A braille scanning pen and a braille scanning method

By combining a color camera and a fingerprint sensor, combined with gyroscopes and pressure sensors for deviation correction, the problem of existing scanning pens being unable to recognize Braille is solved, and high-accuracy Braille recognition is achieved.

CN115171113BActive Publication Date: 2025-10-10IFLYTEK CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210787515.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-04
Publication Date
2025-10-10
Estimated Expiration
2042-07-04

AI Technical Summary

Technical Problem

Existing scanning pens cannot effectively recognize Braille, and fingerprint sensors have low recognition accuracy when scanning quickly.

Method used

A color camera and fingerprint sensor are combined to collect images through the color camera and detect pressure signals through the fingerprint sensor. The gyroscope and pressure sensor are used to correct the deviation and improve the accuracy of Braille recognition.

Benefits of technology

It improves the accuracy of Braille scanning and recognition, and ensures the correctness of horizontal scanning, making it suitable for Braille recognition of large texts.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115171113B_ABST
    Figure CN115171113B_ABST
Patent Text Reader

Abstract

The application discloses a Braille scanning pen and a Braille scanning method, and the Braille scanning pen comprises a body, a color camera, a fingerprint sensor and a main control module arranged in the body; a lens of the color camera is arranged on one end face of the body, the fingerprint sensor is fixed on a first side face of the body close to the end face, and an end face of a detection head of the fingerprint sensor protrudes from the end face of the body; the main control module is in signal connection with the color camera and the fingerprint sensor respectively, and the main control module is used for obtaining a Braille recognition result according to an image obtained by the color camera and a detection result of the fingerprint sensor. The color camera and the fingerprint sensor are simultaneously used for scanning Braille, the recognition results of the two are complementary to each other, and the accuracy of Braille scanning and recognition is improved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Braille recognition, and in particular to a Braille scanning pen and a Braille scanning method. BACKGROUND

[0002] Most of the current scanning pens are used for scanning ordinary characters with white background and black font. The article can be scanned by turning on the black and white camera, and the corresponding characters can be read out or translated through AR recognition. However, since the color of the Braille characters and the background is relatively close, the existing scanning pen cannot scan and recognize Braille.

[0003] Most of the existing Braille scanning pens use fingerprint sensors to recognize Braille. However, the recognition accuracy of the fingerprint sensor is directly affected by the scanning speed. When the scanning speed is fast, the recognition accuracy is low, and therefore the quality of Braille scanning and recognition cannot be guaranteed. SUMMARY

[0004] In view of the above, the present application aims to provide a Braille scanning pen and a Braille scanning method, which simultaneously use a color camera and a fingerprint sensor to scan Braille. The recognition results of the two are complementary to each other, and the accuracy of Braille scanning and recognition is improved.

[0005] The technical solution adopted by the present application is as follows:

[0006] In a first aspect, the present application provides a Braille scanning pen, comprising a body, a color camera, a fingerprint sensor, and a main control module arranged inside the body.

[0007] The lens of the color camera is arranged on one end face of the body, the fingerprint sensor is fixed on the first side face of the body close to the end face, and the end face of the detection head of the fingerprint sensor protrudes from the end face of the body.

[0008] The main control module is signal connected with the color camera and the fingerprint sensor, and is used to obtain a Braille recognition result according to the image obtained by the color camera and the detection result of the fingerprint sensor.

[0009] In one possible implementation, the Braille scanning pen further comprises a gyroscope, which is signal connected with the main control module.

[0010] In one possible implementation, the Braille scanning pen further comprises two pressure sensors, which are arranged on opposite sides of the lens, the end face of the detection head of the fingerprint sensor protrudes from the end face of the pressure sensor, and the pressure sensor is signal connected with the main control module.

[0011] During a normal horizontal scanning process, the two pressure sensors are located on both sides of the Braille in the longitudinal direction, there is a first preset distance between the end of the pressure sensor and the Braille, the detection head of the fingerprint sensor is in contact with the Braille, and there is a second preset distance between the lens of the color camera and the Braille.

[0012] In a second aspect, the present invention further provides a Braille scanning method, comprising:

[0013] receiving an image obtained by a color camera and a detection result of a fingerprint sensor during a horizontal scanning process;

[0014] Analyze and process the image and detection results to obtain Braille recognition results;

[0015] Output Braille recognition results.

[0016] In one possible implementation manner, the method further includes performing a deflection correction operation during the horizontal scanning process.

[0017] In one possible implementation, the correction operation includes:

[0018] receiving a longitudinal displacement signal from a gyroscope and / or a pressure detection signal from a pressure sensor during a transverse scanning process;

[0019] determining whether there is a deviation during the transverse scanning process based on the longitudinal displacement signal and / or the pressure detection signal;

[0020] If there is an offset, an offset signal is output.

[0021] One possible implementation also includes:

[0022] Scanning speed monitoring is performed during the lateral scanning process.

[0023] In one possible implementation, scanning speed monitoring includes:

[0024] receiving a lateral displacement signal of the gyroscope during a lateral scanning process;

[0025] Determine whether the lateral displacement velocity or lateral displacement acceleration is greater than a threshold;

[0026] If so, an overspeed signal is output.

[0027] In one possible implementation, the image and the detection result are analyzed and processed to obtain a Braille recognition result, specifically including:

[0028] Complementarily fusing the first salient point data obtained based on the image and the second salient point data obtained based on the detection result to obtain fused salient point data;

[0029] The Braille database is called to determine the Braille corresponding to the fused salient point data as the Braille recognition result.

[0030] In one possible implementation, after determining the Braille corresponding to the fused salient point data, the method further includes determining the horizontal coordinates and vertical coordinates of the Braille corresponding to the salient point data based on the horizontal displacement signal and the vertical displacement signal of the gyroscope, and using the Braille and its corresponding coordinates as the Braille recognition result.

[0031] In one possible implementation, the first salient point data obtained based on the image specifically includes:

[0032] Preprocessing the image to obtain a preprocessed image;

[0033] Extract salient points from the preprocessed image as first salient point data.

[0034] In one possible implementation, extracting salient points from the preprocessed image includes:

[0035] Selecting a circular structure element with a preset radius to subdivide the preprocessed image to obtain at least one subdivision area, each subdivision area including one or more Braille characters;

[0036] For each subdivided area, the subdivided area is locally segmented using the large law threshold segmentation to obtain at least one local segmented area, each local segmented area includes a salient point;

[0037] According to the linear characteristics of Braille arrangement, isolated points and miscellaneous points in each local segmentation area are removed to obtain salient points in the local segmentation area;

[0038] The set of salient points in all segmented areas is taken as the salient points in the preprocessed image.

[0039] In one possible implementation, preprocessing the image to obtain a preprocessed image specifically includes:

[0040] Performing contrast-limited adaptive histogram equalization on the image to obtain an equalized image;

[0041] The bilinear interpolation method is used to combine adjacent regions in the equalized image to eliminate artificial boundaries and obtain the preprocessed image.

[0042] In one possible implementation manner, the first salient point data is obtained from the image using an augmented reality recognition algorithm.

[0043] The concept of the present invention is to fuse the convex point data obtained from images captured by a color camera with the convex point data obtained from pressure signals collected by a fingerprint sensor, making the two complementary and improving the accuracy of Braille scanning and recognition. Furthermore, the present invention uses a gyroscope and / or a pressure sensor to perform deflection correction during the lateral scanning process to ensure the accuracy of lateral scanning. Furthermore, the present invention determines the coordinates of Braille using the displacement signals detected by the gyroscope, thereby enabling positional recognition of multiple lines of Braille and providing a good foundation for Braille recognition of large text volumes. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:

[0045] Figure 1 A schematic structural diagram of the Braille scanning pen provided by the present invention;

[0046] Figure 2 A structural diagram of the Braille scanning pen provided by the present invention;

[0047] Figure 3 This is a flow chart of the Braille scanning method provided by the present invention. DETAILED DESCRIPTION

[0048] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0049] The concept of the present invention is to fuse the convex point data obtained from images captured by a color camera with the convex point data obtained from pressure signals collected by a fingerprint sensor, making the two complementary and improving the accuracy of Braille scanning and recognition. Furthermore, the present invention uses a gyroscope and / or a pressure sensor to perform deflection correction during the lateral scanning process to ensure the accuracy of lateral scanning. Furthermore, the present invention determines the coordinates of Braille using the displacement signals detected by the gyroscope, thereby enabling positional recognition of multiple lines of Braille and providing a good foundation for Braille recognition of large text volumes.

[0050] In view of the aforementioned core concept, the present invention provides at least one embodiment of a Braille scanning pen, such as Figure 1 and 2 As shown, the Braille scanner 100 includes a body 180 , a color camera 150 , a fingerprint sensor 140 , a main control module 110 disposed inside the body 180 , and a speaker 160 .

[0051] A color camera is a camera that uses the superposition of multiple color channels to obtain the image of the target object. Unlike a black and white camera, a color camera can better identify the foreground features in the image when the foreground and background colors are similar.

[0052] In one possible implementation, the color camera is an RGB camera, which is a camera that obtains an image of a target object by changing the three color channels of red (R), green (G), and blue (B) and superimposing them in different proportions to achieve different colors.

[0053] It can be understood that the color camera can be other color camera devices in the prior art.

[0054] The main body of the color camera 150 is fixed inside the body 180 , and the lens of the color camera 150 is embedded in one end surface of the body 180 . Figure 2 In the illustrated implementation, the lens of the color camera 150 is embedded in the lower end surface 1801 of the body 180 .

[0055] The fingerprint sensor 140 is fixed on the first side surface of the main body 180 close to the lower end surface 1801. The end surface of the detection head of the fingerprint sensor 140 protrudes from the lower end surface 1801 of the main body 180, so that when scanning Braille, the fingerprint sensor 140 abuts against the Braille while the lens of the color camera has a second preset distance from the Braille, which can enable the color camera to capture the image of the Braille.

[0056] The main control module 110 is respectively connected to the color camera 150 (through the MIPI interface or SPI interface) and the fingerprint sensor 140 (I2C1 bus or SPI interface) signals. The main control module 110 receives the image obtained by the color camera 150 and the detection result of the fingerprint sensor 140, and obtains the Braille recognition result based on the image and the detection result.

[0057] The speaker 160 is provided on the main body 180 and is connected to the main control module 110 via an I2S bus signal, and is used to play the Braille recognition result and the alarm information of the main control module 110 .

[0058] It is understood that a USB interface can also be provided on the body 180 for connecting to an external headset 200 so that the user can receive the Braille recognition result. It is understood that the Braille scanner pen is also provided with a power module and a power switch.

[0059] In a preferred implementation, the Braille scanner 100 further includes a gyroscope 120, which is signal-connected to the main control module 110 via an I2C3 bus. The gyroscope 120 is configured to detect the Braille scanner's lateral and longitudinal displacement signals during a lateral scanning process. The main control module 110 determines the Braille scanner's horizontal and vertical coordinates based on the lateral and longitudinal displacement signals, and obtains the Braille scanner's lateral displacement velocity and acceleration, as well as the Braille scanner's longitudinal displacement velocity and acceleration.

[0060] During normal horizontal scanning, the vertical coordinate of the Braille scanner does not change, that is, there is no vertical displacement. If the Braille scanner deviates, the vertical displacement is not zero, and the main control module 110 determines that the scanning has deviated and outputs a deviation signal to the user through the speaker 160 or the earphone 200.

[0061] In a preferred implementation, the Braille scanner 100 further includes two pressure sensors 130, which are respectively arranged on opposite sides of the lens of the color camera. Figure 2 In the illustrated implementation, the two pressure sensors 130 are respectively disposed on extensions of two opposing side surfaces of the body 180, both of which are perpendicular to the first side surface 1802. The end surface of the detection head of the fingerprint sensor 140 protrudes from the lower end surface 1801 of the body 180. During a normal horizontal scanning process, the detection head of the fingerprint sensor 140 abuts the Braille text, while the two pressure sensors 130 are located on either side of the Braille text in the longitudinal direction (without contact). A first predetermined distance is maintained between the ends of the pressure sensors 130 and the Braille text, which is less than or equal to the height of the raised dots of the Braille text.

[0062] It should be noted that on the plane formed by the braille board, braille is written horizontally to form rows, and multiple rows are arranged vertically. That is, the horizontal direction is the left-right direction of the braille, and the vertical direction is the vertical direction. The vertical sides of the braille are the outer sides of the upper end and the outer sides of the lower end of the braille. The distance between the two pressure sensors 130 and the upper and lower ends of the braille is less than the distance between the two braille rows. Therefore, during normal horizontal scanning, the pressure sensors do not contact the braille.

[0063] If the Braille scanner stylus deflects significantly during horizontal scanning, pressure sensor 130 will contact the Braille text and generate a pressure detection signal. Pressure sensor 130 is connected to main control module 110 via the I2C2 bus. If main control module 110 detects the pressure detection signal from pressure sensor 130, indicating scanning deviation, it outputs the deviation signal to the user via speaker 160 or earphones 200.

[0064] Based on the above-mentioned Braille scanning pen, the present invention also provides a Braille scanning method applied to the main control module.Figure 3 As shown, the Braille scanning method includes:

[0065] S310: Receive an image obtained by the color camera and a detection result of the fingerprint sensor during the horizontal scanning process.

[0066] S320: Analyze and process the image and detection results to obtain Braille recognition results.

[0067] It should be noted that during the horizontal scanning process, the color camera and fingerprint sensor transmit images and detection results to the main control module in real time. While scanning horizontally, the main control module analyzes and processes the received data, thereby improving the efficiency of Braille scanning.

[0068] In one possible implementation, analyzing and processing the image and the detection result to obtain the Braille recognition result specifically includes:

[0069] S3201: Complementarily fuse the first salient point data obtained based on the image and the second salient point data obtained based on the detection result to obtain fused salient point data.

[0070] In a possible implementation, the first salient point data obtained based on the image specifically includes:

[0071] P1: Preprocess the image to obtain a preprocessed image.

[0072] In a possible implementation, preprocessing the image to obtain the preprocessed image specifically includes:

[0073] Q1: Perform contrast-constrained adaptive histogram equalization on the image to obtain a balanced image. Specifically, the image is divided into several small regions, and histogram equalization is performed on each small region. The contrast of the grayscale image is adjusted so that the image pixels are evenly distributed across the entire grayscale range of the image, making the image display more balanced.

[0074] Q2: Adjacent regions in the equalized image are combined using bilinear interpolation to eliminate artificial boundaries between adjacent regions, thereby enhancing the image. The interpolated image is used as the preprocessed image.

[0075] In another possible implementation, the first salient point data is obtained from the image using an augmented reality recognition algorithm.

[0076] It is understandable that existing image processing methods can be used to enhance the image.

[0077] P2: Extract salient points from the preprocessed image as the first salient point data.

[0078] In one possible implementation, extracting salient points from the preprocessed image specifically includes:

[0079] R1: Select a circular structuring element with a preset radius to subdivide the preprocessed image, obtaining at least one subdivision region, each containing one or more Braille characters. Image subdivision eliminates image boundary points, shrinks the image boundary inward, removes images smaller than the circular structuring element, and eliminates sharp peaks and convex corners at image edges, as well as fine connections between images.

[0080] In a possible implementation, the preset radius is 1 mm.

[0081] R2: For each subdivision, use the large-law threshold segmentation to locally segment the subdivision to obtain at least one local segmentation region, each of which includes a salient point. Large-law threshold segmentation can minimize the probability of misclassification.

[0082] Specifically, assume an image I(x, y) with an M×N segmented region. Let T be the segmentation threshold for the foreground (target) and background. The ratio of foreground pixels to the total image is W0, and the average grayscale of foreground pixels is v0. The ratio of background pixels to the total image is W1, and the average grayscale of background pixels is v1. The total average grayscale of the image is v, and the between-class variance is g. The number of pixels in the image whose grayscale value is less than the threshold T is N0, and the number of pixels whose grayscale value is greater than the threshold T is N1.

[0083] Then W0=N0 / M×N (1)

[0084] W1=N1 / M×N (2)

[0085] N0+N1=M×N (3)

[0086] W0+W1=1 (4)

[0087] v=W0*v0+W1*v1 (5)

[0088] g=W0×(v0-v) 2 +W1×(v1-v) 2 (6)

[0089] Substituting (5) into (6), we get the formula

[0090] g=W0×W1×(v0-v1) 2 (7)

[0091] Let the grayscale value be t, and then traverse t from the minimum grayscale value to the maximum grayscale value. When the t value maximizes the inter-class variance g, t is the optimal threshold T for segmentation.

[0092] Each sub-region is locally segmented by using the optimal threshold T to obtain a locally segmented region.

[0093] R3: According to the linear feature of the Braille arrangement, remove the isolated points and noise points in each locally segmented region to obtain the convex points in the locally segmented region.

[0094] R4: The set of convex points in all sub-regions is the convex point in the pre-processed image.

[0095] The fingerprint sensor identifies the convex points by detecting the pressure generated by the contact of the head with the convex points of the Braille, forming second convex point data.

[0096] In one possible implementation, the first convex point data and the second convex point data are aligned, and after alignment, there may be overlapping convex points and non-overlapping convex points, and the non-overlapping convex points are convex points that are not detected by the color camera or the fingerprint sensor. The set of all overlapping convex points and non-overlapping convex points is the fused convex point data, thereby complementing the deficiencies of the color camera and the fingerprint sensor, and improving the accuracy of Braille detection.

[0097] In another possible implementation, after the first convex point data and the second convex point data are aligned, for the non-overlapping convex points, the product of the grayscale value of the non-overlapping convex point and the weight of the corresponding color camera or the weight of the fingerprint sensor in the Braille recognition result (the non-overlapping convex point belongs to the first convex point data or the second convex point data) is calculated, and if the product is greater than a threshold, the non-overlapping convex point is taken as the fused convex point data.

[0098] S3202: Call the Braille database to determine the Braille corresponding to the fused convex point data as the Braille recognition result.

[0099] In one possible implementation, the fused convex point data is locally segmented to obtain each part as a first Braille, and the first Braille and the second Braille in the Braille database are compared one by one, and if the arrangement of the convex points of the two is the same, the recognition result of the first Braille is the second Braille.

[0100] It can be understood that the existing convex point recognition technology can be used to obtain the Braille recognition result.

[0101] In a preferred implementation, after determining the Braille corresponding to the fused convex point data, the horizontal displacement signal and the vertical displacement signal of the gyroscope are used to determine the horizontal coordinates and the vertical coordinates of the Braille corresponding to the convex point data (for example, the coordinates of the center of the Braille), and the Braille and its corresponding coordinates are taken as the Braille recognition result, thereby realizing the position recognition of multiple lines of Braille and providing a good foundation for large-scale Braille recognition.

[0102] S330: outputting the Braille recognition result through a loudspeaker or a headphone, and storing the Braille recognition result in a corpus.

[0103] In a preferred implementation, a deviation correction operation is also performed during the lateral scanning.

[0104] In a possible implementation, the master module also receives a longitudinal displacement signal of the gyroscope and / or a pressure detection signal of the pressure sensor during the lateral scanning, and determines whether there is a deviation during the lateral scanning according to the longitudinal displacement signal and / or the pressure detection signal. If there is a deviation, a deviation signal is output to the user through the loudspeaker or the headphone.

[0105] Specifically, the master module calculates a longitudinal displacement speed or acceleration of the Braille scanning pen through the longitudinal displacement signal, and if the longitudinal displacement speed or acceleration is not zero, it indicates that the Braille scanning pen deviates.

[0106] If the master module receives the pressure detection signal of the pressure sensor, it indicates that the pressure sensor touches the Braille, and the Braille scanning pen deviates.

[0107] In another preferred implementation, a scanning speed monitoring is also performed during the lateral scanning.

[0108] In a possible implementation, the master module also receives a lateral displacement signal of the gyroscope during the lateral scanning, and calculates a lateral displacement speed or acceleration according to the lateral displacement signal, and determines whether the lateral displacement speed or acceleration is greater than a threshold. If yes, it indicates that the Braille scanning pen is overspeed, and an overspeed signal is output to the user through the loudspeaker or the headphone.

[0109] In the embodiments of the present application, "at least one" means one or more, and "multiple" means two or more. The "and / or" describes the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Wherein A and B can be singular or plural. The character " / " generally represents an "or" relationship between the front and rear associated objects. "At least one of the following" and the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b, and c can be single or multiple.

[0110] Those skilled in the art can understand that each module, unit and method step described in the embodiments disclosed in the specification can be implemented in electronic hardware, computer software and combination of the electronic hardware and the computer software. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different ways to implement the described functions for each specific application, but the implementation should not be considered beyond the scope of the present application.

[0111] The various embodiments described in the specification can be described in progressive manner, and the same or similar parts between the various embodiments can be referred to each other. In particular, for the device, equipment and other embodiments, since they are basically similar to the method embodiments, the related parts can be referred to the part of the method embodiments. The above described device, equipment and other embodiments are only illustrative, and the modules, units and the like described as separate components can be or can not be physically separated, that is, they can be located in one place, or can be distributed in multiple places, for example, the nodes of system network. Specifically, part or all of the modules, units can be selected to achieve the purpose of the above embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0112] The above embodiments according to the drawings describe the structure, features and effects of the present application, but the above is only the preferred embodiment of the present application, and it should be noted that the technical features involved in the above embodiments and preferred modes can be reasonably combined and matched into various equivalent schemes by those skilled in the art without departing from or changing the design idea and technical effects of the present application. Therefore, the present application is not limited by the drawings, and any change or modification made according to the idea of the present application, or any equivalent embodiment within the scope of the specification and drawings, should be within the protection scope of the present application.

Claims

1. A Braille scanning pen, characterized in that: It includes a body, a color camera, a fingerprint sensor, and a main control module arranged inside the body; The lens of the color camera is arranged on an end surface of the body, the fingerprint sensor is fixed on a first side surface of the body close to the end surface, and the end surface of the detection head of the fingerprint sensor protrudes from the end surface of the body; The main control module is respectively connected to the color camera and the fingerprint sensor signal. The main control module is used to obtain a Braille recognition result based on the image obtained by the color camera and the detection result of the fingerprint sensor, including: complementary fusion of the first salient point data obtained based on the image and the second salient point data obtained based on the detection result to obtain fused salient point data; calling the Braille database to determine the Braille corresponding to the fused salient point data as the Braille recognition result.

2. The Braille scanner pen according to claim 1, characterized in that: The Braille scanning pen further includes a gyroscope, and the gyroscope is signal-connected to the main control module.

3. The Braille scanning pen according to claim 1 or 2, characterized in that: The Braille scanner pen further includes two pressure sensors, which are respectively arranged on opposite sides of the lens. The end surface of the detection head of the fingerprint sensor protrudes from the end surface of the pressure sensor. The pressure sensor is connected to the main control module for signal connection; During a normal horizontal scanning process, the two pressure sensors are respectively located on both sides of the longitudinal direction of the Braille, there is a first preset distance between the end of the pressure sensor and the Braille, the detection head of the fingerprint sensor is in contact with the Braille, and there is a second preset distance between the lens of the color camera and the Braille.

4. A Braille scanning method, characterized in that: include: receiving an image obtained by a color camera and a detection result of a fingerprint sensor during a horizontal scanning process; Analyzing and processing the image and the detection result to obtain a Braille recognition result, including: complementary fusing first salient point data obtained based on the image and second salient point data obtained based on the detection result to obtain fused salient point data; calling a Braille database to determine Braille corresponding to the fused salient point data as the Braille recognition result; The Braille recognition result is output.

5. The Braille scanning method according to claim 4, characterized in that: Also includes: Perform deflection correction during horizontal scanning.

6. The Braille scanning method according to claim 5, characterized in that: The correction operation includes: receiving a longitudinal displacement signal from a gyroscope and / or a pressure detection signal from a pressure sensor during a transverse scanning process; determining whether there is a deviation during the transverse scanning process according to the longitudinal displacement signal and / or the pressure detection signal; If there is an offset, an offset signal is output.

7. The Braille scanning method according to claim 4 or 5, characterized in that: Also includes: Scanning speed monitoring is performed during the lateral scanning process.

8. The Braille scanning method according to claim 7, characterized in that: The scanning speed monitoring includes: receiving a lateral displacement signal of the gyroscope during a lateral scanning process; Calculating the lateral displacement velocity or lateral displacement acceleration according to the lateral displacement signal; Determining whether the lateral displacement velocity or the lateral displacement acceleration is greater than a threshold; If so, an overspeed signal is output.

9. The Braille scanning method according to claim 4, wherein: After determining the Braille corresponding to the fused salient point data, the method further includes determining the horizontal coordinates and vertical coordinates of the Braille corresponding to the salient point data based on the horizontal displacement signal and the vertical displacement signal of the gyroscope, and using the Braille and its corresponding coordinates as the Braille recognition result.

10. The Braille scanning method according to claim 4, characterized in that: The first salient point data obtained based on the image specifically includes: Preprocessing the image to obtain a preprocessed image; Extract salient points from the preprocessed image as the first salient point data.

11. The Braille scanning method according to claim 10, characterized in that: Extract salient points from the preprocessed image, specifically including: Selecting a circular structure element with a preset radius to subdivide the preprocessed image to obtain at least one subdivided area, each of the subdivided areas including one or more Braille characters; For each subdivided area, locally segment the subdivided area using a large-law threshold segmentation method to obtain at least one local segmented area, each of which includes a salient point; According to the linear characteristics of the Braille arrangement, isolated points and miscellaneous points in each local segmentation area are removed to obtain salient points in the local segmentation area; The set of salient points in all subdivided areas is used as the salient points in the preprocessed image.

12. The Braille scanning method according to claim 10, wherein: Preprocess the image to obtain a preprocessed image, specifically including: performing contrast-limited adaptive histogram equalization processing on the image to obtain an equalized image; Adjacent regions in the equalized image are combined using a bilinear interpolation method to eliminate artificial boundaries and obtain a preprocessed image.

13. The Braille scanning method according to claim 4, characterized in that: The first salient point data is obtained from the image using an augmented reality recognition algorithm.

Citation Information

Patent Citations

  • Lateral fingerprint module and mobile terminal possessing same

    CN107506757A

  • Braille point display reading pen

    CN210606228U