CN103456021A - Piece goods blemish detecting method based on morphological analysis - Google Patents
Piece goods blemish detecting method based on morphological analysis Download PDFInfo
- Publication number
- CN103456021A CN103456021A CN2013104377479A CN201310437747A CN103456021A CN 103456021 A CN103456021 A CN 103456021A CN 2013104377479 A CN2013104377479 A CN 2013104377479A CN 201310437747 A CN201310437747 A CN 201310437747A CN 103456021 A CN103456021 A CN 103456021A
- Authority
- CN
- China
- Prior art keywords
- flaw
- image
- value
- piece goods
- blemish
- 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
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Treatment Of Fiber Materials (AREA)
Abstract
The invention discloses a piece goods blemish detecting method based on morphological analysis. The piece goods blemish detecting method comprises the steps that a line-scan digital camera is used for conducting image collection on piece goods in motion, the distance of adjacent tissue points of the piece goods is obtained by conducting basic brightness stretching and binaryzation processing on gray level images of the piece goods, and the distance serves as the reference value of the size of structural elements; then, morphological close operation is conducted, filling is conducted on common weaving point gaps, the number of the weaving points at a blemish position is small and the weaving points at the blemish position are irregular, so that the weaving points stand out in brightness; at least, binaryzation processing is conducted on the images, determining of the threshold of the images is controlled through gamma training selecting of gray level power transform, blemished binary images are counted and marked, blemish information is output, and the blemish detecting process of the piece goods is completed. The piece goods blemish detecting method based on morphological analysis is more accurate and simpler in structure, and improves the calculation speed; can weaken the interference of background brightness unevenness; brings convenience to realization of automatic detection.
Description
Technical field
The present invention relates to a kind of detection method of fabric, be specifically related to a kind of flaw detection method of cloth, especially a kind of Fabric Defect detection method based on morphological analysis.
Background technology
In the textile production process, quality control and inspection is very important, and it is wherein particularly crucial ingredient that Fabric Defect detects.At present, the detection mode of most of textile enterprise is still manual detection, this mode detection speed is low, the impact of easy examined member's subjective factor, flase drop, undetected situation have generation more, and domestic labor cost is more and more higher, thereby manual detection more and more can not meet the requirement of Modern Manufacturing Enterprise.Replace manual detection with advanced Automatic Measurement Technique, can improve well detection efficiency, reduce the labour, reduce labour intensity and further improve the quality of cloth.
At present, have on market that to take the Cyclops perching system of Belgian Barco company, the I-TEX2000 perching system of Israel Evs company, the Fabriscan cloth quality automatic checkout system of Switzerland Uster company etc. be main cloth detection system.The most price of the said equipment is very expensive, and domestic textile enterprise's majority is difficult to bear.
In prior art, the Fabric Defect detection method roughly is divided into 3 classes: based on statistic law, based on modelling with based on Zymography.Comparatively classical method has: the gray scale symbiosis square method based on statistics, Fourier transformation method, small wave converting method, Gabor filtered method and the markov random file method based on model based on analysis of spectrum.Wherein, detection algorithm commonly used be based on analysis of spectrum method or with the combination of additive method, these methods, by the conversion of spatial domain and frequency domain, are extracted correlated characteristic information, but the method calculated amount that relates to frequency domain is larger, changes back and forth the accuracy that makes partial information between territory and be difficult to guarantee.
Along with the development of line-scan digital camera, the high speed acquisition high-definition picture becomes possibility, adds the advantage of the intuitive of spatial domain, and this provides the foundation for some new flaw detection methods based on spatial domain of development.
Summary of the invention
Goal of the invention of the present invention is to provide a kind of Fabric Defect detection method based on morphological analysis, to reduce, Fabric Defect is carried out to the cost automatically detected, the manual detection means of generally using in subrogate country.
To achieve the above object of the invention, the technical solution used in the present invention is: a kind of Fabric Defect detection method based on morphological analysis comprises the steps:
(1), adopt line-scan digital camera to carry out image acquisition to the cloth in motion, the cloth gray level image that to obtain bit depth be 8;
(2), the cloth gray level image is carried out to basic brightness stretching, binary conversion treatment, statistical organization point distribution number, and calculate cloth adjacent tissue dot spacing r, as structural element size reference value;
(3), according to flaw type selecting structure element, for the flaw that is sub-circular, described structural element is taken as the disc structure that radius is r, for the flaw of line style, described structural element is taken as the square structure that the length of side is r;
(4), former cloth gray level image is carried out to contrast stretching by top, end cap conversion, weaken the impact of surround lighting on image simultaneously;
(5), adopt in morphology expand, the mode of the corrosion gray level image after to contrast stretching processed, first expand and corrode again, to fill the dark-coloured gap between conventional organization point, with the contrast in the formation brightness of flaw place;
(6), (5) gained image is carried out to the negative film operation with the gray scale power transform, make flaw partly become the light tone main body, training is simultaneously adjusted in the gray scale power transform
the value, gray-scale value is carried out to Compression and Expansion, until under same ambient brightness intact part with flaw part grey value profile in 127.5 both sides;
(7), (6) gained image is carried out take 127.5 binary conversion treatment that are threshold value, and to make the structural element size be the dilation operation of r, to the flaw filling of caving in;
(8), in the image after binary conversion treatment, the point that is 1 for value utilizes the mode of eight connections to mark connection, the number of connected region is the flaw number; Add up the area that the pixel sum that comprises in each connected region is this flaw, the center that the barycenter that calculates connected region is this flaw, realize the detection of Fabric Defect thus.
In technique scheme, at first input picture is carried out to basic binaryzation operation, analyze and obtain the adjacent distance of knitting of cloth, for follow-up morphology, process the structural element size is provided; The stretching of the mode gray-scale value then converted with the top cap, strengthen picture contrast, makes it be not limited to less tonal range; Then carry out the morphology closed operation, fill a conventional gap of knitting, it is on the low side and irregular that the knitting of flaw place counted, and in brightness, highlighted; Finally, image is carried out to binary conversion treatment, determining of its threshold value controlled by the training adjustment of γ value in the gray scale power transform, adds up and mark two-value flaw figure, exports flaw information, completes the Defect Detection process of cloth.
In technique scheme, in step (2), directly from spatial domain angle geometric ways, add up its tissue and count, thereby calculate the distance between adjacent tissue point, as the reference value of morphological structuring elements size.
In step (6), training
value replaces the threshold value of binaryzation.Carry out the gray scale adjustment when carrying out the negative film operation, the mode of gray scale adjustment is power transform,
,
be respectively the gray level image of conversion front and back,
,
, γ is normal number,
being taken as 1, γ value is obtained by the training adjustment to indefectible image.
Because technique scheme is used, the present invention compared with prior art has following advantages:
The present invention adopts the method for mathematical morphology to solve this complicated realistic problem of Defect Detection from the spatial domain angle by the structure analysis to image.Its difficult point is the determining of size of morphological structuring elements, choosing of structural element is the key that guarantees the morphology operations accuracy, the present invention gathers image with the linear array high definition camera, cloth interlacing point clear in structure, utilize geometric knowledge just can calculate the value reference value of structural element, calculate the scale dependent parameter and need not be transformed into frequency domain, and this method structure is more accurately simple, has improved computing velocity; Another the present invention, when carrying out contrast stretching, is not to use general luminance transformation, but adopts top, end cap mapping mode, because can weaken the interference of background luminance inequality when carrying out top cap conversion; Moreover, when the setting of the threshold value of the present invention during by binary conversion treatment is converted into the gray scale power transform, the γ value chooses, the reason that the different images threshold value is different is that the ambient brightness change affects the gradation of image value, from the gray scale angle, this variable is controlled, and result can be more accurate.
The accompanying drawing explanation
Fig. 1 is the Fabric Defect detection method process flow diagram based on morphological analysis in the embodiment of the present invention;
Fabric Defect sample image when Fig. 2-Fig. 6 is the common varying environment brightness gathered in embodiment;
Fig. 7-Figure 12 be take the Defect Detection procedural image that Fig. 4 is example;
Figure 13-16th, the testing result output image of common Fabric Defect image.
Embodiment
Below in conjunction with drawings and Examples, the invention will be further described:
Embodiment mono-: shown in Figure 1, and be the process flow diagram of the Fabric Defect detection method research based on morphological analysis.At first utilize the image calculation structural element to gathering, with the morphology mode strengthen contrast, the interlacing point gap is filled in closed operation, carries out binaryzation after the computing of compressibility negative film, statistics mark flaw is exported relevant information.
As Fig. 2 to Fig. 6 is the typical flaw image that the present invention gathers, below take Fig. 4 and detection algorithm be specifically addressed as example:
1, computation structure element size
At first calculate the adjacent tissue dot spacing of the cloth image that gathers, when the cloth image is larger, can carry out the sample decomposition collection to the cloth image, as intercept long for a=500, wide for b=200 roughly be 4 sample images that are distributed in former figure, as shown in Figure 7.
Sample image is carried out to the linear luminance stretching, make the gradation of image value be covered in 0~255, take and 127.5 as threshold value, make binary conversion treatment, in image, interlacing point shows as white bright spot (value is 1).The point mark connected component that is 1 to mark value by the mode of eight connections, add up to such an extent that each sample interlacing point distribution number mean value of 500 * 200 pixels is n=1376, and the formula that calculates cloth adjacent tissue dot spacing is
, result is as the structural element size.
According to common flaw type selecting structure element, usually be divided into two classes structure: a class is the flaws such as broken hole that circularity is higher, stain, and as Fig. 2,3,5, structural element adopts disc structure, and disc radius is taken as r; Another kind of is the flaws such as the narrower scarce warp that is similar to line style of width, crapand, and as Fig. 4,11, now structural element is selected square structure, and the square length of side is taken as r.
2, former flaw figure strengthens contrast
Gathered original image is carried out to contrast stretching by the conversion of top cap, weaken the impact of surround lighting on the bright dark inequality of image simultaneously.The gray-scale value of image that camera is clapped is comparatively concentrated, flaw with knit margin of image element a little apart from enough not greatly, as Fig. 4, its gray-scale value scope is 47~186, contrasts not obviously, thereby will strengthen contrast.Mode commonly used is the linear luminance conversion, by 47~186, by linear scaling, is stretched to 0~255, but observes known its brightness irregularities of image, and the linear luminance conversion can not reduce ambient brightness and disturb.
The present invention selects the morphology contrast stretching conversion based on top cap, the conversion of end cap.Cap conversion in top is that former figure is deducted to the result after its opening operation, the inhomogeneous image for background, and when structural element is got suitable value, opening operation obtains background, and top cap transformation results is exactly to get rid of the image of background influence.In like manner, end cap conversion is that former figure is deducted to the result after its closed operation.
Above-mentioned two conversion can be used together for strengthening contrast, are called the morphology contrast variation.
To lack make this contrast stretching through flaw figure result as shown in Figure 8.
3, interlacing point gap filling
Adopt to expand, the mode of the corrosion gray level image after to contrast stretching processed, and first expands and corrodes, fills the dark-coloured gap between conventional organization point, with the contrast in the formation brightness of flaw place.
Wherein,
with
be respectively
with
field of definition, translation parameters
and
?
field of definition in, and
with
?
field of definition in.
Fig. 8 is first corroded to rear expansion, can play the very little i.e. value in trough of gray-scale value is upwards stretched, make the mild effect of trough.All in all, partially bright except other part overall brightnesses of flaw place, flaw is highlighted, as Fig. 9.
4, negative film conversion, binaryzation
Image to the gap filling gained carries out the negative film operation with the gray scale power transform, make flaw partly become the light tone main body, the γ value of power transform is adjusted in training simultaneously, to the suitable Compression and Expansion of gray-scale value, make grey value profile in the intact part of same ambient brightness hypograph and flaw part in 127.5 both sides.
Power transform is the non-linear brightness conversion,
,
, γ is normal number,
by training, obtained,
be taken as 1.γ<1 o'clock, higher numerical value (becoming clear) output is laid particular stress in mapping; γ > 1 o'clock, lower numerical value (gloomy) output is laid particular stress in mapping.
Because gained image graph 9 after closed operation, be not that to only have flaw be dark-coloured, all the other have partial pixel and flaw gap little, dead color will be compressed to than close limit under this environment, so select γ<1.Determining of the concrete numerical value of γ, adopt the training to indefectible image, and this lacks trains to obtain critical value 0.92 through the intact image of image environment of living in, and γ will approach critical value as far as possible, too smallly causes luminance compression too, produces undetectedly, gets 0.90, and result is as Figure 10.As γ<1 under Fig. 2,3 ambient brightness, now need dark-coloured scope compression, and the γ value is less than Fig. 4; As γ under Fig. 5,6 ambient brightness > 1, now need the light tone scope is compressed.Then given tacit consent to binaryzation, and do suitably to expand the flaw depression is filled, removed burrs on edges, as Figure 11.
5, statistics flaw information output
After binary conversion treatment, the image background pixel value is 0(black), the flaw pixel value is 1(white), marking connection for the point that is labeled as 1 by eight modes that are communicated with, the number of connected region is the flaw number; Add up the area that the pixel sum that comprises in each connected region is this flaw, the center that the barycenter that calculates connected region is this flaw.Finally, output flaw image and above-mentioned flaw parameter, as Figure 12.
Fig. 7 to Figure 12 represents the Check processing process of falling vacant through flaw intuitively, and the testing result of other common flaws is as shown in Figure 13 to Figure 16.
Claims (3)
1. the Fabric Defect detection method based on morphological analysis, is characterized in that, comprises the steps:
(1), adopt line-scan digital camera to carry out image acquisition to the cloth in motion, the cloth gray level image that to obtain bit depth be 8;
(2), the cloth gray level image is carried out to basic brightness stretching, binary conversion treatment, statistical organization point distribution number, and calculate cloth adjacent tissue dot spacing r, as structural element size reference value;
(3), according to flaw type selecting structure element, for the flaw that is sub-circular, described structural element is taken as the disc structure that radius is r, for the flaw of line style, described structural element is taken as the square structure that the length of side is r;
(4), former cloth gray level image is carried out to contrast stretching by top, end cap conversion, weaken the impact of surround lighting on image simultaneously;
(5), adopt in morphology expand, the mode of the corrosion gray level image after to contrast stretching processed, first expand and corrode again, to fill the dark-coloured gap between conventional organization point, with the contrast in the formation brightness of flaw place;
(6), (5) gained image is carried out to the negative film operation with the gray scale power transform, make flaw partly become the light tone main body, the γ value in the gray scale power transform is adjusted in training simultaneously, gray-scale value is carried out to Compression and Expansion, until under same ambient brightness intact part with flaw part grey value profile in 127.5 both sides;
(7), (6) gained image is carried out take 127.5 binary conversion treatment that are threshold value, and to make the structural element size be the dilation operation of r, to the flaw filling of caving in;
(8), in the image after binary conversion treatment, the point that is 1 for value utilizes the mode of eight connections to mark connection, the number of connected region is the flaw number; Add up the area that the pixel sum that comprises in each connected region is this flaw, the center that the barycenter that calculates connected region is this flaw, realize the detection of Fabric Defect thus.
2. the Fabric Defect detection method based on morphological analysis according to claim 1, it is characterized in that: in step (2), directly from the spatial domain angle, by geometric ways, adding up its tissue counts, thereby calculate the distance between adjacent tissue point, as the reference value of morphological structuring elements size.
3. the Fabric Defect detection method based on morphological analysis according to claim 1, is characterized in that: in step (6), train the γ value to replace the threshold value of binaryzation; Carry out the gray scale adjustment when carrying out the negative film operation, the mode of gray scale adjustment is power transform,
,
be respectively the gray level image of conversion front and back,
,
, γ is normal number,
being taken as 1, γ value is obtained by the training adjustment to indefectible image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310437747.9A CN103456021B (en) | 2013-09-24 | 2013-09-24 | A kind of Fabric Defect detection method based on morphological analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310437747.9A CN103456021B (en) | 2013-09-24 | 2013-09-24 | A kind of Fabric Defect detection method based on morphological analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103456021A true CN103456021A (en) | 2013-12-18 |
CN103456021B CN103456021B (en) | 2016-05-18 |
Family
ID=49738349
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310437747.9A Active CN103456021B (en) | 2013-09-24 | 2013-09-24 | A kind of Fabric Defect detection method based on morphological analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103456021B (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104142349A (en) * | 2014-07-28 | 2014-11-12 | 云南省机械研究设计院 | Method for detecting heat sealing defects of external packaging transparent film |
CN105354836A (en) * | 2015-10-14 | 2016-02-24 | 合肥安晶龙电子股份有限公司 | Color sorting method |
CN106504251A (en) * | 2016-09-22 | 2017-03-15 | 北京科技大学 | A kind of electric aluminum foil cubic texture detection method of content based on picture processing |
CN107248158A (en) * | 2017-07-20 | 2017-10-13 | 广东工业大学 | A kind of method and system of image procossing |
CN109509187A (en) * | 2018-11-05 | 2019-03-22 | 中山大学 | A kind of efficient check algorithm for the nibs in big resolution ratio cloth image |
CN110400306A (en) * | 2019-07-29 | 2019-11-01 | 恒天嘉华非织造有限公司 | Non-woven fabrics defect detection method based on morphologic filtering and convolutional neural networks |
CN110458761A (en) * | 2019-06-25 | 2019-11-15 | 千奥星科南京生物科技有限公司 | A kind of brain medicine fluorescence video image restorative procedure |
CN110579486A (en) * | 2019-10-18 | 2019-12-17 | 东莞市庆颖智能自动化科技有限公司 | equipment and method for detecting internal flaw image of semiconductor silicon crystal column |
CN113222882A (en) * | 2020-01-21 | 2021-08-06 | 上海铼钠克数控科技股份有限公司 | Cloth seam detection method and device, electronic equipment and storage medium |
CN115311294A (en) * | 2022-10-12 | 2022-11-08 | 启东金耀億华玻纤材料有限公司 | Glass bottle body flaw identification and detection method based on image processing |
CN116579672A (en) * | 2023-07-15 | 2023-08-11 | 图南益(杭州)智能科技有限公司 | Cloth quality control method, system, electronic equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3841761A (en) * | 1973-10-24 | 1974-10-15 | Neotec Corp | Method and apparatus for detecting faults in fabric |
CN101216435A (en) * | 2008-01-03 | 2008-07-09 | 东华大学 | Fabric flaw automatic detection method based on multi-fractal characteristic parameter |
CN102680478A (en) * | 2012-04-25 | 2012-09-19 | 华南农业大学 | Detection method and device of surface defect of mechanical part based on machine vision |
CN103175844A (en) * | 2012-03-16 | 2013-06-26 | 沈阳理工大学 | Detection method for scratches and defects on surfaces of metal components |
-
2013
- 2013-09-24 CN CN201310437747.9A patent/CN103456021B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3841761A (en) * | 1973-10-24 | 1974-10-15 | Neotec Corp | Method and apparatus for detecting faults in fabric |
CN101216435A (en) * | 2008-01-03 | 2008-07-09 | 东华大学 | Fabric flaw automatic detection method based on multi-fractal characteristic parameter |
CN103175844A (en) * | 2012-03-16 | 2013-06-26 | 沈阳理工大学 | Detection method for scratches and defects on surfaces of metal components |
CN102680478A (en) * | 2012-04-25 | 2012-09-19 | 华南农业大学 | Detection method and device of surface defect of mechanical part based on machine vision |
Non-Patent Citations (4)
Title |
---|
HENRY Y.T. NGAN,GRANTHAM K.H. PANG,NELSON H.C. YUNG: "Automated fabric defect detection-A review", 《IMAGE AND VISION COMPUTING》 * |
HONG-GANG BU,JUN WANG,XIU-BAO HUANG: "Fabric defect detection based on multiple fractal features and support vector data description", 《ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE》 * |
K.L.MARK,P.PENG,K.F.C. YIU: "Fabric defect detection using morphological filters", 《IMAGE AND VISION COMPUTING》 * |
王文杰: "瑕疵点检测算法研究及应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104142349A (en) * | 2014-07-28 | 2014-11-12 | 云南省机械研究设计院 | Method for detecting heat sealing defects of external packaging transparent film |
CN105354836A (en) * | 2015-10-14 | 2016-02-24 | 合肥安晶龙电子股份有限公司 | Color sorting method |
CN105354836B (en) * | 2015-10-14 | 2018-08-28 | 合肥安晶龙电子股份有限公司 | Color selection method |
CN106504251A (en) * | 2016-09-22 | 2017-03-15 | 北京科技大学 | A kind of electric aluminum foil cubic texture detection method of content based on picture processing |
CN106504251B (en) * | 2016-09-22 | 2019-02-15 | 北京科技大学 | A kind of electric aluminum foil cubic texture detection method of content based on picture processing |
CN107248158A (en) * | 2017-07-20 | 2017-10-13 | 广东工业大学 | A kind of method and system of image procossing |
CN109509187A (en) * | 2018-11-05 | 2019-03-22 | 中山大学 | A kind of efficient check algorithm for the nibs in big resolution ratio cloth image |
CN109509187B (en) * | 2018-11-05 | 2022-12-13 | 中山大学 | Efficient inspection algorithm for small defects in large-resolution cloth images |
CN110458761A (en) * | 2019-06-25 | 2019-11-15 | 千奥星科南京生物科技有限公司 | A kind of brain medicine fluorescence video image restorative procedure |
CN110400306A (en) * | 2019-07-29 | 2019-11-01 | 恒天嘉华非织造有限公司 | Non-woven fabrics defect detection method based on morphologic filtering and convolutional neural networks |
CN110579486A (en) * | 2019-10-18 | 2019-12-17 | 东莞市庆颖智能自动化科技有限公司 | equipment and method for detecting internal flaw image of semiconductor silicon crystal column |
CN113222882A (en) * | 2020-01-21 | 2021-08-06 | 上海铼钠克数控科技股份有限公司 | Cloth seam detection method and device, electronic equipment and storage medium |
CN115311294A (en) * | 2022-10-12 | 2022-11-08 | 启东金耀億华玻纤材料有限公司 | Glass bottle body flaw identification and detection method based on image processing |
CN116579672A (en) * | 2023-07-15 | 2023-08-11 | 图南益(杭州)智能科技有限公司 | Cloth quality control method, system, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN103456021B (en) | 2016-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103456021A (en) | Piece goods blemish detecting method based on morphological analysis | |
CN102221559B (en) | Online automatic detection method of fabric defects based on machine vision and device thereof | |
CN104268505A (en) | Automatic cloth defect point detection and recognition device and method based on machine vision | |
CN102158653B (en) | Device and method for acquiring digital image with high dynamic range in real time | |
CN103063167B (en) | A kind of method of automatic decision laser cleaning effect | |
CN103698341B (en) | A kind of quality of roll for cotton detection system and quality of roll for cotton image detecting method | |
CN105427275B (en) | Crop field environment wheat head method of counting and device | |
CN102622763A (en) | Method for detecting and eliminating shadow | |
CN102622602B (en) | Cotton foreign fiber image online dividing method and cotton foreign fiber image online dividing system | |
CN104408709B (en) | A kind of irregular quick antidote of linear array CCD camera gradation of image | |
CN105044128A (en) | Method for detecting fruit surface defect based on light background difference image | |
CN103839271A (en) | Method for image texture synthesis based on optimum matching | |
CN107895363A (en) | Textile flaw detection method based on template characteristic | |
CN110378887A (en) | Screen defect inspection method, apparatus and system, computer equipment and medium | |
CN115082744A (en) | Solar energy heat collection efficiency analysis method and system based on artificial intelligence | |
CN107025891B (en) | A kind of display module defect fast repairing method and system | |
US11055839B2 (en) | Method for fast judging and optimizing light emitting quality of light guide plate based on image processing technology | |
CN103471974A (en) | Method for determining theoretic porosity of fabric through image method | |
CN105184256B (en) | A kind of surge test Device under test status lamp image steganalysis method | |
CN113362274B (en) | Rainfall monitoring and calculating method | |
CN103177244A (en) | Method for quickly detecting target organisms in underwater microscopic images | |
CN117576093A (en) | EL image-based solar cell piece hidden crack detection method | |
CN104899863A (en) | Mold protector and implementation method thereof | |
CN103927544A (en) | Machine vision grading method for ginned cotton rolling quality | |
Yogesh et al. | Adaptive thresholding based segmentation of infected portion of pome fruit |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CP02 | Change in the address of a patent holder | ||
CP02 | Change in the address of a patent holder |
Address after: Suzhou City, Jiangsu province 215137 Xiangcheng District Ji Road No. 8 Patentee after: Soochow University Address before: 215123 Suzhou City, Suzhou Province Industrial Park, No. love road, No. 199 Patentee before: Soochow University |