CN114973292B - Character recognition method and system based on irregular surface and storage medium - Google Patents
Character recognition method and system based on irregular surface and storage medium Download PDFInfo
- Publication number
- CN114973292B CN114973292B CN202210905479.8A CN202210905479A CN114973292B CN 114973292 B CN114973292 B CN 114973292B CN 202210905479 A CN202210905479 A CN 202210905479A CN 114973292 B CN114973292 B CN 114973292B
- Authority
- CN
- China
- Prior art keywords
- character recognition
- recognition area
- area
- image data
- character
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/414—Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06V30/1475—Inclination or skew detection or correction of characters or of image to be recognised
- G06V30/1478—Inclination or skew detection or correction of characters or of image to be recognised of characters or characters lines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06V30/18—Extraction of features or characteristics of the image
- G06V30/186—Extraction of features or characteristics of the image by deriving mathematical or geometrical properties from the whole image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Character Input (AREA)
Abstract
The application relates to the technical field of character recognition, in particular to a character recognition method, a system and a storage medium based on an irregular surface, wherein the method comprises the following steps: acquiring image data of a target object; acquiring a character recognition area in image data; acquiring the object space geometric characteristics of the character recognition area; confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object, and correcting the character recognition area by adopting a corresponding mechanism; and performing character recognition on the corrected character recognition area by using OCR character recognition. The method comprises the steps of calculating specific parameters of font deformation by using the space geometric characteristics of an object, further processing the specific parameters by combining image data to obtain an abnormal recognition type existing on the surface of the object, correcting a character recognition area by adopting a corresponding mechanism, and finally restoring font fonts, so that the effect of improving the font recognition rate on an irregular surface is achieved, and meanwhile, the intelligent degree of an automatic system can be further improved.
Description
Technical Field
The application relates to the technical field of character recognition, in particular to a character recognition method based on an irregular surface.
Background
With the widespread adoption of automation technology in various industries, the recognition of characters on the surface of an object becomes a main way for an automation system to identify the object. Thus, when faced with some irregular surfaces, the recognition of the text on the irregular surface becomes a more difficult problem.
At present, an OCR technology is mainly adopted for a character recognition method, the technology has certain requirements on the character font normalization, even if some common reasoning technologies (such as ncnn reasoning algorithm) are used, the technology usually has certain effect only aiming at some inclined character font deformation with smaller amplitude, but under the special conditions such as pipelines, included angles, high-low dislocation and the like, the recognition rate is often lower, and the requirement of industrial production is far from being met.
Disclosure of Invention
In order to improve the recognition rate of characters on the irregular surface and further improve the intelligent degree of an automatic system, the application provides a character recognition method and system based on the irregular surface and a storage medium.
In a first aspect, the application provides a method for recognizing characters based on an irregular surface, which adopts the following technical scheme:
a character recognition method based on an irregular surface comprises the following steps:
acquiring image data of a target object;
acquiring a character recognition area in the image data;
acquiring the object space geometric characteristics of the character recognition area;
confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object, and correcting the character recognition area by adopting a corresponding mechanism;
and performing character recognition on the corrected character recognition area by using OCR character recognition.
By adopting the technical scheme, the specific parameters of font deformation are calculated by utilizing the geometric characteristics of the object space, further processing is carried out by combining image data, the abnormal recognition type existing on the surface of the object is obtained, the character recognition area is corrected by adopting a corresponding mechanism, and the font is finally restored, so that the effect of improving the font recognition rate on the irregular surface is achieved, and meanwhile, the intelligent degree of an automatic system can be further improved.
Optionally, the step of obtaining the object space geometric feature of the character recognition area includes:
identifying solid geometric information in the image data, and positioning the relative position of the character identification area in the surface of the object based on the solid geometric information;
acquiring the object space geometric characteristics of the character recognition area according to the relative position; wherein the object space geometric characteristics include curvature, included angle, and height difference.
Optionally, the step of determining the abnormal recognition type of the object surface according to the object space geometric characteristics and correcting the character recognition area by using a corresponding mechanism includes:
under the condition that the surface of the object has a curved surface and/or an included angle, vertically mapping the characters in the character recognition area to a line segment of the mapping area by adopting point-to-point mapping;
calculating the length of an arc line in the character recognition area and the length of a line segment in the mapping area;
and stretching the character recognition area according to the ratio of the arc length to the line length to obtain the corrected character recognition area.
Optionally, after the step of stretching the character recognition area according to the ratio of the arc length to the line length to obtain the corrected character recognition area, the method further includes:
detecting pixels of the character recognition area;
and under the condition that the pixel is lower than a preset threshold value, supplementing the pixel according to the color of the character recognition area.
Optionally, the step of determining the abnormal recognition type of the object surface according to the object space geometric feature and correcting the character recognition area by using a corresponding mechanism includes:
and under the condition that an included angle exists on the surface of the object, adopting an inclined character reasoning module to carry out reasoning and recognition on the characters in the character recognition area to obtain a corrected character recognition area.
Optionally, the step of determining the abnormal recognition type of the object surface according to the object space geometric feature and correcting the character recognition area by using a corresponding mechanism includes:
under the condition that an irregular plane or geometric dislocation exists on the surface of the object, a dislocation area and a difference value are identified according to the spatial geometric characteristics of the object;
and splicing image pixel data according to the difference value to obtain a corrected character recognition area.
In a second aspect, the application provides a character recognition system based on an irregular surface, which adopts the following technical scheme:
a system for recognizing characters based on irregular surfaces, comprising: the image acquisition module is used for acquiring image data of the target object;
the area identification module is used for acquiring a character identification area in the image data;
the characteristic acquisition module is used for acquiring the object space geometric characteristics of the character recognition area;
the correction module is used for confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object and correcting the character recognition area by adopting a corresponding mechanism;
and the recognition module is used for carrying out character recognition on the corrected character recognition area by utilizing OCR character recognition.
By adopting the technical scheme, the specific parameters of font deformation are calculated by utilizing the geometric characteristics of the object space, further processing is carried out by combining image data, the abnormal recognition type existing on the surface of the object is obtained, the character recognition area is corrected by adopting a corresponding mechanism, and the font is finally restored, so that the effect of improving the font recognition rate on the irregular surface is achieved, and meanwhile, the intelligent degree of an automatic system can be further improved.
In a third aspect, the present application provides a computer-readable storage medium, which adopts the following technical solutions:
a computer readable storage medium storing a computer program capable of being loaded by a processor and performing any of the methods according to the first aspect.
In summary, the present application includes at least one of the following beneficial technical effects:
the method comprises the steps of calculating specific parameters of font deformation by using the space geometric characteristics of an object, further processing the specific parameters by combining image data to obtain an abnormal recognition type existing on the surface of the object, correcting a character recognition area by adopting a corresponding mechanism, and finally restoring font fonts, so that the effect of improving the font recognition rate on an irregular surface is achieved, and meanwhile, the intelligent degree of an automatic system can be further improved.
Drawings
Fig. 1 is a flowchart illustrating a text recognition method according to an embodiment of the present application.
Fig. 2 is a block diagram illustrating a method for correcting a character recognition area according to an embodiment of the present application.
Fig. 3 is a block flow diagram of a text recognition system according to an embodiment of the present application.
Description of the reference numerals: 1. an image acquisition module; 2. a region identification module; 3. a feature acquisition module; 4. a rectification module; 5. and identifying the module.
Detailed Description
The present application is described in further detail below with reference to fig. 1-3.
The embodiment of the application discloses a character recognition method based on an irregular surface, which is beneficial to solving the problem of character acquisition on irregular objects, including but not limited to tubular objects, included angles, uneven object surfaces and the like.
As one embodiment of the character recognition method, as shown in fig. 1, the character recognition method includes:
100, image data of a target object is acquired.
Note that the camera is used to capture high-resolution image data of the target object.
200, acquiring a character recognition area in the image data.
Specifically, the character recognition area in the image data is locked by machine learning. In the machine learning, hog feature detection and svm classification are utilized, and the compiling environment used is vs2010 and opencv2.4.9.
When detecting, dividing a picture to be detected into RGBY four colors to obtain an outer contour of the region; then, edge detection is used, canny edge detection can be selected, the minimum external rectangle of each unconnected edge is calculated, and the position of the rectangle is returned, so that the character recognition area which is interested by people can be obtained. For convenience, we can use the cvFindCounter () function in opencv directly, and the third rectangle of this function can return a rectangular outline. And respectively detecting the returned rectangular outlines, and finally obtaining a desired result.
In this embodiment, first, interest identifiers, such as icons, markers, specific graphic structures, and the like, are determined, regions of the interest points are determined, and then, text regions of adjacent regions are found according to the positions of the interest points.
And 300, acquiring the space geometric characteristics of the object in the character recognition area.
Wherein, step 300 specifically comprises:
301, recognizing solid geometric information in image data, and positioning relative position of character recognition area in surface of object based on solid geometric information;
302, obtaining the object space geometric characteristics of the character recognition area according to the relative position; wherein the object space geometric characteristics include curvature, included angle, and height difference.
Specifically, the method comprises the steps of calling object point cloud data captured by a laser radar or three-dimensional space information captured by a depth camera, identifying solid geometry information in image data based on the object point cloud data or the three-dimensional space information, wherein the solid geometry information comprises XYZ coordinate information of a character identification area, and carrying out equal-proportion correspondence through a mode that the XYZ coordinate information of the character identification area is mapped with the image data coordinate information, so that the relative position of the character identification area in the surface of an object is judged. For example, a lidar has a resolution of 100 × 100 in the area of the plane and a resolution of 300 × 300 in the image data, the area of coordinates (10, 10) to (30, 30) is mapped to the area of (30, 30) to (90, 90).
And if the character recognition area is at the upper left corner of the image data, acquiring object point cloud data at the upper left corner, matching the image data with the object point cloud data, and positioning spatial geometric characteristics of the character recognition area, such as curvature, included angle, height difference and the like. The object point cloud data is fitted into a straight line or a curve, and then the curvature, the included angle, the height difference and the like are obtained through calculation.
It should be noted that, when the spatial geometric features cannot be identified due to unclear image data shooting or poor matching between the image data and the object point cloud data, the process returns to step 100 to shoot the image again.
If no curvature, included angle, height difference and the like exist, the character recognition area is not abnormal, normal recognition information is returned, and OCR character recognition is adopted to recognize the character recognition area locked in the step 200; otherwise step 400 is entered.
And 400, confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object, and correcting the character recognition area by adopting a corresponding mechanism.
In the case where a curvature exists, it is confirmed that a curved surface exists on the surface of the object; under the condition that the included angle exists, the included angle of the surface of the object is confirmed; when the height difference exists, the surface of the object is confirmed to have geometric dislocation or character splitting.
As an embodiment of the method for correcting the character recognition area, with reference to fig. 2, the method includes:
401, under the condition that the surface of the object has a curved surface and/or an included angle, vertically mapping characters in the character recognition area to line segments of the mapping area by adopting point-to-point mapping;
402, calculating the arc length in the character recognition area and the line length in the mapping area;
403, stretching the characters in the character recognition area according to the ratio of the length of the arc line to the length of the line to obtain corrected characters;
404, detecting pixels of the character recognition area;
and 405, if the pixel is lower than a preset threshold value, supplementing the pixel according to the color of the character recognition area.
Specifically, the arcs on the text area in the drawing are the areas where the text exists, i.e. the text recognition areas, and the text is firstly vertically mapped onto the line segments of the mapping area in a point-to-point mapping manner, and then the text is deformed and extruded in the horizontal direction. To restore the characters to the original font, the length of the arc and the length of the line segment need to be calculated, and the characters are stretched according to the ratio of the length of the arc and the length of the line segment to obtain corrected characters.
The method comprises the steps of calculating the length of an arc line and the length of a line segment through object point cloud data captured by a laser radar, firstly calculating the radius of a character area, then determining the range of the arc line, and calculating the length of the corresponding arc line through the radius after the length of the line segment is known.
In the case that the detection pixel is too low, the definition of the whole image is affected, and the recognition of the characters is inconvenient, so that the pixel is supplemented according to the color of the character recognition area to improve the definition of the image data.
Another embodiment of a method for correcting a character recognition area includes:
and under the condition that an included angle exists on the surface of the object, adopting an inclined character reasoning module to carry out reasoning recognition on the characters in the character recognition area to obtain a corrected character recognition area.
It should be noted that the inclined character inference module may adopt an NCNN inference algorithm, and the NCNN algorithm is used to patch and restore the image, so as to obtain a corrected character recognition region.
In another embodiment, a method for correcting a character recognition area includes:
under the condition that the surface of an object has geometric dislocation or character fracture, identifying a dislocation area and a difference value according to the spatial geometric characteristics of the object;
and splicing image pixel data according to the difference value to obtain a corrected character recognition area.
It should be noted that, if the surface of the object has problems of geometric misalignment, character splitting, etc. due to unevenness, etc., the misaligned region and difference are identified according to the spatial geometric features of the object, and then the image pixel data are re-stitched to obtain the corrected character identification region.
Specifically, the internal value and the external value of the character recognition area are obtained by using the point cloud data of the object, wherein the internal value and the external value are coordinate values of different areas in the character recognition area, the height difference is calculated based on the internal value and the external value, and the surface of the object is uneven under the condition that the internal value and the external value are inconsistent. At this time, image pixel data is re-spliced, that is, the external value or the internal value is adjusted to make the internal value and the external value consistent, so that a corrected character recognition area is obtained.
And 500, performing character recognition on the corrected character recognition area by using OCR character recognition.
Based on the character recognition method, the embodiment of the application also discloses a character recognition system based on the irregular surface.
As an embodiment of the character recognition system, as shown in fig. 3, the character recognition system includes:
an image acquisition module 1 for acquiring image data of a target object;
the area identification module 2 is used for acquiring a character identification area in the image data;
the characteristic acquisition module 3 is used for acquiring the object space geometric characteristics of the character recognition area;
the correction module 4 is used for confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object and correcting the character recognition area by adopting a corresponding mechanism;
and the recognition module 5 is used for carrying out character recognition on the corrected character recognition area by using OCR character recognition.
The embodiment of the application also discloses a computer readable storage medium, which stores a computer program that can be loaded by a processor and executes the operation control method of the central ferry vehicle, and the computer readable storage medium comprises: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above embodiments are preferred embodiments of the present application, and the protection scope of the present application is not limited by the above embodiments, so: all equivalent changes made according to the structure, shape and principle of the present application shall be covered by the protection scope of the present application.
Claims (7)
1. A character recognition method based on an irregular surface is characterized by comprising the following steps:
acquiring image data of a target object;
acquiring a character recognition area in the image data;
acquiring the object space geometric characteristics of the character recognition area;
confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object, and correcting the character recognition area by adopting a corresponding mechanism;
performing character recognition on the corrected character recognition area by using OCR character recognition;
the step of obtaining the object space geometric characteristics of the character recognition area comprises the following steps:
calling object point cloud data captured by a laser radar or three-dimensional space information captured by a depth camera;
identifying solid geometric information in the image data based on the object point cloud data or the three-dimensional space information, carrying out equal-proportion correspondence based on a mapping mode of the solid geometric information and image data coordinate information, and positioning the relative position of a character identification area in the surface of the object; the solid geometry information comprises XYZ coordinate information of the character recognition area;
matching the image data with the object point cloud data according to the relative position, fitting the object point cloud data into a straight line or a curve, and acquiring the object space geometric characteristics of the character recognition area; wherein the object space geometric characteristics include curvature, included angle, and height difference.
2. The method according to claim 1, wherein the step of confirming the abnormal recognition type of the object surface according to the object space geometric features and correcting the character recognition area by using a corresponding mechanism comprises:
under the condition that the surface of the object has a curved surface and/or an included angle, vertically mapping characters in the character recognition area to line segments of the mapping area by adopting point-to-point mapping;
calculating the length of an arc line in the character recognition area and the length of a line segment in the mapping area;
and stretching the character recognition area according to the ratio of the arc length to the line segment length to obtain the corrected character recognition area.
3. The method of claim 2, wherein the step of stretching the text recognition area according to the ratio of the arc length to the segment length to obtain a corrected text recognition area further comprises:
detecting pixels of the character recognition area;
and under the condition that the pixel is lower than a preset threshold value, supplementing the pixel according to the color of the character recognition area.
4. The method according to claim 1, wherein the step of confirming the abnormal recognition type of the object surface according to the object space geometric features and correcting the character recognition area by using a corresponding mechanism comprises:
and under the condition that an included angle exists on the surface of the object, adopting an inclined character reasoning module to carry out reasoning and recognition on the characters in the character recognition area to obtain a corrected character recognition area.
5. The method according to claim 1, wherein the step of confirming the abnormal recognition type of the object surface according to the object space geometric features and correcting the character recognition area by using a corresponding mechanism comprises:
under the condition that an irregular plane or geometric dislocation exists on the surface of the object, a dislocation area and a difference value are identified according to the spatial geometric characteristics of the object;
and splicing image pixel data according to the difference value to obtain a corrected character recognition area.
6. An irregular surface based word recognition system comprising: the image acquisition module is used for acquiring image data of the target object;
the area identification module is used for acquiring a character identification area in the image data;
the characteristic acquisition module is used for acquiring the object space geometric characteristics of the character recognition area;
the correction module is used for confirming the abnormal recognition type existing on the surface of the object according to the space geometric characteristics of the object and correcting the character recognition area by adopting a corresponding mechanism;
the recognition module is used for carrying out character recognition on the corrected character recognition area by utilizing OCR character recognition;
the feature acquisition module includes:
calling object point cloud data captured by a laser radar or three-dimensional space information captured by a depth camera;
identifying solid geometric information in the image data based on the object point cloud data or the three-dimensional space information, carrying out equal-proportion correspondence based on a mapping mode of the solid geometric information and image data coordinate information, and positioning the relative position of a character identification area in the surface of the object; the solid geometry information comprises XYZ coordinate information of the character recognition area;
matching the image data with the object point cloud data according to the relative position, fitting the object point cloud data into a straight line or a curve, and acquiring the object space geometric characteristics of the character recognition area; wherein the object space geometric characteristics include curvature, included angle, and height difference.
7. A computer-readable storage medium, characterized in that: a computer program which can be loaded by a processor and which performs the method according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210905479.8A CN114973292B (en) | 2022-07-29 | 2022-07-29 | Character recognition method and system based on irregular surface and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210905479.8A CN114973292B (en) | 2022-07-29 | 2022-07-29 | Character recognition method and system based on irregular surface and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114973292A CN114973292A (en) | 2022-08-30 |
| CN114973292B true CN114973292B (en) | 2022-11-15 |
Family
ID=82969849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210905479.8A Active CN114973292B (en) | 2022-07-29 | 2022-07-29 | Character recognition method and system based on irregular surface and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114973292B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9390342B2 (en) * | 2011-10-17 | 2016-07-12 | Sharp Laboratories Of America, Inc. | Methods, systems and apparatus for correcting perspective distortion in a document image |
| CN104809436B (en) * | 2015-04-23 | 2017-12-15 | 天津大学 | One kind bending written recognition methods |
| CN105590112B (en) * | 2015-09-22 | 2018-12-04 | 成都数联铭品科技有限公司 | Text judgment method is tilted in a kind of image recognition |
| CN111242126A (en) * | 2020-01-15 | 2020-06-05 | 上海眼控科技股份有限公司 | Irregular text correction method and device, computer equipment and storage medium |
| CN111507325B (en) * | 2020-03-16 | 2023-04-07 | 重庆大学 | Industrial visual OCR recognition system and method based on deep learning |
-
2022
- 2022-07-29 CN CN202210905479.8A patent/CN114973292B/en active Active
Non-Patent Citations (3)
| Title |
|---|
| 三维书籍模型的单侧页面提取方法;童立靖等;《计算机工程与设计》;20170630;第38卷(第6期);第1564-1569页 * |
| 基于纸张仿真的表格文档图像数据集生成***的设计与实现;杨静姝;《中国优秀硕士学位论文全文数据库信息科技辑》;20220115;参见第10-11页 * |
| 面向金融业务的表单票据通用识别***研究;任多;《中国优秀硕士学位论文全文数据库信息科技辑》;20220115;第19页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114973292A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107507167B (en) | Cargo tray detection method and system based on point cloud plane contour matching | |
| CN109636815B (en) | Metal plate strip product label information identification method based on computer vision | |
| CN109389121B (en) | Nameplate identification method and system based on deep learning | |
| US9257043B2 (en) | Lane correction system, lane correction apparatus and method of correcting lane | |
| EP3309704A1 (en) | Two-dimensional code partitioning and decoding method and system | |
| CN104794421A (en) | QR (quick response) code positioning and recognizing methods | |
| CN111861979A (en) | Positioning method, positioning equipment and computer readable storage medium | |
| CN112419207A (en) | Image correction method, device and system | |
| CN115713556A (en) | Two-dimensional code positioning method and device and computer readable storage medium | |
| CN110288040B (en) | Image similarity judging method and device based on topology verification | |
| CN117557565B (en) | Detection method and device for lithium battery pole piece | |
| CN114387329A (en) | Building contour progressive regularization method based on high-resolution remote sensing image | |
| CN117382426B (en) | Vehicle-mounted pantograph self-adaptive control method and system | |
| CN112184723B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN114973292B (en) | Character recognition method and system based on irregular surface and storage medium | |
| CN111627069A (en) | Dot detection method, terminal and computer-readable storage medium | |
| CN111178210A (en) | Image identification and alignment method for cross mark | |
| CN113537216B (en) | Dot matrix font text line inclination correction method and device | |
| JP4543675B2 (en) | How to recognize characters and figures | |
| CN115984211A (en) | Visual positioning method, robot and storage medium | |
| CN115205861A (en) | Method for acquiring abnormal character recognition area, electronic equipment and storage medium | |
| CN111667429B (en) | Target positioning correction method for inspection robot | |
| CN114120320A (en) | Image multi-target information identification method, system and medium | |
| CN112304291B (en) | HUD-based lane line display method and computer-readable storage medium | |
| WO2018057082A1 (en) | Curvature-based face detector |
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 |