CN118429978B - Rotary braille square detection method and device based on block detection pretreatment - Google Patents
Rotary braille square detection method and device based on block detection pretreatment Download PDFInfo
- Publication number
- CN118429978B CN118429978B CN202410529923.XA CN202410529923A CN118429978B CN 118429978 B CN118429978 B CN 118429978B CN 202410529923 A CN202410529923 A CN 202410529923A CN 118429978 B CN118429978 B CN 118429978B
- Authority
- CN
- China
- Prior art keywords
- braille
- block
- square
- pictures
- detected
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/146—Aligning or centring of the image pick-up or image-field
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method and a device for detecting a rotary braille square based on block detection pretreatment, wherein the method comprises the following steps: s110: acquiring a braille picture to be detected; s120: acquiring all Braille block information in the Braille picture to be detected based on an M1 network obtained through Rotated-RETINANET network training; s130: cutting all the braille blocks from the braille pictures to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks; s140: rotating the Braille block picture to the 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: and mapping the detected braille square back to the braille picture to be detected based on the braille square information, and integrating all the braille square to obtain a final detection result. The method improves the existing braille detection method, introduces braille block detection pretreatment, solves the problem that the detection effect of the rotating braille is limited in the rotation target detection, and greatly improves the detection effect of the rotating braille in the handwriting braille scene.
Description
Technical Field
The invention relates to the technical field of braille square detection, in particular to a method and a device for detecting a rotary braille square based on block detection pretreatment.
Background
The open eye person does not have special training, and the reading braille is low in efficiency, and the handwriting braille can inevitably be wrong, so that the efficiency of the open eye person for reading braille is further influenced. To convert braille to a chinese document that is legible to an open eye or to process it into speech, it is often necessary to detect braille parties in braille using a target detection technique. The conventional technology directly applies the target detection technology to braille square detection. The braille written by means of the braille writing tool has a rotating braille square, and the target detection technology cannot effectively detect the rotating braille square. Because the characteristics of the rotated braille square are complex, the braille square with a smaller rotation angle and the braille square with a larger rotation angle are easy to be confused, and therefore the detection effect of the rotation target detection technology is limited.
The above problems are currently in need of solution.
Disclosure of Invention
The present invention has been made to overcome at least one of the above-mentioned drawbacks of the prior art, and in one aspect, provides a method for detecting a rotating braille square based on block detection preprocessing, the method comprising: s110: acquiring a braille picture to be detected; s120: acquiring all Braille block information in the Braille picture to be detected based on an M1 network obtained through Rotated-RETINANET network training; s130: cutting all the braille blocks from the braille pictures to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks; s140: rotating the Braille block picture to the 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: and mapping the detected braille square back to the braille picture to be detected based on the braille square information, and integrating all the braille square to obtain a final detection result.
Furthermore, the M1 network is obtained by training through Rotated-RETINANET networks in combination with braille pictures and corresponding braille block labeling information.
Further, the obtaining all the braille block information in the braille pictures to be detected based on the M1 network obtained through Rotated-RETINANET network training comprises: preprocessing the braille pictures to be detected to adjust the format and the size of the braille pictures to be detected; inputting the preprocessed braille pictures to be detected into an M1 network to obtain braille block information in the braille pictures to be detected; the braille block information includes a center coordinate of the braille block, a width of the braille block, a height of the braille block, and a rotation angle of the braille block.
Further, the step of clipping all the braille blocks from the braille pictures to be detected based on the braille block information to obtain corresponding braille block pictures includes: the braille blocks are clipped from the braille pictures to be detected based on the cv2.getrotationmatrix2d function, the cv2.warp affine function and the cv2.bitwise and function.
Further, the rotating the braille block chips to the horizontal direction includes: and sequentially rotating the cut braille block picture to the horizontal direction based on the cv2.getrotationmatrix2D function and the cv2.warp Affine function by taking the center coordinates of the braille blocks as rotation centers.
Further, the rotating the braille block chips to the horizontal direction includes: and sequentially rotating the braille block pictures by taking the central coordinates of the braille blocks as the rotation centers and rotating the braille blocks at the rotation angles so that the braille block pictures are rotated to the horizontal direction.
Furthermore, the M2 network is obtained by training through Rotated-RETINANET network in combination with braille pictures and corresponding braille side labeling information.
Further, the detecting the braille square information existing in all the braille blocks based on the M2 network trained through Rotated-RETINANET network includes: preprocessing the braille block picture; inputting the preprocessed braille block image into an M2 network to obtain braille party information existing in the braille block; the Braille information comprises the center coordinates of the Braille, the width of the Braille, the height of the Braille, the rotation angle of the Braille and the Braille category.
Further, the step of mapping the detected braille directions back to the braille pictures to be detected based on the braille directions information and integrating all the braille directions to obtain a final detection result includes: sequentially mapping braille squares of all the braille blocks back to braille pictures to be detected; rotating the braille square by a corresponding angle in the direction opposite to the step S140; and combining all the rotated braille squares to generate and obtain a final detection result.
In a second aspect, the present invention provides a rotary braille square detection device based on block detection preprocessing, the device comprising: the method comprises the steps of obtaining a braille picture unit to be detected, and being applicable to obtaining braille pictures to be detected; the method comprises the steps of obtaining a Braille block information unit, wherein the Braille block information unit is suitable for obtaining all Braille block information in a Braille picture to be detected based on an M1 network obtained through Rotated-RETINANET network training; the braille block picture unit is used for cutting all braille blocks from the braille picture to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks; the rotating unit is suitable for rotating the braille block picture to the horizontal direction; the braille square information detection unit is suitable for detecting braille square information existing in all braille block pictures based on an M2 network obtained through Rotated-RETINANET network training; and a detection result generation unit which is suitable for mapping the detected braille square back to the braille picture to be detected based on the braille square information and integrating all the braille square to obtain a final detection result.
In a third aspect, the present invention further provides a computer readable storage medium, where one or more instructions are stored, where the computer instructions are configured to cause the computer to perform the above-described method for detecting a rotating braille square based on block detection preprocessing.
In yet another aspect, the present invention provides an electronic device, including: a memory and a processor; at least one program instruction is stored in the memory; the processor loads and executes the at least one program instruction to implement the above-mentioned method for detecting a rotating braille square based on the block detection preprocessing.
The beneficial effects of the invention are as follows: the invention provides a method for detecting a rotary braille square based on block detection pretreatment, which is characterized by comprising the following steps: s110: acquiring a braille picture to be detected; s120: acquiring all Braille block information in the Braille picture to be detected based on an M1 network obtained through Rotated-RETINANET network training; s130: cutting all the braille blocks from the braille pictures to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks; s140: rotating the Braille block picture to the 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: and mapping the detected braille square back to the braille picture to be detected based on the braille square information, and integrating all the braille square to obtain a final detection result. According to the method, the rotating braille sides in the braille pictures are detected through the block detection pretreatment method, the existing braille sides detection method is improved, the braille blocks are introduced to be detected and pretreated, the problem that the rotating braille sides are limited in detection effect due to rotation target detection is solved, and the detection effect of the rotating braille sides in a handwriting braille scene is greatly improved.
Drawings
The invention is further described below with reference to the drawings and examples.
Fig. 1 is a flowchart of a method for detecting a rotating braille square based on block detection preprocessing according to an embodiment of the invention.
Fig. 2 is a schematic diagram of a braille square and a braille block included in a braille picture to be detected according to an embodiment of the invention.
Fig. 3 is a schematic structural diagram of a rotary braille square detecting device based on block detection preprocessing according to an embodiment of the invention.
Fig. 4 is a partial block diagram of an electronic device provided by an embodiment of the invention.
Detailed Description
Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart depicts operations as a sequential process, many of the operations can be performed in parallel, concurrently, or at the same time. Furthermore, the order of the operations may be rearranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figures. The processes may correspond to methods, functions, procedures, subroutines, and the like.
It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
The present invention will now be described in detail with reference to the accompanying drawings. The figure is a simplified schematic diagram illustrating the basic structure of the invention only by way of illustration, and therefore it shows only the constitution related to the invention.
Example 1
As shown in fig. 1, a flow chart of a method for detecting a rotating braille square based on block detection preprocessing is provided.
As an example, in connection with fig. 2, the method comprises:
S110: and obtaining the braille pictures to be detected.
S120: and acquiring all the braille block information in the braille pictures to be detected based on the M1 network obtained through Rotated-RETINANET network training.
Preferably, the M1 network is obtained by training through Rotated-RETINANET network in combination with braille pictures and corresponding braille block labeling information. Specifically, the M1 network is obtained by training a Rotated-RETINANET network (the network is a rotation target detection version of RETINANET, and an angle of a prediction frame is increased based on horizontal detection) in combination with a braille picture and corresponding braille block labeling information (including positions of braille blocks) and is used for detecting braille blocks existing in the picture.
Preferably, the obtaining all the braille block information in the braille pictures to be detected based on the M1 network obtained through Rotated-RETINANET network training includes: preprocessing the braille pictures to be detected to adjust the format and the size of the braille pictures to be detected; inputting the preprocessed braille pictures to be detected into an M1 network to obtain braille block information in the braille pictures to be detected; the braille block information includes a center coordinate of the braille block, a width of the braille block, a height of the braille block, and a rotation angle of the braille block. Specifically, an input braille picture to be detected is adjusted to be in an input and format of the same size, and then all braille block information { b1, b2, …, bk } contained in the picture is acquired through an M1 network, wherein each braille block bi consists 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.
S130: and cutting all the braille blocks from the braille pictures to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks.
Preferably, the step of clipping all the braille blocks from the braille pictures to be detected based on the braille block information to obtain corresponding braille block pictures includes: the braille blocks are clipped from the braille pictures to be detected based on the cv2.getrotationmatrix2d function, the cv2.warp affine function and the cv2.bitwise and function.
S140: and rotating the Braille block picture to the horizontal direction.
Preferably, the rotating the braille block pattern to the horizontal direction includes: and sequentially rotating the cut braille block picture to the horizontal direction based on the cv2.getrotationmatrix2D function and the cv2.warp Affine function by taking the center coordinates of the braille blocks as rotation centers. Specifically, all the braille blocks are cut out from the original braille pictures to be detected according to all the detected braille blocks { b1, b2, …, bk }, and then the cut braille blocks are rotated to the horizontal direction by taking the central coordinates (xi, yi) as the rotation centers according to the rotation angles anglei of the braille blocks.
S150: and detecting braille square information existing in all braille block pictures based on the M2 network obtained through Rotated-RETINANET network training.
Preferably, the M2 network is obtained by training through Rotated-RETINANET network in combination with braille pictures and corresponding braille side labeling information. Specifically, the M2 network is obtained by training through Rotated-RETINANET network in combination with the braille pictures and corresponding braille party labeling information (including positions and categories of braille parties) and is used for detecting braille parties existing in the pictures.
Preferably, the detecting the braille square information existing in all the braille blocks based on the M2 network trained through Rotated-RETINANET network includes: preprocessing the braille block picture; inputting the preprocessed braille block image into an M2 network to obtain braille party information existing in the braille block; the Braille information comprises the center coordinates of the Braille, the width of the Braille, the height of the Braille, the rotation angle of the Braille and the Braille category. Specifically, the braille block pictures are preprocessed to unify the input size and format, then the pictures are input to Rotated-RETINANET detection network M2, all braille sides { br1, br2, …, brm } contained in the braille block pictures are obtained, each braille side brj is composed of { (xj, yj), wj, hj, anglej, clsi } (xj, yj) represents the center coordinates of the braille side, the width of wj at No. four braille sides, hj represents the height of the braille side, anglej represents the rotation angle of the braille side, clsi represents the category of the braille side, and the category of the braille side removes the total of 63 blank braille sides.
S160: and mapping the detected braille square back to the braille picture to be detected based on the braille square information, and integrating all the braille square to obtain a final detection result.
Preferably, the step of mapping the detected braille directions back to the braille pictures to be detected based on the braille directions information and integrating all the braille directions to obtain a final detection result includes: sequentially mapping braille squares of all the braille blocks back to braille pictures to be detected; rotating the braille square by a corresponding angle in the direction opposite to the step S140; and combining all the rotated braille squares to generate and obtain a final detection result. Specifically, for the braille square { br1, br2, …, brm }, according to the braille blocks bi to which each braille square brj belongs, the braille square is mapped to the position on the original picture according to anglei by taking the center coordinates of the braille blocks as rotation points, and then all the braille squares in the braille blocks are combined to obtain the final braille square detection result.
The above embodiment provides a method for detecting a rotating braille square based on block detection pretreatment, wherein the method adds the block detection pretreatment to detect the braille square in the braille picture based on the traditional braille square detection method, converts all the rotating braille square detection into small-angle rotating braille square detection, effectively avoids confusion of small-angle and large-angle rotating braille square characteristics, and obviously improves the rotating braille square detection effect in a handwriting braille scene.
Example 2
Referring to fig. 3, the present embodiment provides a schematic structural diagram of a rotary braille square detecting device based on block detection preprocessing.
As an example, the apparatus includes:
the braille picture to be detected unit 310 is adapted to obtain the braille picture to be detected.
The braille block information obtaining unit 320 is suitable for obtaining all the braille block information in the braille pictures to be detected based on the M1 network obtained through Rotated-RETINANET network training.
And the braille block image obtaining unit 330 is suitable for clipping all braille blocks from the braille images to be detected based on the braille block information to obtain braille block images corresponding to the braille blocks.
And a rotation unit 340 adapted to rotate the braille block pattern to a horizontal direction.
The braille side information detecting unit 350 is adapted to detect braille side information present in all braille block pictures based on the M2 network trained through Rotated-RETINANET network.
The detection result generating unit 360 is adapted to map the detected braille square back to the braille picture to be detected based on the braille square information, and integrate all the braille square to obtain a final detection result.
It is to be noted that this embodiment is a system example corresponding to the first embodiment, and can be implemented in cooperation with the first embodiment. The related technical details mentioned in the first embodiment are still valid in this embodiment, and in order to reduce repetition, a detailed description is omitted here. Accordingly, the related art details mentioned in the present embodiment can also be applied to the first embodiment.
It should be noted that each unit referred to in this embodiment is a logic unit, and in practical application, one logic unit may be one physical unit, or may be a part of one physical unit, or may be implemented by a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, units that are not so close to solving the technical problem presented by the present invention are not introduced in the present embodiment, but this does not indicate that other units are not present in the present embodiment.
Example 3
The embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a rotating braille detection method based on block detection pretreatment, and the rotating braille detection method based on the block detection pretreatment is realized when the rotating braille detection program based on the block detection pretreatment is executed by a processor. Because the storage medium adopts all the technical schemes of all the embodiments, the storage medium has at least all the beneficial effects brought by the technical schemes of the embodiments, and the description is omitted here.
Example 4
Referring to fig. 4, an embodiment of the present invention further provides an electronic device, including: a memory and a processor; at least one program instruction is stored in the memory; the processor implements the method for detecting a rotating braille square based on the block detection preprocessing provided in embodiment 1 by loading and executing the at least one program instruction.
The memory 602 and the processor 601 are connected by a bus, which may include any number of interconnected buses and bridges, which connect together various circuits of the one or more processors 601 and the memory 602. The bus may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or may be a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor 601 is transmitted over a wireless medium via an antenna, which further receives the data and transmits the data to the processor 601.
The processor 601 is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And memory 602 may be used to store data used by processor 601 in performing operations.
The foregoing is merely an embodiment of the present application, and a specific structure and characteristics of common knowledge in the art, which are well known in the scheme, are not described herein, so that a person of ordinary skill in the art knows all the prior art in the application date or before the priority date, can know all the prior art in the field, and has the capability of applying the conventional experimental means before the date, and a person of ordinary skill in the art can complete and implement the present embodiment in combination with his own capability in the light of the present application, and some typical known structures or known methods should not be an obstacle for a person of ordinary skill in the art to implement the present application. It should be noted that modifications and improvements can be made by those skilled in the art without departing from the structure of the present application, and these should also be considered as the scope of the present application, which does not affect the effect of the implementation of the present application and the utility of the patent. The protection scope of the present application is subject to the content of the claims, and the description of the specific embodiments and the like in the specification can be used for explaining the content of the claims.
Claims (8)
1.A method for detecting a rotating braille square based on block detection preprocessing, the method comprising:
s110: acquiring a braille picture to be detected;
s120: acquiring all Braille block information in the Braille picture to be detected based on an M1 network obtained through Rotated-RETINANET network training;
S130: cutting all the braille blocks from the braille pictures to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks;
s140: rotating the Braille block picture to the horizontal direction;
S150: detecting braille square information existing in all braille block pictures based on an M2 network obtained through Rotated-RETINANET network training, wherein the method comprises the following steps: preprocessing the braille block picture; inputting the preprocessed braille block image into an M2 network to obtain braille party information existing in the braille block; the Braille information comprises a central coordinate of the Braille, a width of the Braille, a height of the Braille, a rotation angle of the Braille and a Braille category;
S160: mapping the detected braille square back to the braille picture to be detected based on the braille square information, and integrating all the braille square to obtain a final detection result, wherein the method comprises the following steps: sequentially mapping braille squares of all the braille blocks back to braille pictures to be detected; rotating the braille square by a corresponding angle in the direction opposite to the step S140; combining all the rotated braille squares to generate and obtain a final detection result, including: for the braille square { br1, br2, …, brm }, according to the braille blocks bi to which each braille square brj belongs, the braille square is mapped to the position on the original picture according to the rotation angle of the braille square by taking the center coordinates of the braille blocks as rotation points, and the braille squares in all the braille blocks are combined to obtain the final braille square detection result.
2. The method for detecting the rotary braille square based on the block detection preprocessing according to claim 1, wherein the M1 network is obtained by training Rotated-RETINANET networks in combination with braille pictures and corresponding braille block labeling information.
3. The method for detecting a rotating braille square based on block detection preprocessing according to claim 2, wherein the obtaining all braille block information in the braille pictures to be detected based on the M1 network obtained through Rotated-RETINANET network training comprises:
Preprocessing the braille pictures to be detected to adjust the format and the size of the braille pictures to be detected;
Inputting the preprocessed braille pictures to be detected into an M1 network to obtain braille block information in the braille pictures to be detected;
the braille block information includes a center coordinate of the braille block, a width of the braille block, a height of the braille block, and a rotation angle of the braille block.
4. The method for detecting a rotating braille square based on block detection preprocessing according to claim 1, wherein the step of clipping all braille blocks from the braille pictures to be detected based on the braille block information to obtain corresponding braille block pictures includes:
The braille blocks are clipped from the braille pictures to be detected based on the cv2.getrotationmatrix2d function, the cv2.warp affine function and the cv2.bitwise and function.
5. The method for detecting a rotating braille square based on block detection pretreatment according to claim 1, characterized in that the rotating the braille block chips to a horizontal direction comprises:
And sequentially rotating the cut braille block picture to the horizontal direction based on the cv2.getrotationmatrix2D function and the cv2.warp Affine function by taking the center coordinates of the braille blocks as rotation centers.
6. The method for detecting a rotating braille lettering square based on the block detection preprocessing according to claim 5, characterized in that the rotating the braille block drawing to a horizontal direction comprises:
and sequentially rotating the braille block pictures by taking the central coordinates of the braille blocks as the rotation centers and rotating the braille blocks at the rotation angles so that the braille block pictures are rotated to the horizontal direction.
7. The method for detecting the rotary braille square based on the block detection pretreatment according to claim 1, wherein the M2 network is obtained by training Rotated-RETINANET networks in combination with braille pictures and corresponding braille square labeling information.
8. A rotating braille square detection device based on block detection preprocessing, characterized in that the device comprises:
the method comprises the steps of obtaining a braille picture unit to be detected, and being applicable to obtaining braille pictures to be detected;
the method comprises the steps of obtaining a Braille block information unit, wherein the Braille block information unit is suitable for obtaining all Braille block information in a Braille picture to be detected based on an M1 network obtained through Rotated-RETINANET network training;
The braille block picture unit is used for cutting all braille blocks from the braille picture to be detected based on the braille block information to obtain braille block pictures corresponding to the braille blocks;
The rotating unit is suitable for rotating the braille block picture to the horizontal direction;
The braille square information detection unit is suitable for detecting braille square information existing in all braille block pictures based on an M2 network obtained through Rotated-RETINANET network training, and comprises the following steps: preprocessing the braille block picture; inputting the preprocessed braille block image into an M2 network to obtain braille party information existing in the braille block; the Braille information comprises a central coordinate of the Braille, a width of the Braille, a height of the Braille, a rotation angle of the Braille and a Braille category;
the generating a detection result unit, which is suitable for mapping the detected braille square back to the braille picture to be detected based on the braille square information, and integrating all the braille square to obtain a final detection result, comprises the following steps: sequentially mapping braille squares of all the braille blocks back to braille pictures to be detected; rotating the braille square by a corresponding angle in the direction opposite to the step S140; combining all the rotated braille squares to generate and obtain a final detection result, including: for the braille square { br1, br2, …, brm }, according to the braille blocks bi to which each braille square brj belongs, the braille square is mapped to the position on the original picture according to the rotation angle of the braille square by taking the center coordinates of the braille blocks as rotation points, and the braille squares in all the braille blocks are combined to obtain the final braille square detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410529923.XA CN118429978B (en) | 2024-04-29 | 2024-04-29 | Rotary braille square detection method and device based on block detection pretreatment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410529923.XA CN118429978B (en) | 2024-04-29 | 2024-04-29 | Rotary braille square detection method and device based on block detection pretreatment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118429978A CN118429978A (en) | 2024-08-02 |
| CN118429978B true CN118429978B (en) | 2024-11-15 |
Family
ID=92336772
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410529923.XA Active CN118429978B (en) | 2024-04-29 | 2024-04-29 | Rotary braille square detection method and device based on block detection pretreatment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118429978B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052936A (en) * | 2017-11-03 | 2018-05-18 | 中国科学院计算技术研究所 | A kind of braille image wing drop bearing calibration and system |
| CN108052955A (en) * | 2017-11-03 | 2018-05-18 | 中国科学院计算技术研究所 | A kind of high-precision braille recognition methods and system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2019012929A (en) * | 2017-06-30 | 2019-01-24 | 京セラドキュメントソリューションズ株式会社 | Image forming apparatus |
| JP7031928B2 (en) * | 2018-03-12 | 2022-03-08 | Ｎｅｃソリューションイノベータ株式会社 | Braille block judgment support device, judgment support method, and program |
| CN114565926A (en) * | 2022-03-04 | 2022-05-31 | 浙江大学 | Two-stage Braille detection and identification method based on target detection |
| CN116612490B (en) * | 2023-06-02 | 2025-05-27 | 兰州大学 | Braille image recognition method based on target detection |
| CN116863471A (en) * | 2023-06-21 | 2023-10-10 | 山东浪潮科学研究院有限公司 | A method to improve the accuracy of text recognition of multiple types of documents |
-
2024
- 2024-04-29 CN CN202410529923.XA patent/CN118429978B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108052936A (en) * | 2017-11-03 | 2018-05-18 | 中国科学院计算技术研究所 | A kind of braille image wing drop bearing calibration and system |
| CN108052955A (en) * | 2017-11-03 | 2018-05-18 | 中国科学院计算技术研究所 | A kind of high-precision braille recognition methods and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118429978A (en) | 2024-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR20210152371A (en) | Method and device for image generation, apparatus, and medium | |
| WO2019011249A1 (en) | Method, apparatus, and device for determining pose of object in image, and storage medium | |
| CN110210328A (en) | The method, apparatus and electronic equipment of object are marked in image sequence | |
| CN111832445B (en) | Positioning method and related device for engineering drawing | |
| US11132586B2 (en) | Rolling shutter rectification in images/videos using convolutional neural networks with applications to SFM/SLAM with rolling shutter images/videos | |
| CN110599455A (en) | Display screen defect detection network model, method and device, electronic equipment and storage medium | |
| US20210248808A1 (en) | Posture prediction method, computer device and storage medium | |
| US11055818B2 (en) | Panorama image alignment method and device based on multipath images, and portable terminal | |
| CN115860067B (en) | Method, device, computer equipment and storage medium for generating countermeasure network training | |
| CN119863678B (en) | A method for constructing scene-to-image cross-domain mapping sample sets driven by attribute features | |
| CN109002776A (en) | Face identification method, system, computer equipment and computer readable storage medium | |
| US20220351495A1 (en) | Method for matching image feature point, electronic device and storage medium | |
| CN118429978B (en) | Rotary braille square detection method and device based on block detection pretreatment | |
| CN114387294B (en) | Image processing method and storage medium | |
| CN118429979B (en) | Method and device for detecting smeared braille square based on block detection pretreatment | |
| CN118333909B (en) | A multi-eye image acquisition and preprocessing system and method for sea surface scenes | |
| CN110688511A (en) | Fine-grained image retrieval method and device, computer equipment and storage medium | |
| CN112614210B (en) | Engineering drawing display method, system and related device | |
| CN111054072A (en) | Role model trailing method, device, equipment and storage medium | |
| CN117333937A (en) | Human body posture estimation method and device based on classification and distillation and electronic equipment | |
| CN116645473A (en) | A 3D human body reconstruction method for multi-sensor multi-view data fusion | |
| CN113763513B (en) | Interactive marking method for target object in image | |
| CN111161174A (en) | Blood vessel model establishing method and device, computer equipment and storage medium | |
| CN116152055B (en) | Pixel-level infrared image stitching method based on deep learning | |
| CN110610135A (en) | Display method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |