CN103345750A - Battery tail end peeling detection method - Google Patents
Battery tail end peeling detection method Download PDFInfo
- Publication number
- CN103345750A CN103345750A CN2013102676461A CN201310267646A CN103345750A CN 103345750 A CN103345750 A CN 103345750A CN 2013102676461 A CN2013102676461 A CN 2013102676461A CN 201310267646 A CN201310267646 A CN 201310267646A CN 103345750 A CN103345750 A CN 103345750A
- Authority
- CN
- China
- Prior art keywords
- picture
- tail end
- value
- pixel
- decortication
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a battery tail end peeling detection method. Binarization is carried out on an original picture to obtain I1, certain noise contained in the I1 is set to be 255 in the binarization process, small-area communication areas are removed for processing, and therefore most noise points can be removed. When the closed operation is carried out, the value of the radius R can guarantee the given accuracy. When the number of communication areas in I3 is counted, the bigger the K value is, the larger the number of obvious breaking points in the original picture is. On the premise that the battery tail end peeling can be judged, complicated spectral analysis is not adopted, the noise points can be removed through a morphological method, therefore, the aim of judging scars can be achieved, calculation amount is greatly reduced, and the efficiency is improved.
Description
Technical field
The present invention relates to a kind of detection method of battery, especially relate to a kind of battery tail end decortication detection method.
Background technology
Field of machine vision is an emerging field in recent years, but its application has been diffused into all trades and professions, and caters to current production automation megatrend especially.Because machine replaces manually having irreplaceable advantage, as: speed is fast, the precision height, and stream time is long, and can be operated in some intolerable environment.Machine vision one of widespread use in production automation field is that line detects, replace human eye with camera, computer replaces human brain, carry out various detections, there has been at present the field of application to comprise that the capsule surface scar of pharmaceuticals industry detects, the scar that bottle is made in the industry detects, in the battery production to detection of battery tail end scar etc.In the battery production field, scar detection method for the battery bottom is human eye detection mostly at present, at first the duplication of labour expends a large amount of manpower and materials, secondly efficient and precision are subjected to considerable influence, and the application that machine vision detects at production line for manufacturing battery will help to reduce resource cost and enhance productivity.In the testing process to the battery tail end, a kind of scar is arranged decortication, its typical construction is that the iron sheet of battery tail end outer ring exists small wearing and tearing or comes off, because it is very trickle, be difficult to detect after battery tail end picture is taken, this can influence the precision of detection system and follow-up scar kind is judged.Therefore need a kind of method that can judge whether to have the battery of decortication scar fast of development.Wherein, how to give prominence to decortication scar and extraction useful information and help to judge to be the difficult point place.In the present online detection range, for the judgement of the small scar of this class of decortication, be that picture is carried out spectrum analysis, analyze the specific frequency spectrum of scar sample, the method for obtaining a result.There is drawback in frequency spectrum analysis method: can be because little scar and noise (" noise " means because of some disturbing factors of taking the former of hardware thereby producing at picture) can't well be distinguished and make differentiation precision not enough herein
Summary of the invention
Technical matters to be solved by this invention provides a kind of simple and efficient, the accurate battery tail end decortication of judgement detection method
The present invention solves the problems of the technologies described above the technical scheme that adopts: a kind of battery tail end decortication detection method is characterized in that comprising that concrete steps are as follows:
S1: correlation parameter:
Pending picture I pixel is M * N; Grey scale pixel value is 0 to 255 in the picture; Definition: area is pixel number that comprises with the gray-scale value zone in the picture, and area threshold is x; Connected region S is that the grey scale pixel value of a panel region in the picture is 255, and all pixels of this panel region adjacency are 0; Radius is n pixel number sum that is in a straight line continuous in the picture; Structural element is for carrying out the set of pixels of a given shape of closed operation to picture;
S2: utilize maximum variance between clusters that I is carried out binaryzation and obtain two-value picture I
1, traversal I
1All connected regions all put 0 with area less than 255 pixel regions below the x, obtain picture I
2
S3: be 5 the I of circular configuration element R with radius
2In after all connected regions carry out closed operation, obtain two-value picture I
3, statistics I
3The number K of middle connected region if K, judges then that the geometric properties that picture shows is the tail end decortication greater than 10, otherwise is judged to be non-decortication.
Compared with prior art, advantage of the present invention is former picture to be carried out binaryzation obtain I
1, I
1In some noises of comprising in the process of binaryzation, can be put 255, by the small size connected region is removed processing, can remove most of noise; When carrying out closed operation, the radius value of R can guarantee set precision; Statistics I
3In the connected region number time, the K value is more big, illustrates that then the obvious breakpoint of former picture is more many.Under the prerequisite that can judge the decortication of battery tail end, do not adopt complicated spectrum analysis, and adopt morphological method to eliminate noise, reach the purpose of differentiating scar, significantly reduced operand, improved efficient.
Description of drawings
Fig. 1 is the logical organization block diagram to battery tail end decortication picture testing process of the embodiment of the invention.
Embodiment
Describe in further detail below in conjunction with the present invention of accompanying drawing embodiment.
Embodiment one: a kind of battery tail end decortication detection method specifically comprises following concrete steps:
S1: correlation parameter:
The pixel of pending picture I is 912 * 912; Grey scale pixel value is 0 to 255 in the picture; Definition: area is pixel number that comprises with the gray-scale value zone in the picture; Connected region S is that the grey scale pixel value of a panel region in the picture is 255, and all pixels of this panel region adjacency are 0; Radius is the pixel number sum that 5 are in a straight line continuous in the picture; Structural element is for carrying out the set of pixels of a given shape of closed operation to picture; Area threshold is 30;
S2: utilize maximum variance between clusters that I is carried out binaryzation and obtain two-value picture I
1, traversal I
1All connected regions all put 0 with area less than 255 pixel regions below 30, obtain picture I
2
S3: be 5 the I of circular configuration element R with radius
2In after all connected regions carry out closed operation, obtain binary map I
3, statistics I
3The number K of middle connected region, K=19 judges that then battery is the tail end decortication.
Embodiment two: a kind of battery tail end decortication detection method specifically comprises following concrete steps:
S1: correlation parameter:
The pixel of pending picture I is 912 * 910; Grey scale pixel value is 0 to 255 in the picture; Definition: area is pixel number that comprises with the gray-scale value zone in the picture; Connected region S is that the grey scale pixel value of a panel region in the picture is 255, and all pixels of this panel region adjacency are 0; Radius is the pixel number sum that 5 are in a straight line continuous in the picture; Structural element is for carrying out the set of pixels of a given shape of closed operation to picture; Area threshold is 30;
S2: utilize maximum variance between clusters that I is carried out binaryzation and obtain two-value picture I
1, traversal I
1All connected regions all put 0 with area less than 255 pixel regions below 30, obtain picture I
2
S3: be 5 the I of circular configuration element R with radius
2In after all connected regions carry out closed operation, obtain binary map I
3, statistics I
3The number K of middle connected region, K=4 judges that then the geometric properties that picture shows is non-tail end decortication.
Claims (1)
1. battery tail end decortication detection method is characterized in that comprising that concrete steps are as follows:
S1: correlation parameter:
Pending picture I pixel is M * N; Grey scale pixel value is 0 to 255 in the picture; Definition: area is pixel number that comprises with the gray-scale value zone in the picture, and area threshold is x; Connected region S is that the grey scale pixel value of a panel region in the picture is 255, and all pixels of this panel region adjacency are 0; Radius is n pixel number sum that is in a straight line continuous in the picture; Structural element is for carrying out the set of pixels of a given shape of closed operation to picture;
S2: utilize maximum variance between clusters that I is carried out binaryzation and obtain two-value picture I
1, traversal I
1All connected regions all put 0 with area less than 255 pixel regions below the x, obtain picture I
2
S3: be 5 the I of circular configuration element R with radius
2In after all connected regions carry out closed operation, obtain two-value picture I
3, statistics I
3The number K of middle connected region if K, judges then that the geometric properties that picture shows is the tail end decortication greater than 10, otherwise is judged to be non-decortication.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013102676461A CN103345750A (en) | 2013-06-28 | 2013-06-28 | Battery tail end peeling detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013102676461A CN103345750A (en) | 2013-06-28 | 2013-06-28 | Battery tail end peeling detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103345750A true CN103345750A (en) | 2013-10-09 |
Family
ID=49280543
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013102676461A Pending CN103345750A (en) | 2013-06-28 | 2013-06-28 | Battery tail end peeling detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103345750A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101527044A (en) * | 2009-03-16 | 2009-09-09 | 江苏银河电子股份有限公司 | Automatic segmenting and tracking method of multiple-video moving target |
CN101976437A (en) * | 2010-09-29 | 2011-02-16 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
KR20120137158A (en) * | 2011-06-10 | 2012-12-20 | 성균관대학교산학협력단 | A hybrid segmentation method of x-ray ct images |
-
2013
- 2013-06-28 CN CN2013102676461A patent/CN103345750A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101527044A (en) * | 2009-03-16 | 2009-09-09 | 江苏银河电子股份有限公司 | Automatic segmenting and tracking method of multiple-video moving target |
CN101976437A (en) * | 2010-09-29 | 2011-02-16 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
KR20120137158A (en) * | 2011-06-10 | 2012-12-20 | 성균관대학교산학협력단 | A hybrid segmentation method of x-ray ct images |
Non-Patent Citations (2)
Title |
---|
姚文伟: "印刷电路板缺陷检测技术及系统实现研究", 《中国优秀硕士学位论文库》 * |
王磊: "基于机器视觉的电池表面缺陷检测技术研究", 《中国硕士学位论文数据库》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107742286B (en) | Method for detecting EL test crack defects of polycrystalline silicon solar cell | |
CN105891233B (en) | Lens surface defect intelligent checking system and its implementation based on machine vision | |
CN101403602B (en) | Embedded high-speed on-line machine vision detection method and device | |
Tonti et al. | An automated approach to the segmentation of HEp-2 cells for the indirect immunofluorescence ANA test | |
CN105447851A (en) | Glass panel sound hole defect detection method and system | |
CN105069447B (en) | A kind of recognition methods of human face expression | |
CN104408449A (en) | Intelligent mobile terminal scene character processing method | |
CN112837290A (en) | Crack image automatic identification method based on seed filling algorithm | |
CN103065122A (en) | Facial expression recognition method based on facial motion unit combination features | |
CN113674226A (en) | Tea leaf picking machine tea leaf bud tip detection method based on deep learning | |
CN103345632A (en) | Method for detecting surface scratch defects of battery tail end | |
CN109086772A (en) | A kind of recognition methods and system distorting adhesion character picture validation code | |
CN112560941A (en) | Coal and gangue identification method based on image detection | |
CN104048966B (en) | The detection of a kind of fabric defect based on big law and sorting technique | |
CN108776823A (en) | Cervical carcinoma lesion analysis method based on cell image recognition | |
CN112200813A (en) | Coal and gangue identification method and system considering illumination factor | |
CN103091330A (en) | Copper strip surface defect identifying device for simulating human visual perception mechanism and method thereof | |
Lin et al. | Surface defect detection of machined parts based on machining texture direction | |
CN107516315A (en) | A kind of development machine based on machine vision is slagged tap monitoring method | |
Li et al. | A method of surface defect detection of irregular industrial products based on machine vision | |
CN110599232A (en) | Consumption group analysis method based on big data | |
CN109858501A (en) | A kind of two phase flow pattern feature extracting method | |
CN102313740B (en) | Solar panel crack detection method | |
Wu et al. | Automatic gear sorting system based on monocular vision | |
Sidehabi et al. | The Development of Machine Vision System for Sorting Passion Fruit using Multi-Class Support Vector Machine. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131009 |
|
RJ01 | Rejection of invention patent application after publication |