training includes: preprocessing the Braille block image; inputting the preprocessed Braille block image into the M2 network to obtain the Braille square information existing in the Braille block; the Braille square information includes the center coordinates of the Braille square, the width of the Braille square, the height of the Braille square, the rotation angle of the Braille square and the Braille square category. Specifically, the braille block image is preprocessed to unify the input size and format, and then the image is input into the Rotated-RetinaNet detection network M2 to obtain all the braille squares {br1, br2, ..., brm} contained in the braille block image. Each braille square brj is composed of {(xj, yj), wj, hj, anglej, clsi}, (xj, yj) represents the center coordinates of the braille square, wj represents the width of the braille square, hj represents the height of the braille square, anglej represents the rotation angle of the braille square, and clsi represents the category of the braille square. There are 63 types of braille squares excluding blank braille squares.

[0039] S160: Mapping the detected Braille squares back to the Braille image to be detected based on the Braille square information, and integrating all the Braille squares to obtain a final detection result.

[0040] Preferably, mapping the detected Braille squares back to the Braille image to be detected based on the Braille square information, and integrating all Braille squares to obtain the final detection result includes: mapping the Braille squares of each Braille block back to the Braille image to be detected in turn; rotating the Braille squares by a corresponding angle in the opposite direction to step S140; and merging all the rotated Braille squares to generate the final detection result. Specifically, for the Braille squares {br1, br2, ..., brm}, according to the Braille block bi to which each Braille square brj belongs, the Braille square is rotated with the center coordinates of the Braille block as the rotation point, mapped to the position on the original image according to anglei, and then the Braille squares in all Braille blocks are merged to obtain the final Braille square detection result.

[0041] The above embodiment provides a method for detecting a rotated Braille square based on block detection preprocessing, wherein the method adds block detection preprocessing to the traditional Braille square detection method to detect the Braille squares in the Braille image, and converts all rotated Braille square detections into small-angle rotated Braille square detections, effectively avoiding confusion between rotated Braille square features at smaller angles and larger angles, and significantly improving the detection effect of rotated Braille squares in handwritten Braille scenarios.

[0042] Example 2

[0043] See also Figure 3 , this embodiment provides a structural schematic diagram of a rotating Braille square detection device based on block detection preprocessing.

[0044] As an example, the device comprises:

[0045] The unit 310 for obtaining a Braille picture to be detected is adapted to obtain a Braille picture to be detected.

[0046] The braille block information acquisition unit 320 is adapted to acquire all braille block information in the braille image to be detected based on the M1 network obtained by training the Rotated-RetinaNet network.

[0047] The braille block image obtaining unit 330 is adapted to crop all braille blocks from the braille image to be detected based on the braille block information to obtain a corresponding braille block image.

[0048] The rotation unit 340 is adapted to rotate the braille block image to a horizontal direction.

[0049] The Braille square information detection unit 350 is adapted to detect the Braille square information existing in all Braille block images based on the M2 network obtained by training the Rotated-RetinaNet network.

[0050] The detection result generating unit 360 is adapted to map the detected Braille squares back to the Braille image to be detected based on the Braille square information, and integrate all the Braille squares to obtain a final detection result.

[0051] It is not difficult to find that this embodiment is a system embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the first embodiment.

[0052] It is worth mentioning that all units involved in this embodiment are logical units. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by the present invention, but this does not mean that there are no other units in this embodiment.

[0053] Example 3

[0054] The embodiment of the present invention further provides a storage medium, on which a rotational Braille square detection method based on block detection preprocessing is stored, and when the rotational Braille square detection program based on block detection preprocessing is executed by a processor, the steps of the rotational Braille square detection method based on block detection preprocessing as described above are implemented. Since the storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described one by one here.

[0055] Example 4

[0056] See also Figure 4 An embodiment of the present invention further provides an electronic device, comprising: a memory and a processor; the memory stores at least one program instruction; the processor implements the rotating Braille square detection method based on block detection preprocessing provided in Example 1 by loading and executing the at least one program instruction.

[0057] The memory 602 and the processor 601 are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors 601 and the memory 602 together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor 601 is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor 601.

[0058] The processor 601 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management and other control functions. The memory 602 can be used to store data used by the processor 601 when performing operations.

[0059] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can know all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for detecting a rotating Braille square based on block detection preprocessing, characterized in that: The method comprises: S110: Obtain a Braille image to be detected; S120: Acquire all Braille block information in the Braille image to be detected based on the M1 network trained by the Rotated-RetinaNet network; S130: based on the braille block information, cropping all braille blocks from the braille image to be detected to obtain a braille block image corresponding thereto; S140: Rotate the braille block image to a horizontal direction; S150: Detecting the Braille square information in all the Braille block images based on the M2 network obtained by training the Rotated-RetinaNet network, including: preprocessing the Braille block images; inputting the preprocessed Braille block images into the M2 network to obtain the Braille square information in the Braille blocks; the Braille square information includes the center coordinates of the Braille square, the width of the Braille square, the height of the Braille square, the rotation angle of the Braille square and the Braille square category; S160: mapping the detected Braille squares back to the Braille image to be detected based on the Braille square information, and integrating all the Braille squares to obtain the final detection result, including: mapping the Braille squares of each Braille block back to the Braille image to be detected in turn; rotating the Braille squares by a corresponding angle in the opposite direction to step S140; merging all the rotated Braille squares to generate the final detection result, including: for the Braille squares {br1, br2, …, brm}, according to the Braille block bi to which each Braille square brj belongs, the Braille square is rotated with the center coordinates of the Braille block as the rotation point, mapped to the position on the original image according to the rotation angle of the Braille square, and the Braille squares in all the Braille blocks are merged to obtain the final Braille square detection result.

2. The method for detecting a rotating Braille square based on block detection preprocessing according to claim 1, characterized in that: The M1 network is trained by combining the Rotated-RetinaNet network with the Braille image and the corresponding Braille block annotation information.

3. The method for detecting a rotating Braille square based on block detection preprocessing according to claim 2, characterized in that: The method of obtaining all the Braille block information in the Braille image to be detected based on the M1 network obtained through Rotated-RetinaNet network training includes: Preprocessing the Braille image to be detected to adjust the format and size of the Braille image to be detected; Inputting the preprocessed Braille image to be detected into the M1 network to obtain Braille block information in the Braille image to be detected; The braille block information includes the center coordinates of the braille block, the width of the braille block, the height of the braille block and the rotation angle of the braille block.

4. The method for detecting a rotating Braille square based on block detection preprocessing according to claim 1, characterized in that: The step of cropping all the braille blocks from the to-be-detected braille picture based on the braille block information to obtain a braille block picture corresponding thereto comprises: The braille block is cropped from the braille image to be detected based on the cv2.getRotationMatrix2D function, the cv2.warpAffine function and the cv2.bitwise_and function.

5. The method for detecting a rotating Braille square based on block detection preprocessing according to claim 1, characterized in that: The step of rotating the braille block image to a horizontal direction comprises: The braille block image is rotated to the horizontal direction based on the cv2.getRotationMatrix2D function and the cv2.warpAffine function with the center coordinates of the braille block as the rotation center.

6. The method for detecting a rotating Braille square based on block detection preprocessing according to claim 5, characterized in that: The step of rotating the braille block image to a horizontal direction comprises: The braille block image is rotated in sequence with the central coordinates of the braille block as the rotation center and the rotation angle of the braille block, so that the braille block image is rotated to a horizontal direction.

7. The method for detecting a rotating Braille square based on block detection preprocessing according to claim 1, characterized in that: The M2 network is trained by combining the Rotated-RetinaNet network with the Braille image and the corresponding Braille annotation information.

8. A rotating Braille square detection device based on block detection preprocessing, characterized in that: The device comprises: A unit for obtaining a Braille image to be detected is used for obtaining a Braille image to be detected; Obtaining the Braille block information unit, which is suitable for obtaining all the Braille block information in the Braille image to be detected based on the M1 network trained by the Rotated-RetinaNet network; A braille block image acquisition unit is used to crop all braille blocks from the braille image to be detected based on the braille block information to obtain a braille block image corresponding thereto; A rotation unit, adapted to rotate the braille block image to a horizontal direction; A unit for detecting braille square information is used to detect the braille square information existing in all braille block images based on the M2 network obtained by training the Rotated-RetinaNet network, including: preprocessing the braille block image; inputting the preprocessed braille block image into the M2 network to obtain the braille square information existing in the braille block; the braille square information includes the center coordinates of the braille square, the width of the braille square, the height of the braille square, the rotation angle of the braille square and the braille square category; A detection result generating unit is adapted to map the detected Braille squares back to the Braille image to be detected based on the Braille square information, and integrate all the Braille squares to obtain a final detection result, including: mapping the Braille squares of each Braille block back to the Braille image to be detected in turn; rotating the Braille squares by a corresponding angle in a direction opposite to step S140; merging all the rotated Braille squares to generate a final detection result, including: for the Braille squares {br1, br2, …, brm}, according to the Braille block bi to which each Braille square brj belongs, the Braille square is rotated with the center coordinates of the Braille block as a rotation point, mapped to a position on the original image according to the rotation angle of the Braille square, and the Braille squares in all the Braille blocks are merged to obtain a final Braille square detection result.