CN103605973A - Image character detection and identification method - Google Patents
Image character detection and identification method Download PDFInfo
- Publication number
- CN103605973A CN103605973A CN201310511403.8A CN201310511403A CN103605973A CN 103605973 A CN103605973 A CN 103605973A CN 201310511403 A CN201310511403 A CN 201310511403A CN 103605973 A CN103605973 A CN 103605973A
- Authority
- CN
- China
- Prior art keywords
- image
- character
- white image
- black white
- gray
- 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.)
- Granted
Links
Images
Abstract
The invention provides an image character detection and identification method which comprises the following steps: first, making a standard black-and-white image of a corresponding character; second, acquiring a test image in real time, extracting the position of the character and processing the test image into a test black-and-white image; and third, comparing the test black-and-white image obtained from the second step with the standard black-and-white image made in the first step to judge whether the test black-and-white image is qualified. The method can completely solve the problem that image characters cannot be identified in automatic flat-panel television test, promotes realization of automatic detection on flat-panel television core components and improves product quality and labor efficiency.
Description
Technical field
The present invention relates to a kind of image character detection and identification method, particularly relate to a kind of method of flat panel TV field of automatic testing to the image character detection and identification gathering that be applicable to.
Background technology
In the process in flat panel TV cartridge assemblies processing technology, because of supplied materials device with production technology, limit, the cartridge assemblies after having welded has defect in various degree.The method of existing test-based examination is that manual wiring or needle-bar connect display screen substantially, by artificial visually examine, the quality of image deflects is judged, the method exists the misjudgement that artificial subjective factor causes, the impact of failing to judge.And along with the continuous rising of human cost, robotization detects progressively application.In automatically detecting, need the image character to gathering to detect, the mode of the general purpose O CR character recognition of more employing in industry, owing to gathering the problems such as brightness, thickness, stroke fracture or adhesion of character in image, causing in OCR identification has more erroneous judgement, identification difficulty in actual applications.
Summary of the invention
The technical problem to be solved in the present invention be to provide a kind of erroneous judgement less, identification is easy to for to image character detection and identification method.
The technical solution used in the present invention is as follows: a kind of image character detection and identification method, and its method step is: one, make corresponding character standard black white image; Two, Real-time Collection test pattern, extracts and is treated to test black white image to character position; Three, the standard black white image that test black white image step 2 being obtained is made with step 1 contrast judge whether qualified.
As it is further improved, the concrete grammar step of described step 1 is: 1, choose a width by the testing standard image of Real-time Collection; 2, the testing standard image that frame selects required detection character picture position to obtain step 1 extracts; 3, set a gray threshold P, the part that gradation of image is greater than to gray threshold P is defined as character pixels point, by the image extracting in step 2 is carried out to binary conversion treatment by gray threshold P, obtains standard black white image.
As it is further improved, the bitmap images of described testing standard image for going out by image capturing system Real-time Collection.
As it is further improved, the 8Bit-256 color bitmap image of take is example, and described black white image is for being that gray-scale value is that 0 black, background definition is entirely that gray-scale value is 255 complete white 8Bit-256 color bitmap image by character definition.
As it is further improved, the concrete grammar step that described gray threshold P sets is: A, the RGB bitmap images gathering by capture card is converted to gray level image by formula Gray=R*0.299+G*0.587+B*0.114; B, by the gray level image obtaining by carrying out grey level quantization between N gray area, draw N histogram; C, from N histogram, choose gray-scale pixels two the most intensive intervals that distribute, the intermediate value that defines these two interval gray scales is respectively Gray1 and Gray2; D, selected two gray-scale value P=(Gray1+ Gray2)/2 in the middle of interval in step B are defined as to gray threshold.
As it is further improved, described N is 10.
As it is further improved, in described step 3, contrast concrete grammar and be: the whole housing that extracts character contrasts judgement.
As it is further improved, described in extract the concrete grammar that the whole housing of character contrasts and be: a, to read binaryzation standard black white image and test black white image pixel be that identical x row * y is capable, and defining image is f (x, y); B, from image coordinate initial point start to standard black white image and test black white image carry out boundary search simultaneously; C, by columns and rows, search for respectively, by row in the pixel that to search containing gray-scale value by row be 0, minimum and maximal value is defined as respectively x
minand x
max; , by by line search, to row in the pixel that is 0 containing gray-scale value, minimum and maximal value is defined as respectively y
minand y
max; D, according to the character boundary maximal value and the minimum value that search in row, column, i.e. f (x
min, y
min) and f (x
max, y
max) 2 whole housings of can frame selecting image character, the whole housing of standard black white image and test black white image character is carried out to pointwise contrast, consistance reach setting value be judged as qualified.
As it is further improved, the concrete grammar that contrasts judgement in described step 3 also comprises: qualified if whole housing contrasts, carry out the point-to-point contrast judgement of single character, its concrete grammar is: from (the x of standard black white image and test black white image f (x, y)
min, y
min) coordinate points starts, and row or column is searched for, the 8Bit-256 color bitmap image column of take search is example, to standard black white image and black white image according to formula
,
by row search, when running into first E=0, the value of x may be defined as first character row right margin x
1; Thus can be by (x
min, y
min) and (x
1, y
min) 2 frames select first character; Search for second character from the pixel (x of f (x, y)
1, y
min) start by row search, when E ≠ 0, be the left margin of second character, the value that records x is x
2; While continuing to search for to next E=0 by row, the value of x is defined as second character row right margin x
2; Thus can be by (x
2, y
min) and (x
3, y
max) 2 frames select second character; The like until search for the pixel (x to f (x, y)
max, y
max) can complete search; The character that standard black white image and artwork master frame are selected is order 1,2,3～N successively, carries out one by one each pixel contrast, consistance reach setting value be judged as qualified.
Compared with prior art, the invention has the beneficial effects as follows: the method can solve the problem of image character None-identified in flat panel TV is tested automatically completely, advance flat panel TV cartridge assemblies to realize robotization and detect, Improving The Quality of Products and people's effect.
Accompanying drawing explanation
Fig. 1 is the wherein binary image housing search schematic diagram of an embodiment of the present invention.
Fig. 2 is the wherein method flow schematic diagram of an embodiment of the present invention.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
Disclosed arbitrary feature in this instructions (comprising any accessory claim, summary and accompanying drawing), unless narration especially all can be replaced by other equivalences or the alternative features with similar object.That is,, unless narration especially, each feature is an example in a series of equivalences or similar characteristics.
An image character detection and identification method, its method step is: one, make corresponding character standard black white image; Two, Real-time Collection test pattern, extracts and is treated to test black white image to character position; Three, the standard black white image that test black white image step 2 being obtained is made with step 1 contrast judge whether qualified.
In this specific embodiment, the concrete grammar step of described step 1 is: 1, choose a width by the testing standard image of the LVDS output of Real-time Collection; 2, the testing standard image that frame selects required detection character picture position to obtain step 1 extracts; 3, set a gray threshold P, the part that gradation of image is greater than to gray threshold P is defined as character pixels point, by the image extracting in step 2 is carried out to binary conversion treatment by gray threshold P, obtains standard black white image.
In this specific embodiment, the 8Bit-256 color bitmap image of described testing standard image for going out by image capturing system Real-time Collection.Black white image is for being that gray-scale value is that 0 black, background definition is entirely that gray-scale value is 255 complete white 8Bit-256 color bitmap image by character definition.
In this specific embodiment, the concrete grammar step that described gray threshold P sets is: A, the RGB bitmap images gathering by capture card is converted to gray level image by formula Gray=R*0.299+G*0.587+B*0.114; B, by the gray level image obtaining by carrying out grey level quantization between 10 (optimum analysis scheme) gray areas, draw 10 histograms; C, from 10 histograms, choose gray-scale pixels two the most intensive intervals that distribute, the intermediate value that defines these two interval gray scales is respectively Gray1 and Gray2; D, selected two gray-scale value P=(Gray1+ Gray2)/2 in the middle of interval in step B are defined as to gray threshold.
The whole housing that extracts character in step 3 contrasts judgement.Concrete grammar is: a, to read binaryzation standard black white image and test black white image pixel be that identical x row * y is capable, and defining image is f (x, y); B, from image coordinate initial point start to standard black white image and test black white image carry out boundary search simultaneously; C, by columns and rows, search for respectively, by row in the pixel that to search containing gray-scale value by row be 0, minimum and maximal value is defined as respectively x
minand x
max; , by by line search, to row in the pixel that is 0 containing gray-scale value, minimum and maximal value is defined as respectively y
minand y
max; D, according to the character boundary maximal value and the minimum value that search in row, column, i.e. f (x
min, y
min) and f (x
max, y
max) 2 whole housings of can frame selecting image character, the whole housing of standard black white image and test black white image character is carried out to pointwise contrast, consistance reach setting value be judged as qualified.In this specific embodiment, this setting value is 99.9%, is judged as defectively lower than 99.9%, and it is qualified to equal to be judged as higher than 99.9%.
As it is further improved, the concrete grammar that contrasts judgement in described step 3 also comprises: qualified if whole housing contrasts, carry out the point-to-point contrast judgement of single character, its concrete grammar is: from (the x of standard black white image and test black white image f (x, y)
min, y
min) coordinate points starts, and row or column is searched for, and take that to be listed as search be example, to standard black white image and black white image according to formula
,
by row search, when running into first E=0, the value of x may be defined as first character row right margin x
1; Thus can be by (x
min, y
min) and (x
1, y
min) 2 frames select first character; Search for second character from the pixel (x of f (x, y)
1, y
min) start by row search, when E ≠ 0, be the left margin of second character, the value that records x is x
2; While continuing to search for to next E=0 by row, the value of x is defined as second character row right margin x
2; Thus can be by (x
2, y
min) and (x
3, y
max) 2 frames select second character; The like until search for the pixel (x to f (x, y)
max, y
max) can complete search; The character that standard black white image and artwork master frame are selected is order 1,2,3～N successively, carries out one by one each pixel contrast, consistance reach setting value be judged as qualified.In this specific embodiment, this setting value is 99.99%, is judged as defectively lower than this value, and it is qualified to equal to be judged as higher than this value, thereby further improves accuracy rate.
Claims (9)
1. an image character detection and identification method, its method step is: one, make corresponding character standard black white image; Two, Real-time Collection test pattern, extracts and is treated to test black white image to character position; Three, the standard black white image that test black white image step 2 being obtained is made with step 1 contrast judge whether qualified.
2. method according to claim 1, the concrete grammar step of described step 1 is: 1, choose a width by the testing standard image of Real-time Collection; 2, the testing standard image that frame selects required detection character picture position to obtain step 1 extracts; 3, set a gray threshold P, the part that gradation of image is greater than to gray threshold P is defined as character pixels point, by the image extracting in step 2 is carried out to binary conversion treatment by gray threshold P, obtains standard black white image.
3. method according to claim 2, the bitmap images of described testing standard image for going out by image capturing system Real-time Collection.
4. method according to claim 3, the 8Bit-256 color bitmap image of take is example, described black white image is for being that gray-scale value is that 0 black, background definition is entirely that gray-scale value is 255 complete white 8Bit-256 color bitmap image by character definition.
5. method according to claim 2, the concrete grammar step that described gray threshold P sets is: A, the RGB bitmap images gathering by capture card is converted to gray level image by formula Gray=R*0.299+G*0.587+B*0.114; B, by the gray level image obtaining by carrying out grey level quantization between N gray area, draw N histogram; C, from N histogram, choose gray-scale pixels two the most intensive intervals that distribute, the intermediate value that defines these two interval gray scales is respectively Gray1 and Gray2; D, selected two gray-scale value P=(Gray1+ Gray2)/2 in the middle of interval in step B are defined as to gray threshold.
6. method according to claim 5, described N is 10.
7. method according to claim 1, contrasts concrete grammar and is in described step 3: the whole housing that extracts character contrasts judgement.
8. method according to claim 7, described in extract the concrete grammar that the whole housing of character contrasts and be: a, to read binaryzation standard black white image and test black white image pixel be that identical x row * y is capable, and defining image is f (x, y); B, from image coordinate initial point start to standard black white image and test black white image carry out boundary search simultaneously; C, by columns and rows, search for respectively, by row in the pixel that to search containing gray-scale value by row be 0, minimum and maximal value is defined as respectively x
minand x
max; , by by line search, to row in the pixel that is 0 containing gray-scale value, minimum and maximal value is defined as respectively y
minand y
max; D, according to the character boundary maximal value and the minimum value that search in row, column, i.e. f (x
min, y
min) and f (x
max, y
max) 2 whole housings of can frame selecting image character, the whole housing of standard black white image and test black white image character is carried out to pointwise contrast, consistance reach setting value be judged as qualified.
9. method according to claim 8, the concrete grammar that contrasts judgement in described step 3 also comprises: qualified if whole housing contrasts, carry out the point-to-point contrast judgement of single character, its concrete grammar is: from (the x of standard black white image and test black white image f (x, y)
min, y
min) coordinate points starts, and row or column is searched for, the 8Bit-256 color bitmap image column of take search is example, to standard black white image and black white image according to formula
,
by row search, when running into first E=0, the value of x may be defined as first character row right margin x
1; Thus can be by (x
min, y
min) and (x
1, y
min) 2 frames select first character; Search for second character from the pixel (x of f (x, y)
1, y
min) start by row search, when E ≠ 0, be the left margin of second character, the value that records x is x
2; While continuing to search for to next E=0 by row, the value of x is defined as second character row right margin x
2; Thus can be by (x
2, y
min) and (x
3, y
max) 2 frames select second character; The like until search for the pixel (x to f (x, y)
max, y
max) can complete search; The character that standard black white image and artwork master frame are selected is order 1,2,3～N successively, carries out one by one each pixel contrast, consistance reach setting value be judged as qualified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310511403.8A CN103605973B (en) | 2013-10-25 | 2013-10-25 | A kind of image character detection and identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310511403.8A CN103605973B (en) | 2013-10-25 | 2013-10-25 | A kind of image character detection and identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103605973A true CN103605973A (en) | 2014-02-26 |
| CN103605973B CN103605973B (en) | 2017-03-01 |
Family
ID=50124193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310511403.8A Active CN103605973B (en) | 2013-10-25 | 2013-10-25 | A kind of image character detection and identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103605973B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104240348A (en) * | 2014-09-12 | 2014-12-24 | 福建省智慧物联网研究院有限责任公司 | Admittance identity authentication method based on image identification |
| CN104504385A (en) * | 2014-12-10 | 2015-04-08 | 上海大学 | Recognition method of handwritten connected numerical string |
| CN105424190A (en) * | 2015-09-30 | 2016-03-23 | 广州超音速自动化科技股份有限公司 | Grayscale detection method of product appearance |
| CN110427885A (en) * | 2019-07-31 | 2019-11-08 | Tcl王牌电器(惠州)有限公司 | Detection method, device and the computer readable storage medium of nameplate |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010016059A1 (en) * | 1999-12-22 | 2001-08-23 | Andreas Krahn | Inspection device for packages |
| CN1567340A (en) * | 2003-06-23 | 2005-01-19 | 中国科学院研究生院 | False proof bill, false proof method of bill and system thereof |
| CN1975804A (en) * | 2006-12-15 | 2007-06-06 | 华南理工大学 | Education robot with character-learning and writing function and character recognizing method thereof |
| CN103049743A (en) * | 2012-12-27 | 2013-04-17 | 天津普达软件技术有限公司 | Detection system and detection method for characters at bottom of bowl |
-
2013
- 2013-10-25 CN CN201310511403.8A patent/CN103605973B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010016059A1 (en) * | 1999-12-22 | 2001-08-23 | Andreas Krahn | Inspection device for packages |
| CN1567340A (en) * | 2003-06-23 | 2005-01-19 | 中国科学院研究生院 | False proof bill, false proof method of bill and system thereof |
| CN1975804A (en) * | 2006-12-15 | 2007-06-06 | 华南理工大学 | Education robot with character-learning and writing function and character recognizing method thereof |
| CN103049743A (en) * | 2012-12-27 | 2013-04-17 | 天津普达软件技术有限公司 | Detection system and detection method for characters at bottom of bowl |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104240348A (en) * | 2014-09-12 | 2014-12-24 | 福建省智慧物联网研究院有限责任公司 | Admittance identity authentication method based on image identification |
| CN104504385A (en) * | 2014-12-10 | 2015-04-08 | 上海大学 | Recognition method of handwritten connected numerical string |
| CN104504385B (en) * | 2014-12-10 | 2017-12-22 | 上海大学 | The recognition methods of hand-written adhesion numeric string |
| CN105424190A (en) * | 2015-09-30 | 2016-03-23 | 广州超音速自动化科技股份有限公司 | Grayscale detection method of product appearance |
| CN110427885A (en) * | 2019-07-31 | 2019-11-08 | Tcl王牌电器(惠州)有限公司 | Detection method, device and the computer readable storage medium of nameplate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103605973B (en) | 2017-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107845087B (en) | Method and system for detecting uneven brightness defect of liquid crystal panel | |
| CN109879005B (en) | Belt tearing detection device and method | |
| CN104616275A (en) | Defect detecting method and defect detecting device | |
| US7978903B2 (en) | Defect detecting method and defect detecting device | |
| CN106851264B (en) | Camera module lens surface detection method and device | |
| CN108460757A (en) | A kind of mobile phone TFT-LCD screens Mura defects online automatic detection method | |
| WO2019001438A1 (en) | Display panel gray scale adjustment method and device | |
| EP3242268A1 (en) | Method for automatically identifying and calibrating medical color and gray-scale images | |
| CN108764352B (en) | Method and device for detecting repeated page content | |
| CN110009607B (en) | Display screen dead pixel detection method and device, computer equipment and storage medium | |
| CN105913093A (en) | Template matching method for character recognizing and processing | |
| CN104135660A (en) | Detection method of contamination of image pickup module and detection system | |
| CN104978748A (en) | Liquid crystal screen defect detection method based on local pixel values | |
| WO2003019274A1 (en) | Method for sorting ununiformity of liquid crystal display panel, sorting apparatus, and information recorded medium with recorded program for executing this sorting | |
| WO2018010387A1 (en) | Reversed element detection method and system | |
| WO2018010386A1 (en) | Method and system for component inversion testing | |
| CN111289538A (en) | PCB element detection system and detection method based on machine vision | |
| CN103605973A (en) | Image character detection and identification method | |
| CN111261079B (en) | Method for detecting abnormal phenomena of bright and dark spots | |
| CN112819844B (en) | Image edge detection method and device | |
| CN110415223B (en) | No-reference spliced image quality evaluation method and system | |
| CN114119591A (en) | Display screen picture quality detection method | |
| CN104166250A (en) | Flat panel display panel uniformity detection method and system | |
| CN101807297B (en) | Medical ultrasonic image line detection method | |
| CN104573743B (en) | A kind of facial image detection filter method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |