CN109540925A - Complicated ceramic tile surface defect inspection method based on difference shadow method and local variance measurement operator - Google Patents
Complicated ceramic tile surface defect inspection method based on difference shadow method and local variance measurement operator Download PDFInfo
- Publication number
- CN109540925A CN109540925A CN201910062683.6A CN201910062683A CN109540925A CN 109540925 A CN109540925 A CN 109540925A CN 201910062683 A CN201910062683 A CN 201910062683A CN 109540925 A CN109540925 A CN 109540925A
- Authority
- CN
- China
- Prior art keywords
- image
- defect
- ceramic tile
- local variance
- tile
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/956—Inspecting patterns on the surface of objects
- G01N21/95623—Inspecting patterns on the surface of objects using a spatial filtering method
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
Landscapes
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses it is a kind of based on difference shadow method and local variance measurement operator complicated ceramic tile surface defect inspection method, this method using based on image calculus of differences and improved local variance measurement operator instead of defect profile extraction by the way of.Firstly, being partitioned into the salient region in image using clustering method to each block of defect ceramic tile；Then, the substantially defect area of image is obtained using difference shadow method；Finally, the profile of defect ceramic tile is extracted by the local variance value for calculating Defect Edge pixel and its surrounding pixel point, it is final to obtain accurate complete defect area after the morphological operations such as smooth and filling.The present invention, which carries out defect inspection method to complicated ceramic tile surface using difference shadow method and improved local variance measurement operator, has the effect of rotational invariance, computation complexity is low, and detection efficiency is high, has preferable robustness, especially also has preferable effect to the identification of low quality ceramic tile defect.
Description
Technical field
The application that the present invention relates to computer vision techniques in field of target recognition, more particularly to one kind based on poor shadow
The complicated ceramic tile surface defect inspection method of method and local variance measurement operator.
Background technique
Surface defects detection technology based on computer vision is a very important link in automatic detection, wide
The general surface defects detection for being applied to the products such as glass, ceramic tile, steel ball, fabric, timber, is a kind of very convenient reliable detection
Method, compared at high cost, the low efficiency of human eye detection, automatic detection is since stability is strong, it is fast, at low cost etc. to detect speed
Advantage, thus be widely used in defects detection field.For the defects detection of ceramic tile surface, it is generally divided into simple brick surface
Defects detection and complicated brick surface defects detection, simple brick surface disturbance is small, and detection is easier；And complicated brick surface is by spending
Line, background and defect composition, surface disturbance is big, moreover due to ceramic tile manufacturing process complexity, so that the surface of ceramic tile generates
Subtle, local material defect, these defects can substantially be divided into scratch, spot, fall dirty and four class of scatterplot, be the table of ceramic tile
Planar defect detection increases difficulty.Forefathers are mostly based on simple ceramic tile surface defect and detect, and detection method is single, due to porcelain
The diversity on brick surface can not be suitable for the detection of all ceramic tiles.
The two images that difference shadow method is directed under same background do calculus of differences, by the area for extracting different gray values in image
Domain, achievees the purpose that detection, and existing research personnel are applied in surface defects detection.Zhang Jun et al. using wavelet transformation with
The difference method of Morphological Fusion extracts the edge and background information of defect.Wang Yiwen et al. proposes that a kind of circular contour is extraneous
Rectangle and difference method realize the detection to steel ball surface defect.Hanzaei et al. proposes one kind and converts the image into two
System matrix carries out calculus of differences, then extracts to the fringe region of defect, although achieving good results, this
Method is only applied to the detection of simple ceramic tile surface, and narrow scope of application has certain limitation.
Local variance measurement operator is that the image outline that is applied to improved to local binary pattern (LBP) extracts
A kind of algorithm has the advantages that rotational invariance and to illumination-insensitive.It is applied to the detection of ceramic tile surface defect, not only
Calculating cost is reduced, and reduces calculation amount.By calculating between image deflects edge pixel point and its surrounding pixel point
Variance yields extract defect profile, to the contour area extracted by calculating the prominent defect of local variance Weighted information entropy
Fringe region preferably remains the detailed information of defect, strengthens visual effect, while also avoiding due to manual setting side
Error caused by edge threshold value reduces the false detection rate of ceramic tile.
Summary of the invention
In order to overcome the deficiencies in the prior art, the purpose of the present invention is to provide one kind to be based on difference shadow method and part
Variance measures the complicated ceramic tile surface defect inspection method of operator, to establish a kind of effective, accurate and stable ceramic tile surface
Defect inspection method.
Technical solution: to achieve the above object, the technical solution adopted by the present invention are as follows:
A kind of complicated ceramic tile surface defect inspection method based on difference shadow method and local variance measurement operator, feature exist
In sequence executes following steps:
Step 1) image preprocessing: input standard ceramic tile I and tile image D to be detected, using median filtering technology to it
The noise spot that pretreatment operation filters out imaging surface is carried out, the quality of image is improved, obtaining pretreated image is I1 and D1；
Step 2) k-means cluster segmentation: carrying out Clustering to the image-region of pretreated image I1 and D1, will
It is divided into the foreground and background that significant part includes image, completes the segmentation to tile image, obtains standard tile image
With salient region the image I2 and D2 in tile image to be detected；
Step 3) difference shadow method: by the standard tile image I2 and tile image to be detected after pretreatment and image segmentation
Calculus of differences is done between D2, the error image that the pixel value of their respective coordinates is subtracted each other is ceramic tile defect image G；
Step 4) local variance measures operator: for the edge picture of obtained every width ceramic tile defect image G in step 3)
Vegetarian refreshments calculates the local variance of the pixel Yu its surrounding pixel point, carries out smooth and padding to obtained defect image,
To obtain final complete ceramic tile defect map.
As the difference shadow method in the preferred step 3) two images are carried out with formula when calculus of differences are as follows:
Wherein, G (u, v) is the defect image after calculus of differences, and I (x, y) is standard picture, and D (x', y') is defect map
Picture, u, v, x, y, what x', y' were then represented is the corresponding coordinate position of each image, and the pixel value of their respective coordinates is made the difference
Point, it is identical to be then labeled as " 0 ", otherwise it is labeled as " 1 ".
As in the preferred step 4) to obtained every width ceramic tile defect image G in step 3), using based on part
Variance weighted comentropy extracts the profile of ceramic tile defect.First pass through formula (2) calculate local window G in pixel with
The local variance of its surrounding pixel point, obtains the defect area of ceramic tile, calculates the local variance formula of image are as follows:
Wherein
Wherein P is the number of neighbor point, and R is the radius of neighbourhood, and local variance measures operator by calculating defect surrounding pixel
The variance yields of point extracts its profile, avoids error caused by manual setting profile threshold value, but uses this method to lacking
The precision of images that sunken extraction obtains is lower, so also needing to correct using the local variance Weighted information entropy of image-region, formula
It is as follows:
N is the sum of all pixels of local window, and k is that r (1≤r≤n) kind that the defects of local window image G contains is different
Pixel grey scale (1≤k≤r), PkThe probability occurred by different pixels gray scale.The part side of image is calculated by formula (3)
Poor Weighted information entropy remains the detailed information of defect image, more fully reflects the information content of defect image, improves scarce
The precision for falling into image, reduces the labour cost of factory.
The utility model has the advantages that the present invention provides a kind of complicated ceramic tile surface defect based on difference shadow method and local variance measurement operator
Detection method has rotational invariance and stronger robustness, keeps the accuracy rate of ceramic tile surface defects detection higher, pass through part
The method that variance weighted comentropy extracts its defect profile is simple and efficient, and can reach detection by the detection of individual ceramic tile
Purpose reduces the cost of false detection rate and factory, improves detection efficiency, can also reach more to the identification of low quality tile image
Good effect.
Detailed description of the invention
Fig. 1 is the complicated ceramic tile surface defect inspection method of the invention that operator is measured based on difference shadow method and local variance
Flow chart；
Fig. 2 is the image schematic diagram after original piece image and defect tile image of the invention are preprocessed；
Wherein upper row picture a) is original image, lower row picture b) is that treated image；
Fig. 3 is the schematic diagram of the invention used after k-means Threshold segmentation to pretreated tile image；
Wherein left figure a) is standard picture, and right figure b) is defect image；
Fig. 4 is the ceramic tile surface defect inspection method using of the invention based on difference shadow method and local variance measurement operator
Defect effect picture；
Fig. 5 is the effect picture of the invention using after morphological operation；
Fig. 6 is the comparison result of method and existing defect inspection method of the invention；
Fig. 7 is the picture library example in the practical example 1 of the present invention.
Specific embodiment
The present invention will be further explained with reference to the accompanying drawing
As shown in Figure 1, a kind of complicated ceramic tile surface defect based on difference shadow method and local variance measurement operator of the invention
Detection method, comprising the following steps:
Step 1) image preprocessing: in Fig. 2 a) figure be input original piece image, including standard tile image I and lack
Tile image D is fallen into, the noise spot that pretreatment operation filters out imaging surface is carried out to it using median filtering technology, improves image
Quality, while also preferably retaining the marginal portion of image, pretreated image is obtained as shown in b) figure, respectively standard porcelain
Brick image I1 and defect tile image D1；
Step 2) k-means cluster segmentation: to the image of standard tile image I1 after pretreatment and defect tile image D1
Region carries out Clustering, is divided into significant part, because the present invention relates to the defects detection of complicated ceramic tile surface, institute
To include the foreground and background region of image, by being clustered to image pixel, the segmentation to tile image is completed, porcelain is obtained
Salient region image I2 and D2 in brick image；As shown in Figure 3；
Step 3) difference shadow method: by the standard tile image I2 and defect tile image D2 after pretreatment and image segmentation
Between do calculus of differences, the difference value image that the pixel value of their respective coordinates is subtracted each other be defect image G, process can
It is indicated with formula (1):
Wherein, G (u, v) is the defect image after calculus of differences, and I (x, y) is standard picture, and D (x', y') is defect map
Picture, u, v, x, y, what x', y' were then represented is the corresponding coordinate position of each image, and the pixel value of their respective coordinates is made the difference
Point, it is identical to be then labeled as " 0 ", otherwise it is labeled as " 1 ".For the surface defects detection of complicated tile image, due to surface
Decorative pattern is complicated, interfere it is larger, detection get up it is relatively difficult, so the obtained testing result of difference shadow method further includes some non-defective
Region is not completely defect image, also further to extract to the profile of defect to obtain complete defect area
Domain.Other methods are compared to, the method that difference shadow method operates the ceramic tile under same background reduces ceramic tile due to background
Error brought by the gray value difference of point, simultaneously because it is calculated simply, thus is applied to the detection efficiency of ceramic tile.This hair
It is bright be directed to 18 groups of variety classes, 4 kinds of different defects tile image in one group of tile image detected；
Step 4) local variance measures operator: for obtained every width ceramic tile defect image G in step 3), the present invention is adopted
It is extracted with based on profile of the local variance Weighted information entropy to ceramic tile defect.The pixel in local window G is calculated first
With the local variance of its surrounding pixel point, the formula of local variance detection are as follows:
Wherein
Wherein P is the number of neighbor point, and R is the radius of neighbourhood, gPFor the gray value of neighborhood territory pixel, VARP, RThere is continuous value
And do not change with the variation of gray scale.This method can obtain preferable defect image profile, avoid by manual setting threshold value institute
Bring error, but it is lower using the precision of images that this method extracts defect, so in order to make the defect extracted
Profile is more clear accurately, therefore certain amendment is carried out to Weighted information entropy with local variance, can qualitatively characterize partial zones
The clarity of defect profile in domain under different grey-scale retains the detailed information of defect.Office is used to the defect image of acquirement
Portion's variance weighted comentropy formula are as follows:
If n is the sum of all pixels of local window, k is that the r (1≤r≤n) that the defects of local window image G contains is planted not
Same pixel grey scale (1≤k≤r), PkIt is the probability that different pixels gray scale occurs.By the local variance weighting for calculating image
Comentropy remains the detailed information of defect image, more fully reflects the information content of defect image as shown in figure 4, improving
The precision of defect image reduces the false detection rate of ceramic tile and the cost of factory.
For the radius of neighbourhood R in formula (2), when calculating the local variance value of image, selected radius of neighbourhood value is different
Fixed, then the accuracy value, detection time and false detection rate for detecting defect area are all not necessarily, so choosing after experiment optimal
Radius of neighbourhood value R=7 be applied to the extraction of ceramic tile Defect Edge profile, experimental result is as shown in table 1.Then, pass through formula
(3) information content for increasing defect image remains the detailed information of defect, has in terms of ceramic tile defects detection and mentions significantly
It is high.Finally, carrying out smooth and padding to obtained defect image, obtained to obtain final complete ceramic tile defect map
Preferable effect is as shown in Figure 5；
Table 1
Embodiment 1:
The embodiment of the present invention uses the tile image library shot under laboratory environment.It altogether include 18 groups of different ceramic tiles
Different classes of defect in image and 4.Every width tile image general 25 or so total 100 width gray level images compositions, image ruler
Very little is 316 × 318.In order to verify the present invention to the characterization effect of ceramic tile surface defects detection, it is carried out with existing algorithm
Compare, effect as shown in fig. 6, our experiments show that, the method applied in the present invention verification and measurement ratio obviously than single k-means, LBP calculate
The result accuracy rate of method is high, and false detection rate is small, time-consuming short.Experimental result is as shown in table 2:
Table 2
Claims (3)
1. a kind of complicated ceramic tile surface defect inspection method based on difference shadow method and local variance measurement operator, which is characterized in that
Sequence executes following steps:
Step 1) image preprocessing: input standard tile image I and tile image D to be detected, using median filtering technology to it
The noise spot that pretreatment operation filters out imaging surface is carried out, the quality of image is improved, obtaining pretreated image is I1 and D1；
Step 2) k-means cluster segmentation: Clustering is carried out to the image-region of pretreated image I1 and D1, by its point
Include foreground and background two parts of image at significant part, completes the segmentation to tile image, obtain standard ceramic tile figure
Salient region image I2 and D2 in picture and tile image to be detected；
Step 3) difference shadow method: by through pretreatment and image segmentation after standard tile image I2 and tile image D2 to be detected it
Between do calculus of differences, the error image that the pixel value of their respective coordinates is subtracted each other be ceramic tile defect image G；
Step 4) local variance measures operator: it is directed to the edge pixel point of obtained every width ceramic tile defect image G in step 3),
The local variance for calculating the pixel Yu its surrounding pixel point carries out smooth and padding to obtained defect image, thus
Obtain final complete ceramic tile defect map.
2. a kind of complicated ceramic tile surface defect based on difference shadow method and local variance measurement operator according to claims
Detection method, it is characterised in that: the difference shadow method in the step 3) carries out formula when calculus of differences to two images are as follows:
Wherein, G (u, v) is the defect image after calculus of differences, and I (x, y) is standard picture, and D (x', y') is defect image,
What u, v, x, y, x', y' were then represented is the corresponding coordinate position of each image, and the pixel value of their respective coordinates is done difference,
It is identical to be then labeled as " 0 ", otherwise it is labeled as " 1 ".
3. a kind of complicated ceramic tile surface defect based on difference shadow method and local variance measurement operator according to claims
Detection method, it is characterised in that: to obtained every width ceramic tile defect image G in step 3) in the step 4), using being based on
Local variance Weighted information entropy extracts the profile of ceramic tile defect, specifically: it first passes through formula (2) and calculates local window G
The local variance of interior pixel and its surrounding pixel point, obtains the defect area of ceramic tile, calculates the local variance formula of image
Are as follows:
Wherein P is the number of neighbor point, and R is the radius of neighbourhood；
It is modified again by the local variance Weighted information entropy that formula (3) (4) calculate image-region, qualitatively characterizes partial zones
The clarity of defect profile in domain under different grey-scale retains the detailed information of defect, and formula is as follows:
N is the sum of all pixels of local window, and k is that the r (1≤r≤n) that the defects of local window image G contains plants different pictures
Plain gray scale (1≤k≤r), PkThe probability occurred by different pixels gray scale.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910062683.6A CN109540925B (en) | 2019-01-23 | 2019-01-23 | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910062683.6A CN109540925B (en) | 2019-01-23 | 2019-01-23 | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109540925A true CN109540925A (en) | 2019-03-29 |
| CN109540925B CN109540925B (en) | 2021-09-03 |
Family
ID=65838124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910062683.6A Active CN109540925B (en) | 2019-01-23 | 2019-01-23 | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109540925B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110349133A (en) * | 2019-06-25 | 2019-10-18 | 杭州汇萃智能科技有限公司 | Body surface defect inspection method, device |
| CN112419228A (en) * | 2020-10-14 | 2021-02-26 | 惠州高视科技有限公司 | Method and device for detecting three-dimensional edge defect of cover plate |
| CN112819812A (en) * | 2021-02-25 | 2021-05-18 | 西安铂力特增材技术股份有限公司 | Powder bed defect detection method based on image processing |
| CN113554604A (en) * | 2021-07-01 | 2021-10-26 | 常州大学 | Melt-blown fabric defect area detection method based on machine vision |
| CN114782418A (en) * | 2022-06-16 | 2022-07-22 | 深圳市信润富联数字科技有限公司 | Detection method and device for ceramic tile surface defects and storage medium |
| CN115082482A (en) * | 2022-08-23 | 2022-09-20 | 山东优奭趸泵业科技有限公司 | Metal surface defect detection method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968790A (en) * | 2012-10-25 | 2013-03-13 | 西安电子科技大学 | Remote sensing image change detection method based on image fusion |
| CN204422436U (en) * | 2015-03-02 | 2015-06-24 | 三峡大学 | A kind of ceramic tile patterns defect detecting device based on machine vision |
| CN107452010A (en) * | 2017-07-31 | 2017-12-08 | 中国科学院长春光学精密机械与物理研究所 | A kind of automatically stingy nomography and device |
| CN108010047A (en) * | 2017-11-23 | 2018-05-08 | 南京理工大学 | A kind of moving target detecting method of combination unanimity of samples and local binary patterns |
-
2019
- 2019-01-23 CN CN201910062683.6A patent/CN109540925B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968790A (en) * | 2012-10-25 | 2013-03-13 | 西安电子科技大学 | Remote sensing image change detection method based on image fusion |
| CN204422436U (en) * | 2015-03-02 | 2015-06-24 | 三峡大学 | A kind of ceramic tile patterns defect detecting device based on machine vision |
| CN107452010A (en) * | 2017-07-31 | 2017-12-08 | 中国科学院长春光学精密机械与物理研究所 | A kind of automatically stingy nomography and device |
| CN108010047A (en) * | 2017-11-23 | 2018-05-08 | 南京理工大学 | A kind of moving target detecting method of combination unanimity of samples and local binary patterns |
Non-Patent Citations (3)
| Title |
|---|
| 王义文 等: "钢球表面缺陷的图像差分检测算法", 《计算机辅助设计与图形学学报》 * |
| 郭明 等: "基于区域和方向方差加权信息熵的图像融合", 《系统工程与电子技术》 * |
| 高倩倩: "瓷砖表面质量视觉检测技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110349133A (en) * | 2019-06-25 | 2019-10-18 | 杭州汇萃智能科技有限公司 | Body surface defect inspection method, device |
| CN110349133B (en) * | 2019-06-25 | 2021-11-23 | 杭州汇萃智能科技有限公司 | Object surface defect detection method and device |
| CN112419228A (en) * | 2020-10-14 | 2021-02-26 | 惠州高视科技有限公司 | Method and device for detecting three-dimensional edge defect of cover plate |
| CN112419228B (en) * | 2020-10-14 | 2022-04-05 | 高视科技(苏州)有限公司 | Method and device for detecting three-dimensional edge defect of cover plate |
| CN112819812A (en) * | 2021-02-25 | 2021-05-18 | 西安铂力特增材技术股份有限公司 | Powder bed defect detection method based on image processing |
| CN112819812B (en) * | 2021-02-25 | 2024-05-31 | 西安铂力特增材技术股份有限公司 | Powder bed defect detection method based on image processing |
| CN113554604A (en) * | 2021-07-01 | 2021-10-26 | 常州大学 | Melt-blown fabric defect area detection method based on machine vision |
| CN113554604B (en) * | 2021-07-01 | 2024-02-02 | 常州大学 | Melt-blown cloth defect area detection method based on machine vision |
| CN114782418A (en) * | 2022-06-16 | 2022-07-22 | 深圳市信润富联数字科技有限公司 | Detection method and device for ceramic tile surface defects and storage medium |
| CN114782418B (en) * | 2022-06-16 | 2022-09-16 | 深圳市信润富联数字科技有限公司 | Detection method and device for ceramic tile surface defects and storage medium |
| CN115082482A (en) * | 2022-08-23 | 2022-09-20 | 山东优奭趸泵业科技有限公司 | Metal surface defect detection method |
| CN115082482B (en) * | 2022-08-23 | 2022-11-22 | 山东优奭趸泵业科技有限公司 | Metal surface defect detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109540925B (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109540925A (en) | Complicated ceramic tile surface defect inspection method based on difference shadow method and local variance measurement operator | |
| CN104794491B (en) | Based on the fuzzy clustering Surface Defects in Steel Plate detection method presorted | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN108921819B (en) | Cloth inspecting device and method based on machine vision | |
| CN109816652A (en) | A kind of intricate casting defect identification method based on gray scale conspicuousness | |
| CN112651968B (en) | Wood board deformation and pit detection method based on depth information | |
| CN106600600A (en) | Wafer defect detection method based on characteristic matching | |
| CN115082419A (en) | Blow-molded luggage production defect detection method | |
| CN106529559A (en) | Pointer-type circular multi-dashboard real-time reading identification method | |
| CN105334219A (en) | Bottleneck defect detection method adopting residual analysis and dynamic threshold segmentation | |
| CN109087286A (en) | A kind of detection method and application based on Computer Image Processing and pattern-recognition | |
| CN107230203A (en) | Casting defect recognition methods based on human eye vision attention mechanism | |
| CN103870833A (en) | Method for extracting and evaluating pavement crack based on concavity measurement | |
| CN109409290B (en) | Thermometer verification reading automatic identification system and method | |
| CN110334760B (en) | Optical component damage detection method and system based on RESUnet | |
| CN109359604B (en) | Method for identifying instrument under shadow interference facing inspection robot | |
| CN113643276B (en) | Textile texture defect automatic detection method based on statistical analysis | |
| CN108802052A (en) | A kind of detecting system and its detection method about slide fastener defect | |
| CN116563279B (en) | Measuring switch detection method based on computer vision | |
| CN108537787A (en) | A kind of quality judging method of facial image | |
| CN115100206A (en) | Printing defect identification method for textile with periodic pattern | |
| CN110458809B (en) | Yarn evenness detection method based on sub-pixel edge detection | |
| CN108492306A (en) | A kind of X-type Angular Point Extracting Method based on image outline | |
| CN112085699A (en) | Pavement crack extraction method based on two-dimensional image | |
| CN118037730A (en) | Crack length detection system based on computer image processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221021 Address after: No. 3, Tonghuan Road, Tongsha, Dongcheng Street, Dongguan City, Guangdong Province, 523000 Patentee after: Guangdong Jiyang Vision Technology Co.,Ltd. Address before: 330063 No. 696 Feng Nan Road, Jiangxi, Nanchang Patentee before: NANCHANG HANGKONG University |
|
| TR01 | Transfer of patent right |