CN105303565A - Product appearance LOGO detection method - Google Patents
Product appearance LOGO detection method Download PDFInfo
- Publication number
- CN105303565A CN105303565A CN201510646226.3A CN201510646226A CN105303565A CN 105303565 A CN105303565 A CN 105303565A CN 201510646226 A CN201510646226 A CN 201510646226A CN 105303565 A CN105303565 A CN 105303565A
- Authority
- CN
- China
- Prior art keywords
- logo
- product
- appearance
- preset value
- feature
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a product appearance LOGO detection method. The product appearance LOGO detection method comprises steps that, template contour characteristics of a standard product workpiece are established, processing is further carried out, and a first LOGO area is acquired; contour characteristics of a to-be-detected product workpiece are established to acquire a second LOGO area; whether similarity between the template contour characteristics and the contour characteristics reaches a preset value and a target quantity is satisfied is determined, if yes, a difference area is acquired through processing according to the first LOGO area and the second LOGO area, the total quantity of all pixel points in the difference area is acquired, a radius of the largest inscribed circle of the difference area is acquired, when the total quantity of the pixel points is greater than a preset value and the radius of the largest inscribed circle is greater than a preset internal diameter value, LOGO appearance of the to-be-detected product workpiece is qualified, if not, the LOGO appearance of the to-be-detected product workpiece is not qualified. Through the product appearance LOGO detection method, manual contrast detection on LOGO appearance of products is not necessary, an error caused by manual detection can be avoided, detection efficiency is high, and precision is high.
Description
Technical field
The present invention relates to product appearance vision detection technology, particularly relate to the contrast detection method of product appearance.
Background technology
The product appearance of numerous products such as mobile phone, automobile, air-conditioning, computer is provided with LOGO pattern or printed words often through spraying coating process.LOGO is the English saying of logo or trade mark, product appearance sprays LOGO normally in order to identify company, plays the effect identifying and promote, consumer can be allowed to remember company's main body and brand culture by the LOGO of image.Spraying in production run at the LOGO etc. of product appearance, for guaranteeing coating quality, needing the product to having sprayed to carry out outward appearance detection.At present, the outward appearance detection mode that producer adopts is all desk checking, and desk checking has certain subjectivity usually, cause error large, assay is often uncertain, and it judges the place one's entire reliance upon working experience of workman and subjective consciousness, causes product quality unstable.On the other hand, workman repeats the work of high strength every day, and they accomplish accurately to judge each product and unrealistic to want requirement, therefore, because artificial carelessness probably causes customer complaint even to return goods, the efficiency of desk checking is simultaneously low, needs to take a large amount of manpowers and time.
Summary of the invention
For above-mentioned technical matters, the object of the present invention is to provide a kind of LOGO detection method of product appearance, its accuracy of detection is high, speed is fast.
For achieving the above object, the present invention adopts following technical scheme:
The LOGO detection method of product appearance, comprises the steps:
Step one: template contours feature is set up to standardized product workpiece;
Step 2: carry out binary conversion treatment according to template contours feature and obtain a LOGO region;
Step 3: treat testing product workpiece and set up contour feature;
Step 4: carry out binary conversion treatment according to contour feature and obtain the 2nd LOGO region;
Step 5: whether the similarity of the contour feature of the template contours characteristic sum product workpiece to be detected of criterion product workpiece reaches preset value and meet target number, if so, then performs step 6, otherwise, assert that product workpiece LOGO outward appearance to be detected is defective;
Step 6: obtain diff area according to a LOGO region and the 2nd LOGO regional processing, and obtain the sum of all pixels in diff area, and obtain the radius of maximum inscribed circle of this diff area, when the sum of pixel is greater than pixel preset value and the radius of maximum inscribed circle is greater than internal diameter preset value, then product workpiece LOGO outward appearance to be detected is qualified, otherwise, assert that product workpiece LOGO outward appearance to be detected is defective.
Preferably, step one comprises following sub-step:
Step 1.1: the first gray scale picture obtaining standardized product workpiece, and a region of search is chosen on this first gray scale picture;
Step 1.2: contrast the contrast between pixel in region of search, forms template contours feature when contrast is greater than the first preset value;
Step 1.3: the first angular range change preset template contours feature, obtains a series of template contours feature;
Step 1.4: the template contours feature of preserving series.
Preferred further, described first preset value is 60, and described the first default angular range is ± 10 °.
Preferably, step 3 comprises following sub-step:
Step 3.1: the second gray scale picture obtaining product workpiece to be detected, and a region of search is chosen on this second gray scale picture;
Step 3.2: contrast the contrast between neighbor in region of search, forms contour feature when contrast is greater than the second preset value;
Step 3.3: the second angular range change preset contour feature, obtains a series of contour feature;
Step 3.4: the contour feature preserving series.
Preferred further, described second preset value is 40, and described the second default angular range is ± 10 °.
Preferably, the preset value in described step 5 is 40%, and target number is 1.
Compared to existing technology, beneficial effect of the present invention is:
The present invention carries out contrast without the need to the artificial LOGO outward appearance to product and detects, and avoid the error that manual detection is brought, measuring method efficiency is high, and precision is high, and applicability is strong.
Accompanying drawing explanation
Fig. 1 is the workflow diagram of the LOGO detection method of product appearance of the present invention.
Embodiment
Below, by reference to the accompanying drawings and embodiment, the present invention is described further:
See Fig. 1, the invention provides a kind of LOGO detection method of product appearance, using the LOGO of a standardized product as reference, compare with product to be detected, concrete steps are as follows:
S1: template contours feature is set up to standardized product workpiece;
S2: carry out binary conversion treatment according to template contours feature and obtain a LOGO region; A usual appearance images comprises target object, background etc., want to extract target object from many-valued image, the most frequently used is exactly the binary conversion treatment adopting image, set a threshold value T, with threshold value T, the data of image are divided into two parts, be greater than the pixel group of T and be less than the pixel group of T, exactly the gray-scale value of the pixel on image being set to 0 or 255, namely whole image being presented and significantly only have black and white visual effect;
S3: treat testing product workpiece and set up contour feature;
S4: carry out binary conversion treatment according to contour feature and obtain the 2nd LOGO region;
S5: whether the similarity of the contour feature of the template contours characteristic sum product workpiece to be detected of criterion product workpiece reaches preset value and meet target number, if so, then performs S6, otherwise, assert that product workpiece LOGO outward appearance to be detected is defective; The effect of target number weighs air-conditioning workpiece whether to mate qualified necessary condition, target number is the contour feature number that will find, can be 1, the preset value of both similarities is 40%, namely when the similarity of the contour feature of the template contours characteristic sum air-conditioning workpiece to be detected of standard air-conditioning workpiece reaches 40% and target number meets 1, it is qualified to be, if defective, then selects other product workpiece to restart to measure.
S6: obtain diff area according to a LOGO region and the 2nd LOGO regional processing, and obtain the sum of all pixels in diff area, and obtain the radius of maximum inscribed circle of this diff area, judge whether the sum of pixel is greater than pixel preset value and whether the radius of maximum inscribed circle is greater than internal diameter preset value time, then product workpiece LOGO outward appearance to be detected is qualified, otherwise, assert that product workpiece LOGO outward appearance to be detected is defective.In this step, process obtains the mode of diff area for be subtracted each other in a LOGO region and the 2nd LOGO region, and obtain diff area, the quantity of diff area not necessarily only has one, may have more than one.
Wherein, the process of establishing of step S1 and step S2 is similar, in S1 step, comprises sub-step as follows:
S1.1: the first gray scale picture obtaining standardized product workpiece, and a region of search is chosen on this first gray scale picture; Obtain the first gray scale picture of standardized product workpiece by acquisitions of taking pictures, the gray scale picture of the picture of tentatively taking pictures not necessarily black and white, but subsequent treatment for convenience, gained picture processing of taking pictures the most at last is gray scale picture;
S1.2: contrast the contrast between pixel in region of search, forms template contours feature when contrast is greater than the first preset value; Way of contrast in this step adopts halcon algorithm to carry out contrast level parameter process, when contrast is greater than the first preset value, then forms point, multiple point is connected to form template contours feature; The first preset value about contrast can regulate setting voluntarily according to display screen matrix other actual conditions not equal, and this first preset value in the present embodiment is preferably 60;
S1.3: the first angular range change preset template contours feature, obtains a series of template contours feature; This first angular range preset is ± 10 °; Obtain a series of contour feature to the first angular range change that template contours feature is preset, namely in the certain limit of certain angle value, a series of module profile feature is exactly a lot of profiles, as circle, straight line, circular arc etc.;
S1.4: the template contours feature of preserving series.
S3 step comprises following sub-step:
S3.1: the second gray scale picture obtaining product workpiece to be detected, and a region of search is chosen on this second gray scale picture; Same, the mode obtaining the second gray scale picture obtains by taking pictures;
S3.2: contrast the contrast between neighbor in region of search, forms contour feature when contrast is greater than the second preset value; In this step, the same halcon algorithm that adopts carries out contrast level parameter process, and this second preset value is preferably 40;
S3.3: the second angular range change preset contour feature, obtains a series of contour feature; This second angular range preset is ± 10 °;
S3.4: the contour feature preserving series.
To one skilled in the art, according to technical scheme described above and design, other various corresponding change and deformation can be made, and all these change and deformation all should belong within the protection domain of the claims in the present invention.
Claims (6)
1. the LOGO detection method of product appearance, is characterized in that, comprise the steps:
Step one: template contours feature is set up to standardized product workpiece;
Step 2: carry out binary conversion treatment according to template contours feature and obtain a LOGO region;
Step 3: treat testing product workpiece and set up contour feature;
Step 4: carry out binary conversion treatment according to contour feature and obtain the 2nd LOGO region;
Step 5: whether the similarity of the contour feature of the template contours characteristic sum product workpiece to be detected of criterion product workpiece reaches preset value and meet target number, if so, then performs step 6, otherwise, assert that product workpiece LOGO outward appearance to be detected is defective;
Step 6: obtain diff area according to a LOGO region and the 2nd LOGO regional processing, and obtain the sum of all pixels in diff area, and obtain the radius of maximum inscribed circle of this diff area, when the sum of pixel is greater than pixel preset value and the radius of maximum inscribed circle is greater than internal diameter preset value, then product workpiece LOGO outward appearance to be detected is qualified, otherwise, assert that product workpiece LOGO outward appearance to be detected is defective.
2. the LOGO detection method of product appearance as claimed in claim 1, it is characterized in that, step one comprises following sub-step:
Step 1.1: the first gray scale picture obtaining standardized product workpiece, and a region of search is chosen on this first gray scale picture;
Step 1.2: contrast the contrast between pixel in region of search, forms template contours feature when contrast is greater than the first preset value;
Step 1.3: the first angular range change preset template contours feature, obtains a series of template contours feature;
Step 1.4: the template contours feature of preserving series.
3. the LOGO detection method of product appearance as claimed in claim 2, it is characterized in that, described first preset value is 60, and described the first default angular range is ± 10 °.
4. the LOGO detection method of product appearance as claimed in claim 1, it is characterized in that, step 3 comprises following sub-step:
Step 3.1: the second gray scale picture obtaining product workpiece to be detected, and a region of search is chosen on this second gray scale picture;
Step 3.2: contrast the contrast between neighbor in region of search, forms contour feature when contrast is greater than the second preset value;
Step 3.3: the second angular range change preset contour feature, obtains a series of contour feature;
Step 3.4: the contour feature preserving series.
5. the LOGO detection method of product appearance as claimed in claim 4, it is characterized in that, described second preset value is 40, and described the second default angular range is ± 10 °.
6. the LOGO detection method of product appearance as claimed in claim 1, it is characterized in that, the preset value in described step 5 is 40%, and target number is 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510646226.3A CN105303565A (en) | 2015-09-30 | 2015-09-30 | Product appearance LOGO detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510646226.3A CN105303565A (en) | 2015-09-30 | 2015-09-30 | Product appearance LOGO detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105303565A true CN105303565A (en) | 2016-02-03 |
Family
ID=55200787
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510646226.3A Pending CN105303565A (en) | 2015-09-30 | 2015-09-30 | Product appearance LOGO detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105303565A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106501271A (en) * | 2016-11-24 | 2017-03-15 | 深圳市博视科技有限公司 | product appearance detection method |
| CN107561087A (en) * | 2017-08-31 | 2018-01-09 | 广东工业大学 | A kind of mouse logo positioning and defect inspection method based on machine vision |
| CN108734185A (en) * | 2017-04-18 | 2018-11-02 | 北京京东尚科信息技术有限公司 | Image verification method and apparatus |
| CN109559294A (en) * | 2017-09-26 | 2019-04-02 | 凌云光技术集团有限责任公司 | A kind of detection method and device of drop circular hole quality |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1869667A (en) * | 2006-06-08 | 2006-11-29 | 李贤伟 | Profile analysing method for investigating defect of printed circuit board |
| CN103630070A (en) * | 2013-04-08 | 2014-03-12 | 苏州工业园区凯艺精密科技有限公司 | Detection method for image detector and image detector |
| CN104441526A (en) * | 2014-09-24 | 2015-03-25 | 上海智觉光电科技有限公司 | Online mold monitoring and protecting system and method based on contour matching |
| CN104680509A (en) * | 2013-11-30 | 2015-06-03 | 中国科学院沈阳自动化研究所 | Real-time circular printing image defect detecting method |
-
2015
- 2015-09-30 CN CN201510646226.3A patent/CN105303565A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1869667A (en) * | 2006-06-08 | 2006-11-29 | 李贤伟 | Profile analysing method for investigating defect of printed circuit board |
| CN103630070A (en) * | 2013-04-08 | 2014-03-12 | 苏州工业园区凯艺精密科技有限公司 | Detection method for image detector and image detector |
| CN104680509A (en) * | 2013-11-30 | 2015-06-03 | 中国科学院沈阳自动化研究所 | Real-time circular printing image defect detecting method |
| CN104441526A (en) * | 2014-09-24 | 2015-03-25 | 上海智觉光电科技有限公司 | Online mold monitoring and protecting system and method based on contour matching |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106501271A (en) * | 2016-11-24 | 2017-03-15 | 深圳市博视科技有限公司 | product appearance detection method |
| CN108734185A (en) * | 2017-04-18 | 2018-11-02 | 北京京东尚科信息技术有限公司 | Image verification method and apparatus |
| CN108734185B (en) * | 2017-04-18 | 2022-01-04 | 北京京东尚科信息技术有限公司 | Image verification method and device |
| CN107561087A (en) * | 2017-08-31 | 2018-01-09 | 广东工业大学 | A kind of mouse logo positioning and defect inspection method based on machine vision |
| CN109559294A (en) * | 2017-09-26 | 2019-04-02 | 凌云光技术集团有限责任公司 | A kind of detection method and device of drop circular hole quality |
| CN109559294B (en) * | 2017-09-26 | 2021-01-26 | 凌云光技术股份有限公司 | Method and device for detecting quality of circular hole of drop |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110476056B (en) | Visual detection method, detection equipment and robot | |
| CN106097361B (en) | Defect area detection method and device | |
| CN105424190A (en) | Grayscale detection method of product appearance | |
| CN103885217B (en) | A kind of method detecting liquid crystal panel cylindrical spacer defect and device | |
| JP6325520B2 (en) | Unevenness inspection system, unevenness inspection method, and unevenness inspection program | |
| US9892504B2 (en) | Image inspection method and inspection region setting method | |
| CN105303565A (en) | Product appearance LOGO detection method | |
| KR20190075707A (en) | Method for sorting products using deep learning | |
| CN102253048A (en) | Machine vision detection method and system for detection of various products | |
| CN104458748A (en) | Aluminum profile surface defect detecting method based on machine vision | |
| CN111833303A (en) | Product detection method and device, electronic equipment and storage medium | |
| CN107515481B (en) | Display panel detection method and device | |
| US20140050387A1 (en) | System and Method for Machine Vision Inspection | |
| CN102042812A (en) | Visual machine detection method | |
| CN111310753B (en) | Meter alignment method and device | |
| CN106373124A (en) | Gray-level co-occurrence matrix and RANSAC-based industrial product surface defect visual detection method | |
| CN112381765A (en) | Equipment detection method, device, equipment and storage medium based on artificial intelligence | |
| CN115331132A (en) | Detection method and device for automobile parts, electronic equipment and storage medium | |
| CN106501271A (en) | product appearance detection method | |
| CN105136819A (en) | Method for detecting appearance contour of air conditioner | |
| CN107084989A (en) | The adding method and system in a kind of AOI device datas storehouse | |
| CN111380474A (en) | Aperture detection method based on binocular vision | |
| CN108230285A (en) | Baffle ring quality determining method based on machine vision | |
| CN112819717A (en) | Shaving board image gray value correction method and device | |
| CN110874837A (en) | Automatic defect detection method based on local feature distribution |
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: 20160203 |
|
| RJ01 | Rejection of invention patent application after publication |