CN110619331A - Color distance-based color image field positioning method - Google Patents
Color distance-based color image field positioning method Download PDFInfo
- Publication number
- CN110619331A CN110619331A CN201910891636.2A CN201910891636A CN110619331A CN 110619331 A CN110619331 A CN 110619331A CN 201910891636 A CN201910891636 A CN 201910891636A CN 110619331 A CN110619331 A CN 110619331A
- Authority
- CN
- China
- Prior art keywords
- color
- image
- distance
- hsv
- connected domains
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Input (AREA)
- Image Analysis (AREA)
Abstract
A color image field positioning method based on color distance comprises the steps of preprocessing an image, carrying out image correction on the inclination condition of a shot image, identifying the color of an image pixel point, carrying out color filtering, removing background and non-text information, then extracting possible connected domains in the image, obtaining single character position information according to the similarity of the color and the position distance of the connected domains in characters, and finally carrying out connected domain fusion on the single character connected domain and the connected domains in the neighborhood according to the characteristics of the size of the characters in the same line and the distance between the characters and extending the single character connected domain and the connected domains in the neighborhood to possible paragraph boundaries so as to position the whole paragraph.
Description
Technical Field
The invention belongs to the field of image recognition and video analysis, and particularly relates to a color image field positioning method based on color distance.
Background
The text information is the most core part of the image, and in order to extract the text information in the image, the OCR (Optical Character Recognition) technology is developed and is undergoing continuous development and iteration. In the current big data development era, data is a foundation, an OCR technology is a crucial link, and the data structuralization of the OCR technology has great significance for the realization of subsequent automatic understanding, indexing and retrieval of semantics. In the former OCR, for full text recognition, all characters are extracted from unstructured image data, and the characters corresponding to different information have different colors, sizes, brightnesses, etc. the full text uses the same parameters to perform the OCR operation, which easily causes information omission or even ignores key information with smaller fonts and lighter colors, and the full text is mapped into a field, so that information classification and extraction are difficult in the later stage. Therefore, text paragraph positioning is indispensable before character recognition, and the quality of text positioning effect directly affects character recognition.
In a color image, people often embody different information through different colors, for example, in the financial field, financial bills are marked with different key contents or license plate information by different colors in the color image, and characters are highlighted by blue background and white color.
Patent CN105868757A "a method and apparatus for locating characters in image characters" this method removes images of non-character parts on the original image after edge extraction, edge point detection and binarization method on the basis of the original image, and retains images of character parts, thus realizing character location. The edge-based method focuses on the contrast between the text and the background, but is only suitable for scenes with high text-to-background contrast and prominent text edges.
The patent CN103440487B "a natural scene character positioning method with local hue difference", this invention utilizes the textural features of characters, and combines the characteristics of different hues of character region and surrounding region to position. Although the text positioning method based on the character texture features can detect scenes with small contrast between characters and a background, the calculation amount is large, the timeliness of the algorithm is poor, and the method cannot be well adapted to images with noise interference.
The patent CN107301414A discloses a Chinese positioning, segmenting and identifying method in natural scene images, which utilizes accurate extraction of character stroke characteristics and a depth residual error neural network technology to position and identify characters. The method based on machine learning needs a large amount of learning and training, and the reality scene is diversified and cannot be comprehensive, so the obtained training result has no strong robustness.
Disclosure of Invention
Technical problem to be solved
The present invention is directed to a color distance-based color image field positioning method, so as to solve the problems of the background art.
(II) technical scheme
In order to achieve the purpose, the invention provides the following technical scheme: a color distance-based color image field positioning method comprises the following steps:
step 1: preprocessing an image, and performing image correction on the condition that a shot image is inclined;
step 2: identifying the color of the image pixel, performing color filtering, and removing the background color and the non-text color pixel;
and step 3: extracting possible connected domains in the image, and filtering a non-character structure;
and 4, step 4: acquiring single character position information according to the similarity of the color and the distance of the connected domain in the character;
and 5: according to the character size and space of the same line of characters, the single character is expanded to extend to the possible paragraph boundary, so as to position the whole paragraph.
Preferably, the step 1 comprises the following steps:
(a) finding out all straight lines in the image according to a method for detecting the straight lines by HOUGH;
(b) finding a maximum straight line smaller than the image size among the searched straight lines;
(c) calculating an angle V between the maximum straight line and the horizontal direction;
(d) if the angle V is not a multiple of 90 degrees, the image is rotated by V degrees.
Preferably, the step 2 comprises the following steps:
(a) the HSV color model re-maps the RGB model, and thus can be more visually intuitive than the RGB model. Therefore, the image is mapped to an HSV color space, color components of three channels in the HSV are extracted, and pixel points in the image are classified into different colors according to different ranges of the three HSV color components;
(b) and carrying out color filtering on the image, and removing the background color and the non-text color pixel points.
(c) And (4) carrying out secondary processing on the pixel points after color filtering, and if the pixel points in 8 neighborhoods are different from the pixel points and are in the same color system, replacing the color of the pixel point with the color of the neighborhood.
Preferably, the step 3 comprises the following steps:
(a) taking the pixel points subjected to color extraction and filtering as foreground pixel points, and extracting all connected domains in the image based on the foreground pixel points;
(b) the minimum value M1 and the maximum value M2 are set according to the size characteristics of the individual characters in the color image and the different structures contained in the characters. And removing the connected domain of which the pixel point contained in the connected domain is smaller than M1 or the pixel point contained in the connected domain is larger than M2.
Preferably, the step 4 comprises the following steps:
(a) setting a smaller threshold value N1 for the extracted connected domain, and calculating the distance position between the connected domain in the neighborhood and the connected domain;
(b) and fusing the connected domains with the connected domains which have the same color and the position distance smaller than N1 in the adjacent domains, and circling the connected domains by using the minimum circumscribed rectangle of the connected domains to position a single character.
Preferably, the step 5 comprises the following steps:
(a) for a rectangular frame of a single font defined in an image, calculating color distances between a left rectangular frame and a right rectangular frame and the rectangular frame, and a position distance, wherein the color distance between the two rectangular frames is performed in an HSV color channel, a model of an HSV (hue, saturation) color space corresponds to a conical subset in a cylindrical coordinate system, and the top surface of a cone corresponds to V1. The color represented by the RGB model comprises three planes of R1, G1 and B1, and is brighter. The color H is given by the rotation angle around the V-axis. In an HSV (hue, saturation and value) cone with the length of the oblique edge R, the radius of the bottom surface circle R and the height H, a coordinate axis is established in the positive direction of the x axis by taking the center of the ground as the origin and taking H as 0. The three-dimensional coordinates (x, y, z) of the point where the color value is (H, S, V) are then:
calculating average color component points (h1, s1, v1) and (h2, s2, v2) of pixels in the two connected domains to be converted into three-dimensional coordinate points (x1, y1, z1) and (x2, y2, z2), wherein the Euclidean distance of the three-dimensional coordinate points is the color distance of the two connected domains;
(b) setting a color distance threshold N2 and a position distance threshold N3, fusing two rectangular frames with color distance smaller than N2, position distance smaller than N3 and consistent height, and positioning the fields.
(III) advantageous effects
The invention aims to provide a color image field positioning method based on color distance. First, an image is preprocessed, and image correction is performed on the condition that the photographed image is inclined. And secondly, identifying the color of the image pixel point, performing color filtering, and removing background and non-text information. Then, extracting possible connected domains in the image, obtaining position information of a single character according to the similarity of the color and the position distance of the connected domains in the character, and finally fusing the connected domains of the single character connected domain and the adjacent connected domains thereof according to certain conditions and extending to possible paragraph boundaries according to the characteristics of the size of the same line of characters and the character distance so as to position the whole paragraph.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of the algorithm of the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides a technical solution: a color distance-based color image field positioning method comprises the following steps:
step 1: preprocessing an image, and performing image correction on the condition that a shot image is inclined;
step 2: identifying the color of the image pixel, performing color filtering, and removing the background color and the non-text color pixel;
and step 3: extracting possible connected domains in the image, and filtering a non-character structure;
and 4, step 4: acquiring single character position information according to the similarity of the color and the distance of the connected domain in the character;
and 5: according to the character size and space of the same line of characters, the single character is expanded to extend to the possible paragraph boundary, so as to position the whole paragraph.
Preferably, the step 1 comprises the following steps:
(a) finding out all straight lines in the characters according to a method for detecting straight lines by HOUGH;
(b) finding a maximum straight line smaller than the image size among the searched straight lines;
(c) calculating an angle V between the maximum straight line and the horizontal direction;
(d) if the angle V is not a multiple of 90 degrees, the image is rotated by V degrees.
Preferably, the step 2 comprises the following steps:
(a) the HSV color model re-maps the RGB model, and thus can be more visually intuitive than the RGB model. Therefore, the image is mapped to an HSV color space, color components of three channels in the HSV are extracted, and pixel points in the image are classified into different colors according to different ranges of the three HSV color components;
(b) and carrying out color filtering on the image, and removing the background color and the non-text color pixel points.
(c) And (4) carrying out secondary processing on the pixel points after color filtering, and if the pixel points in 8 neighborhoods are different from the pixel points and are in the same color system, replacing the color of the pixel point with the color of the neighborhood.
Preferably, the step 3 comprises the following steps:
(a) taking the pixel points subjected to color extraction and filtering as foreground pixel points, and extracting all connected domains in the image based on the foreground pixel points;
(b) the minimum value M1 and the maximum value M2 are set according to the size characteristics of the individual characters in the color image and the different structures contained in the characters. And removing the connected domain of which the pixel point contained in the connected domain is smaller than M1 or the pixel point contained in the connected domain is larger than M2.
Preferably, the step 4 comprises the following steps:
(a) setting a smaller threshold value N1 for the extracted connected domain, and calculating the distance position between the connected domain in the neighborhood and the connected domain;
(b) and fusing connected domains which have the same color and are positioned at a distance less than N1 in the neighborhood of the connected domains, and circling the connected domains by using the minimum circumscribed rectangle of the connected domains to position a single character.
Preferably, the step 5 comprises the following steps:
(a) for a rectangular frame of a single font defined in an image, calculating color distances between a left rectangular frame and a right rectangular frame and the rectangular frame, and a position distance, wherein the color distance between the two rectangular frames is performed in an HSV color channel, a model of an HSV (hue, saturation) color space corresponds to a conical subset in a cylindrical coordinate system, and the top surface of a cone corresponds to V1. The color represented by the RGB model comprises three planes of R1, G1 and B1, and is brighter. The color H is given by the rotation angle around the V-axis. In an HSV (hue, saturation and value) cone with the length of the oblique edge R, the radius of the bottom surface circle R and the height H, a coordinate axis is established in the positive direction of the x axis by taking the center of the ground as the origin and taking H as 0. The three-dimensional coordinates (x, y, z) of the point where the color value is (H, S, V) are then:
calculating average color component points (h1, s1, v1) and (h2, s2, v2) of pixels in the two connected domains to be converted into three-dimensional coordinate points (x1, y1, z1) and (x2, y2, z2), wherein the Euclidean distance of the three-dimensional coordinate points is the color distance of the two connected domains;
(b) setting a color distance threshold N2 and a position distance threshold N3, fusing two rectangular frames with color distance smaller than N2, position distance smaller than N3 and consistent height, and positioning the fields.
The invention fully utilizes the color characteristics in the color image, and distinguishes the background and the characters through the color of the characters in the image containing the text, wherein in some images, the background is white, and the characters are brighter colors. In particular, for example, in the image of an invoice for a value added tax, the color of the box surrounding the font is black, the font of the heading for each entry is red, and the specific content corresponding to each entry is blue or green. This allows for fast localization of text positions based on the sensitivity of the human eye to color. The invention distinguishes the font and the background by the principle. And for the extraction of the color, the HSV color channel is utilized, and the HSV color model remaps the RGB model, so that the visual intuition is better than that of the RGB model.
The traditional method generally utilizes the brightness, edge or texture information of the image and is sensitive to illumination and brightness change. Different from the characteristic of traditional bill character recognition according to gray level, the method makes full use of the color information of the color image, and can well remove the influence of illumination and brightness change. In addition, compared with the traditional method for positioning the text by simply utilizing the color information, the text combines the color, the position and the geometric characteristics of the characters with the same attribute as the paragraph, can distinguish the title, the guide word and the content information, better segments the characters, and provides a powerful basis for character recognition, storage and big data processing
In the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (6)
1. A color distance-based color image field positioning method is characterized in that:
the method comprises the following steps:
step 1: preprocessing an image, and performing image correction on the condition that a shot image is inclined;
step 2: identifying the color of the image pixel, performing color filtering, and removing the background color and the non-text color pixel;
and step 3: extracting possible connected domains in the image, and filtering a non-character structure;
and 4, step 4: acquiring single character position information according to the similarity of the color and the distance of the connected domain in the character;
and 5: according to the character size and space of the same line of characters, the single character is expanded to extend to the possible paragraph boundary, so as to position the whole paragraph.
2. The method as claimed in claim 1, wherein the color distance-based color image field positioning method comprises: the step 1 comprises the following steps:
(a) finding out all straight lines in the image according to a method for detecting the straight lines by HOUGH;
(b) finding a maximum straight line smaller than the image size among the searched straight lines;
(c) calculating an angle V between the maximum straight line and the horizontal direction;
(d) if the angle V is not a multiple of 90 degrees, the image is rotated by V degrees.
3. The method as claimed in claim 1, wherein the color distance-based color image field positioning method comprises: the step 2 comprises the following steps:
(a) the HSV color model re-maps the RGB model, so that the visual intuition is better than that of the RGB model, the image is mapped to the HSV color space, the color components of three channels in the HSV are extracted, and pixel points in the image are classified into different colors according to different ranges of the three HSV color components;
(b) carrying out color filtering on the image, and removing pixel points of background color and non-text color;
(c) and (4) carrying out secondary processing on the pixel points after color filtering, and if the pixel points in 8 neighborhoods are different from the pixel points and are in the same color system, replacing the color of the pixel point with the color of the neighborhood.
4. The method as claimed in claim 1, wherein the color distance-based color image field positioning method comprises: the step 3 comprises the following steps:
(a) taking the pixel points subjected to color extraction and filtering as foreground pixel points, and extracting all connected domains in the image based on the foreground pixel points;
(b) setting a minimum value M1 and a maximum value M2 according to the size characteristics of a single character in the color image and different structures contained in the character; and removing the connected domain of which the pixel point contained in the connected domain is smaller than M1 or the pixel point contained in the connected domain is larger than M2.
5. The method as claimed in claim 1, wherein the color distance-based color image field positioning method comprises: the step 4 comprises the following steps:
(a) setting a smaller threshold value N1 for the extracted connected domain, and calculating the distance position between the connected domain in the neighborhood and the connected domain;
(b) and fusing connected domains which have the same color and are positioned at a distance less than N1 in the neighborhood of the connected domains, and circling the connected domains by using the minimum circumscribed rectangle of the connected domains to position a single character.
6. The method as claimed in claim 1, wherein the color distance-based color image field positioning method comprises: the step 5 comprises the following steps:
(a) for a rectangular frame of a single font defined in an image, calculating color distances and position distances between a left rectangular frame and a right rectangular frame of the rectangular frame and the rectangular frame, wherein the color distance between the two rectangular frames is carried out in an HSV color channel, a model of an HSV (hue, saturation) color space corresponds to a conical subset in a cylindrical coordinate system, and the top surface of a cone corresponds to V1; the color represented by the three planes of R1, G1 and B1 in the RGB model is brighter; color H is given by the rotation angle around the V-axis; in an HSV (hue, saturation and value) cone with the length of a bevel edge R, the radius of a bottom surface circle R and the height of H, a coordinate axis is established in the positive direction of an x axis by taking the center of the ground as an origin and taking H as 0; the three-dimensional coordinates (x, y, z) of the point whose color value is (H, S, V) are
Calculating average color component points (h1, s1, v1) and (h2, s2, v2) of pixels in the two connected domains to be converted into three-dimensional coordinate points (x1, y1, z1) and (x2, y2, z2), wherein the Euclidean distance of the three-dimensional coordinate points is the color distance of the two connected domains;
(b) setting a color distance threshold N2 and a position distance threshold N3, fusing two rectangular frames with color distance smaller than N2, position distance smaller than N3 and similar height, and positioning the fields.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910891636.2A CN110619331A (en) | 2019-09-20 | 2019-09-20 | Color distance-based color image field positioning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910891636.2A CN110619331A (en) | 2019-09-20 | 2019-09-20 | Color distance-based color image field positioning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110619331A true CN110619331A (en) | 2019-12-27 |
Family
ID=68923631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910891636.2A Pending CN110619331A (en) | 2019-09-20 | 2019-09-20 | Color distance-based color image field positioning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110619331A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798528A (en) * | 2020-07-01 | 2020-10-20 | 西交利物浦大学 | Fabric color difference detection method, device, storage medium and equipment |
| CN113780260A (en) * | 2021-07-27 | 2021-12-10 | 浙江大学 | Computer vision-based intelligent barrier-free character detection method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1516073A (en) * | 1998-02-06 | 2004-07-28 | ��ʿͨ��ʽ���� | Colour image processor and patterns extractor |
| CN1588431A (en) * | 2004-07-02 | 2005-03-02 | 清华大学 | Character extracting method from complecate background color image based on run-length adjacent map |
| CN101339556A (en) * | 2008-08-18 | 2009-01-07 | 福建四通石材有限公司 | Method for implementing image color similarity contrast by taper cone coordinate |
| CN104112128A (en) * | 2014-06-19 | 2014-10-22 | 中国工商银行股份有限公司 | Digital image processing system applied to bill image character recognition and method |
| CN106156711A (en) * | 2015-04-21 | 2016-11-23 | 华中科技大学 | The localization method of line of text and device |
-
2019
- 2019-09-20 CN CN201910891636.2A patent/CN110619331A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1516073A (en) * | 1998-02-06 | 2004-07-28 | ��ʿͨ��ʽ���� | Colour image processor and patterns extractor |
| CN1588431A (en) * | 2004-07-02 | 2005-03-02 | 清华大学 | Character extracting method from complecate background color image based on run-length adjacent map |
| CN101339556A (en) * | 2008-08-18 | 2009-01-07 | 福建四通石材有限公司 | Method for implementing image color similarity contrast by taper cone coordinate |
| CN104112128A (en) * | 2014-06-19 | 2014-10-22 | 中国工商银行股份有限公司 | Digital image processing system applied to bill image character recognition and method |
| CN106156711A (en) * | 2015-04-21 | 2016-11-23 | 华中科技大学 | The localization method of line of text and device |
Non-Patent Citations (6)
| Title |
|---|
| 万燕 等: "自然场景下中文文本定位关键技术的研究", 《计算机应用与软件》 * |
| 哈明虎 等: "《视觉文档图像识别预处理》", 31 July 2007, 河北大学出版社 * |
| 徐平华 等: "民族服饰色彩构成分析方法研究", 《丝绸》 * |
| 李长云 等: "《智能感知技术及在电气工程中的应用》", 31 May 2017, 电子科技大学出版社 * |
| 杜刚: "银行票据识别系统的研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 蒋先刚: "《数字图像模式识别工程项目研究》", 31 March 2014, 西南交通大学出版社 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798528A (en) * | 2020-07-01 | 2020-10-20 | 西交利物浦大学 | Fabric color difference detection method, device, storage medium and equipment |
| CN113780260A (en) * | 2021-07-27 | 2021-12-10 | 浙江大学 | Computer vision-based intelligent barrier-free character detection method |
| CN113780260B (en) * | 2021-07-27 | 2023-09-19 | 浙江大学 | Barrier-free character intelligent detection method based on computer vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
| USRE47889E1 (en) | System and method for segmenting text lines in documents | |
| CN107688806B (en) | Affine transformation-based free scene text detection method | |
| WO2019085971A1 (en) | Method and apparatus for positioning text over image, electronic device, and storage medium | |
| CN105654072A (en) | Automatic character extraction and recognition system and method for low-resolution medical bill image | |
| Harraj et al. | OCR accuracy improvement on document images through a novel pre-processing approach | |
| CN104361336A (en) | Character recognition method for underwater video images | |
| CN109409356B (en) | Multi-direction Chinese print font character detection method based on SWT | |
| CN104463134B (en) | A kind of detection method of license plate and system | |
| CN103336961A (en) | Interactive natural scene text detection method | |
| CN112434699A (en) | Automatic extraction and intelligent scoring system for handwritten Chinese characters or components and strokes | |
| CN107122775A (en) | A kind of Android mobile phone identity card character identifying method of feature based matching | |
| Sun et al. | A visual attention based approach to text extraction | |
| CN112101108B (en) | Left-right pass-through sign recognition method based on graph pole position characteristics | |
| CN112818952A (en) | Coal rock boundary recognition method and device and electronic equipment | |
| CN108038458B (en) | Method for automatically acquiring outdoor scene text in video based on characteristic abstract diagram | |
| CN110619331A (en) | Color distance-based color image field positioning method | |
| CN113392819B (en) | Batch academic image automatic segmentation and labeling device and method | |
| CN111046866B (en) | Method for detecting RMB crown word number region by combining CTPN and SVM | |
| Zhang et al. | A novel approach for binarization of overlay text | |
| CN112418210A (en) | Intelligent classification method for tower inspection information | |
| CN116416624A (en) | Document electronization method and device based on layout correction and storage medium | |
| CN115393748A (en) | Method for detecting infringement trademark based on Logo recognition | |
| Sushma et al. | Text detection in color images | |
| Li et al. | Multi-polarity text segmentation using graph theory |
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 | ||
| CB02 | Change of applicant information |
Address after: No.268 Hanzhong Road, Gulou District, Nanjing, Jiangsu Province 210001 Applicant after: CLP Hongxin Information Technology Co.,Ltd. Address before: No.268 Hanzhong Road, Gulou District, Nanjing, Jiangsu Province 210001 Applicant before: JIANGSU HONGXIN SYSTEM INTEGRATION Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191227 |
|
| RJ01 | Rejection of invention patent application after publication |