CN103871053A - Vision conspicuousness-based cloth flaw detection method - Google Patents
Vision conspicuousness-based cloth flaw detection method Download PDFInfo
- Publication number
- CN103871053A CN103871053A CN201410063927.XA CN201410063927A CN103871053A CN 103871053 A CN103871053 A CN 103871053A CN 201410063927 A CN201410063927 A CN 201410063927A CN 103871053 A CN103871053 A CN 103871053A
- Authority
- CN
- China
- Prior art keywords
- significantly
- gray
- scale
- detection method
- target area
- 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
- Treatment Of Fiber Materials (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a vision conspicuousness-based cloth flaw detection method which comprises the following steps: (1) collecting an image; (2) processing brightness features; (3) processing direction features; (4) performing multichannel superposition normalizing processing; (5) processing a grey-scale map; (6) performing binaryzation processing; (7) judging a flaw area. Compared with the traditional cloth flaw detection method, the vision conspicuousness-based cloth flaw detection method has the advantages that the operation complexity is reduced, the recognition rate is increased, accurate positioning can be realized, false detection easily caused under the condition that a gray value of a conspicuousness image of a perfection image of detected cloth is higher than a gray value of a perfection part in a flaw image is avoided, the interference of a background during detection is effectively reduced, and the condition that a target area obtained by performing adaptive threshold segmentation on an image of perfect cloth is mistakenly judged to be the flaw area is reduced.
Description
Technical field
The present invention relates to a kind of Fabric Defect detection method, be specifically related to a kind of Fabric Defect detection method based on vision significance.
Background technology
In modern textile is produced, the control of quality is extremely important with detection, the Defect Detection of cloth is wherein particularly crucial ingredient, at present, the detection mode of domestic textile enterprise is mainly with being manually main, human eye detection speed is limited, and testing result is subject to subjective factor impact, and flase drop, undetected etc. easily occurs.Replacing artificial Fabric Defect to detect with advanced Automatic Measurement Technique, is the important measures that improve detection efficiency, minimizing labour, reduce hand labor intensity and assurance cloth quality.Chinese scholars has obtained the achievement of a lot of brilliances in the research of automatic testing method.
Fabric Defect detection algorithm is mainly to carry out flaw judgement according to the texture shape of the cloth that detects, roughly has 3 classes: statistic law, Zymography and based on modelling.The Spectral Analysis Method that wherein the modal utilization Fourier transform that is, wavelet transformation and Gabor analyze.Fourier transform is that figure is carried out to global change, thereby can not accurately locate flaw; The change detection better performances that Gabor analyzes, but need to carry out two-dimensional filtering and fusion to hyperchannel direction, greatly increase computation complexity; Wavelet transformation has the advantages such as good local time's frequency analysis, computing velocity be fast, but directional selectivity is poor, makes it can not fine description X-Y scheme characteristic, causes detecting effect undesirable.
The mankind can fast and effeciently identify defect, no matter reflective power, the variation of defect shape, difference that various processes causes etc.When cloth is presented on before human eye, its flaw part is comparatively remarkable, can cause vision attention.That is to say, in figure, information to be found often concentrates on some crucial regions, is conventionally referred to as marking area.If can extract exactly these marking areas, i.e. the region of arresting visually, we just can effectively find the figure information needing.In order to extract marking area from figure, people have proposed the computation model based on vision attention.In these models, represent the conspicuousness of visual zone with remarkable figure.Significantly figure is a width X-Y scheme, the conspicuousness size of the former figure corresponding pixel points of value representation of its each pixel.The present invention is based on vision significance principle, and a kind of method of cloth figure being carried out to Defect Detection is provided.
Summary of the invention
The object of this invention is to provide a kind of Fabric Defect detection method based on vision significance, simulation people's vision noticing mechanism, take GBVS(Graph-based Visual Saliency) model be basis, by extracting brightness and the direction character of cloth, generating vision significantly schemes, by analyzing the intensity profile relation of vision remarkable figure, judge that whether vision significantly scheme concentration of energy, and then judge whether cloth exists flaw.
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 vision significance, comprises the steps:
(1), gather image, cloth is carried out to figure collection, obtain initial gray-scale map;
(2), GBVS model generation specific image, according to the comprehensive significantly figure of initial gray-scale map described in GBVS model generation;
(3), gray-scale map processing, reduce the gray-scale value of the each pixel of described comprehensive significantly figure, thereby obtain weakening the final significantly figure after background conspicuousness;
(4), binary conversion treatment, according to maximum variance between clusters, described final significantly figure is carried out to adaptive threshold and cuts apart, obtain having the binary map of outstanding significant target part;
(5), defect areas judgement, by the grey sum of the squares of the degree and described comprehensive grey sum of the squares of the degree of significantly scheming whole region of corresponding target area in more described final significantly figure, to judge that whether target area is as flaw.
Further technical scheme, step (2) comprises as follows step by step:
(a), brightness processing, described initial gray-scale map is carried out to gaussian pyramid filtering, obtain one group of brightness filtering result under different scale, to each described brightness filtering result, set up Markov chain separately according to the difference between its pixel and Euclidean distance, then ask its Markov balance
Distribute, the brightness filtering result under the described one group of different scale after Markov balanced distribution is normalized, obtain a brightness and significantly scheme;
(b), direction character processing, described initial gray-scale map is carried out to the filtering of Gabor pyramid on four direction, correspondence obtains four groups of direction character filtering results under different scale, to each described direction character filtering result, set up Markov chain separately according to the difference between its pixel and Euclidean distance, then ask its Markov balanced distribution, direction character filtering result under described one group of different scale after Markov balanced distribution is normalized, obtains a direction character and significantly scheme;
(c), hyperchannel stack normalization, by the significantly remarkable figure linear, additive of described direction character in figure and step (b) being normalized of the described brightness in step (a), obtain comprehensively significantly scheming;
Wherein, described step (a) and step (b) are without sequencing.
Further technical scheme, step (3) is adopted with the following method: calculate the described comprehensive significantly average gray of each pixel of figure, whether the gray-scale value of described each pixel of judgement is greater than described average gray, if described gray-scale value is less than described average gray, described gray-scale value is carried out to deal with return to zero, otherwise described gray-scale value equals the difference of itself and described average gray, obtain weakening the final significantly figure of background conspicuousness.
Further technical scheme, step (5) is adopted with the following method: demarcating described final significantly figure is target area corresponding to the region of described target part, calculate respectively the pixel ash sum of the squares of the degree of the described final significantly target area of figure and the pixel ash sum of the squares of the degree in the described comprehensive significantly whole region of figure, and then calculate the ratio of described target area ash sum of the squares of the degree and described whole area grayscale quadratic sum, if described ratio is during lower than judgment threshold, described target area is defect areas, otherwise described target area is not defect areas.
Further technical scheme, the span of described judgment threshold is 0.15~0.4.
Because technique scheme is used, the present invention compared with prior art has following advantages:
1, by the cloth figure of input is carried out to vision significance analysis, judge whether to obtain having the binary map of outstanding significant target area, and then judge whether cloth exists flaw, compared with traditional Fabric Defect detection method, the present invention has reduced the complexity of computing, improve discrimination, and can accurately locate;
2, by grey scale pixel value in comprehensive significantly figure is reduced higher than the gray-scale value of the point of mean value, the gray-scale value of the sub-average point of grey scale pixel value in comprehensive significantly figure is made zero, to weaken background (depending on the intact region of cloth as background) conspicuousness, make the conspicuousness in defect areas and intact region have obvious gap, avoid due to the described comprehensive significantly cause of map generalization principle, the gray-scale value in the intact region of cloth of detecting is than the flase drop easily causing in the taller situation of the intact part gray-scale value in defect areas, effectively reduce the interference of background in testing process,
3, because human eye always has the concentrated region of relative notice in the time that observation texture is evenly schemed, therefore the target part that the detection method of the Fabric Defect based on vision significance easily obtains intact cloth figure through over-segmentation is mistaken for defect areas, carry out energy judgement by the figure characteristic information that extracts target area, judge whether this target area is concentration of energy region, final definite defect areas, calculate the pixel ash sum of the squares of the degree of target area and the pixel ash sum of the squares of the degree in the described comprehensive significantly whole region of figure corresponding to described target part in described final significantly figure, and then calculate described both ratio, judge by ratio size whether energy is relatively concentrated in this target area, ratio is greater than judgment threshold, be that energy is relatively concentrated, described target area is flaw, otherwise described target area is not flaw, by above technical scheme, reduce the generation of erroneous judgement.
Accompanying drawing explanation
Fig. 1 is the implementing procedure figure of the Fabric Defect detection method based on vision significance disclosed by the invention;
Fig. 2 is the progressively testing result figure of the intact figure of cloth disclosed by the invention;
Fig. 3 is the progressively testing result figure of cloth typical case flaw figure disclosed by the invention;
Fig. 4, Fig. 5, Fig. 6, Fig. 7 is the testing result figure of other flaws of cloth disclosed by the invention figure.
Embodiment
Below in conjunction with drawings and Examples, the present invention is further described.
Embodiment: referring to Fig. 1, as shown in legend wherein, above-mentioned Fabric Defect detection method comprises the steps:
(1), gather image, gather the figure of cloth by industrial camera, obtain initial gray-scale map
, if Fig. 2 (a) is the initial graph of the intact figure of cloth, if Fig. 3 (a) is the initial gray-scale map of cloth typical case flaw figure.
(2), brightness processing:
A, by 2-d gaussian filters device to input above-mentioned initial gray-scale map
carry out gaussian pyramid filtering, pyramid filtering refers to constantly carries out 1/2 down-sampled and filtering to initial gray-scale map, the scale factor of filtering reduces and reduces with figure's, obtain one group of filtering result under different scale, in this example, pyramid grade is 2, obtain the different brightness filtering results under two yardsticks, i.e. two brightness figure
;
Wherein,
initial gray-scale map,
in scale factor to brightness
the lower gaussian pyramid obtaining with Gaussian function filtering,
for the Gaussian function of 2-d gaussian filters device,
for the position of pixel in corresponding figure,
for the scale factor of corresponding figure;
B, to brightness figure every described
carry out Markov balanced distribution, and then normalized, obtain brightness and significantly scheme
;
If each brightness figure
for
, the distance between any two pixels is
, any two pixels
with
distance:
By characteristic pattern
every two pixels all connect between two, and every limit represents the weight of point-to-point transmission, point
to point
weight
for:
Wherein
for Gauss's weighting parameters;
That is to say weight between any two pixels by the gray difference between two pixels and dot spacing from common decision, reverse direction weight is identical, said process is the process of asking the eigenvalue of maximum characteristic of correspondence vector of every brightness figure respective weights matrix, asks the equilibrium state process of every brightness figure;
By two brightness balanced distribution figure
be normalized, obtain a brightness and significantly scheme
.
(3), direction character processing:
A, by Gabor wave filter to input initial gray-scale map
on four direction, carry out the filtering of Gabor pyramid, pyramid filtering refers to constantly carries out 1/2 down-sampled and filtering to initial gray-scale map, the scale factor of filtering reduces and reduces with figure's, obtain four groups of filtering results under different scale, in this example, pyramid grade is 2, obtain four groups of the different directions characteristic filtering results under two yardsticks, i.e. eight direction character figure
;
Wherein,
the Gabor pyramid that direction character is obtained in the Gabor function filtering of yardstick direction,
for the Gabor function of Gabor wave filter,
for the position of pixel in corresponding figure,
for scale factor,
for direction factor (is got
),
for the sine wave freuqency of Gabor wave filter;
B, to every direction character figure
carry out Markov balanced distribution, and then normalized, obtain brightness and significantly scheme
,
If each direction character figure
for
, the distance between any two pixels is
, any two pixels
with
distance:
By characteristic pattern
every two pixels all connect between two, and every limit represents the weight of point-to-point transmission, point
to point
weight
for:
Wherein
for Gauss's weighting parameters;
That is to say weight between any two pixels by the gray difference between two pixels and dot spacing from common decision, reverse direction weight is identical, said process is the process of asking the eigenvalue of maximum characteristic of correspondence vector of every direction character figure respective weights matrix
,ask the equilibrium state process of every direction character figure;
By eight direction character balanced distribution figure
be normalized, obtain direction character and significantly scheme
;
(4), stack normalization, above-mentioned brightness is significantly schemed
significantly scheme with direction character
linear, additive is also normalized, and obtains and above-mentioned initial gray-scale map
comprehensive significantly figure of the same size
, if Fig. 2 (b) is the comprehensive significantly figure of the intact figure of cloth, if Fig. 3 (b) is the comprehensive significantly figure of cloth typical case flaw figure.
(5), gray-scale map processing, calculate above-mentioned comprehensive significantly figure
average gray, more above-mentioned comprehensive significantly figure
each pixel gray-scale value and average gray, if above-mentioned pixel gray-scale value is less than average gray, above-mentioned pixel gray-scale value is carried out to deal with return to zero, otherwise above-mentioned pixel gray-scale value equals the difference of itself and average gray, obtain weakening the final significantly figure of background gray scale
, if Fig. 2 (c) is the final significantly figure of the intact figure of cloth, if Fig. 3 (c) is the final significantly figure of cloth typical case flaw figure.
(6), binary conversion treatment, according to maximum variance between clusters, to above-mentioned final significantly figure
carry out adaptive threshold and cut apart, obtain having the binary map of outstanding significant target part
, if Fig. 2 (d) is the binary map of the intact figure of cloth, if Fig. 3 (d) is the binary map of cloth typical case flaw figure.
(7), defect areas judgement, demarcate above-mentioned final significantly figure
in be target area corresponding to the region of described target part, calculate above-mentioned final significantly figure
pixel ash sum of the squares of the degree and the described comprehensive significantly figure of target area
the pixel ash sum of the squares of the degree in whole region, and then calculate both ratio.Judgment threshold span is 0.15~0.4 in the methods of the invention, this routine judgment threshold is taken as 0.2, if above-mentioned ratio is during lower than judgment threshold 0.2, above-mentioned target area is defect areas, otherwise above-mentioned target area is not defect areas, if Fig. 2 (e) is the judged result figure of the intact figure of cloth, if Fig. 3 (e) is the judged result figure of cloth typical case flaw figure.
Calculate the ratio of the grey sum of the squares of the degree in divided region and whole region, judge by ratio size whether energy is relatively concentrated in this target area.Ratio is greater than judgment threshold, and energy is relatively concentrated, is flaw; Otherwise be not flaw.
Referring to Fig. 4-Fig. 7, be the testing result figure of other several typical flaw figure of cloth, wherein (a) is initial graph, is (b) comprehensive significantly figure, (c) is judged result figure.
Claims (5)
1. the Fabric Defect detection method based on vision significance, is characterized in that, comprises the steps:
(1), gather image, cloth is carried out to figure collection, obtain initial gray-scale map;
(2), GBVS model generation specific image, according to the comprehensive significantly figure of initial gray-scale map described in GBVS model generation;
(3), gray-scale map processing, reduce the gray-scale value of the each pixel of described comprehensive significantly figure, thereby obtain weakening the final significantly figure after background conspicuousness;
(4), binary conversion treatment, according to maximum variance between clusters, described final significantly figure is carried out to adaptive threshold and cuts apart, obtain having the binary map of outstanding significant target part;
(5), defect areas judgement, by the grey sum of the squares of the degree and described comprehensive grey sum of the squares of the degree of significantly scheming whole region of corresponding target area in more described final significantly figure, to judge that whether target area is as flaw.
2. the Fabric Defect detection method based on vision significance according to claim 1, is characterized in that, step (2) comprises as follows step by step:
(a), brightness processing, described initial gray-scale map is carried out to gaussian pyramid filtering, obtain one group of brightness filtering result under different scale, to each described brightness filtering result, set up Markov chain separately according to the difference between its pixel and Euclidean distance, then ask its Markov balance
Distribute, the brightness filtering result under the described one group of different scale after Markov balanced distribution is normalized, obtain a brightness and significantly scheme;
(b), direction character processing, described initial gray-scale map is carried out to the filtering of Gabor pyramid on four direction, correspondence obtains four groups of direction character filtering results under different scale, to each described direction character filtering result, set up Markov chain separately according to the difference between its pixel and Euclidean distance, then ask its Markov balanced distribution, direction character filtering result under described one group of different scale after Markov balanced distribution is normalized, obtains a direction character and significantly scheme;
(c), hyperchannel stack normalization, by the significantly remarkable figure linear, additive of described direction character in figure and step (b) being normalized of the described brightness in step (a), obtain comprehensively significantly scheming;
Wherein, described step (a) and step (b) are without sequencing.
3. according to the Fabric Defect detection method based on vision significance described in claim 1, it is characterized in that, step (3) is adopted with the following method: calculate the described comprehensive significantly average gray of each pixel of figure, whether the gray-scale value of described each pixel of judgement is greater than described average gray, if described gray-scale value is less than described average gray, described gray-scale value is carried out to deal with return to zero, otherwise described gray-scale value equals the difference of itself and described average gray, obtain weakening the final significantly figure of background conspicuousness.
4. according to the Fabric Defect detection method based on vision significance described in claim 1, it is characterized in that, step (5) is adopted with the following method: demarcating described final significantly figure is target area corresponding to the region of described target part, calculate respectively the pixel ash sum of the squares of the degree of the described final significantly target area of figure and the pixel ash sum of the squares of the degree in the described comprehensive significantly whole region of figure, and then calculate the ratio of described target area ash sum of the squares of the degree and described whole area grayscale quadratic sum, if described ratio is during lower than judgment threshold, described target area is defect areas, otherwise described target area is not defect areas.
5. according to Fabric Defect detection method claimed in claim 4, it is characterized in that, the span of described judgment threshold is 0.15~0.4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410063927.XA CN103871053B (en) | 2014-02-25 | 2014-02-25 | Vision conspicuousness-based cloth flaw detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410063927.XA CN103871053B (en) | 2014-02-25 | 2014-02-25 | Vision conspicuousness-based cloth flaw detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103871053A true CN103871053A (en) | 2014-06-18 |
CN103871053B CN103871053B (en) | 2017-02-01 |
Family
ID=50909558
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410063927.XA Active CN103871053B (en) | 2014-02-25 | 2014-02-25 | Vision conspicuousness-based cloth flaw detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103871053B (en) |
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104198497A (en) * | 2014-09-12 | 2014-12-10 | 苏州大学 | Surface defect detection method based on visual saliency map and support vector machine |
CN104574353A (en) * | 2014-09-12 | 2015-04-29 | 苏州大学 | Surface defect judgment method based on visual saliency |
CN104700412A (en) * | 2015-03-17 | 2015-06-10 | 苏州大学 | Calculating method of visual salience drawing |
CN105021628A (en) * | 2015-07-19 | 2015-11-04 | 中北大学 | Detection method for surface defects of optical fiber image inverter |
CN105678767A (en) * | 2016-01-07 | 2016-06-15 | 无锡信捷电气股份有限公司 | SoC software and hardware collaborative design-based cloth surface blemish detection method |
CN106157314A (en) * | 2015-12-23 | 2016-11-23 | 河海大学 | A kind of real-time fault diagnosis method of paper making equipment |
CN107145824A (en) * | 2017-03-29 | 2017-09-08 | 纵目科技(上海)股份有限公司 | A kind of lane line dividing method and system, car-mounted terminal based on significance analysis |
CN107220962A (en) * | 2017-04-07 | 2017-09-29 | 北京工业大学 | A kind of image detecting method and device of tunnel crackle |
CN107316295A (en) * | 2017-07-02 | 2017-11-03 | 苏州大学 | A kind of fabric defects detection method based on deep neural network |
CN107895363A (en) * | 2017-10-31 | 2018-04-10 | 常州大学 | Textile flaw detection method based on template characteristic |
CN108268866A (en) * | 2016-12-30 | 2018-07-10 | 纵目科技(上海)股份有限公司 | A kind of vehicle checking method and system |
CN108629778A (en) * | 2017-03-15 | 2018-10-09 | 纵目科技(上海)股份有限公司 | Image significance object detection method and system based on local feature weighting enhancing |
CN108629225A (en) * | 2017-03-15 | 2018-10-09 | 纵目科技(上海)股份有限公司 | A kind of vehicle checking method based on several subgraphs and saliency analysis |
CN108629806A (en) * | 2017-03-15 | 2018-10-09 | 纵目科技(上海)股份有限公司 | A kind of image significance object detection method and system based on picture contrast |
CN108960255A (en) * | 2018-06-28 | 2018-12-07 | 西安工程大学 | Conspicuousness fabric defect detection method based on color similarity and position aggregation |
CN109816631A (en) * | 2018-12-25 | 2019-05-28 | 河海大学 | A kind of image partition method based on new cost function |
CN109961437A (en) * | 2019-04-04 | 2019-07-02 | 江南大学 | A kind of conspicuousness fabric defect detection method under the mode based on machine teaching |
CN109960984A (en) * | 2017-12-25 | 2019-07-02 | 大连楼兰科技股份有限公司 | Vehicle checking method based on contrast and significance analysis |
CN110031461A (en) * | 2019-02-14 | 2019-07-19 | 江苏恒力化纤股份有限公司 | A kind of polyester filament dye uniformity test method |
CN110298835A (en) * | 2019-07-02 | 2019-10-01 | 广东工业大学 | A kind of leather surface damage testing method, system and relevant apparatus |
CN110706205A (en) * | 2019-09-07 | 2020-01-17 | 创新奇智(重庆)科技有限公司 | Method for detecting cloth hole-breaking defect by using computer vision technology |
CN110751135A (en) * | 2019-12-24 | 2020-02-04 | 广东博智林机器人有限公司 | Drawing checking method and device, electronic equipment and storage medium |
CN111402225A (en) * | 2020-03-13 | 2020-07-10 | 浙江工业大学 | Cloth folding false detection defect discrimination method |
CN111667476A (en) * | 2020-06-09 | 2020-09-15 | 创新奇智(广州)科技有限公司 | Cloth flaw detection method and device, electronic equipment and readable storage medium |
CN111861996A (en) * | 2020-06-23 | 2020-10-30 | 西安工程大学 | Printed fabric defect detection method |
CN112669306A (en) * | 2021-01-06 | 2021-04-16 | 北京信息科技大学 | Solar cell defect detection method and system based on saliency map |
CN113638104A (en) * | 2021-10-14 | 2021-11-12 | 海门市恒昌织带有限公司 | Intelligent yarn cleaning control method and system for bobbin winder |
CN114486903A (en) * | 2021-12-06 | 2022-05-13 | 浙江大学台州研究院 | Gray-scale self-adaptive coiled material detection system, device and algorithm |
CN114511851A (en) * | 2022-01-30 | 2022-05-17 | 南水北调中线干线工程建设管理局 | Hairspring algae cell statistical method based on microscope image |
CN115598138A (en) * | 2022-11-23 | 2023-01-13 | 惠州威尔高电子有限公司(Cn) | Power control circuit board flaw detection method and system based on significance detection |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109300127B (en) * | 2018-09-21 | 2021-02-05 | 深圳新视智科技术有限公司 | Defect detection method, defect detection device, computer equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102214298A (en) * | 2011-06-20 | 2011-10-12 | 复旦大学 | Method for detecting and identifying airport target by using remote sensing image based on selective visual attention mechanism |
CN102799882A (en) * | 2012-07-09 | 2012-11-28 | 武汉市科迅智能交通设备有限公司 | License plate positioning method based on visual saliency |
CN103247059A (en) * | 2013-05-27 | 2013-08-14 | 北京师范大学 | Remote sensing image region of interest detection method based on integer wavelets and visual features |
CN103309982A (en) * | 2013-06-17 | 2013-09-18 | 武汉大学 | Remote sensing image retrieval method based on vision saliency point characteristics |
CN103544709A (en) * | 2013-11-07 | 2014-01-29 | 东北大学 | Method for detecting micro defects on surface of silicon steel plate based on significant active contour model |
-
2014
- 2014-02-25 CN CN201410063927.XA patent/CN103871053B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102214298A (en) * | 2011-06-20 | 2011-10-12 | 复旦大学 | Method for detecting and identifying airport target by using remote sensing image based on selective visual attention mechanism |
CN102799882A (en) * | 2012-07-09 | 2012-11-28 | 武汉市科迅智能交通设备有限公司 | License plate positioning method based on visual saliency |
CN103247059A (en) * | 2013-05-27 | 2013-08-14 | 北京师范大学 | Remote sensing image region of interest detection method based on integer wavelets and visual features |
CN103309982A (en) * | 2013-06-17 | 2013-09-18 | 武汉大学 | Remote sensing image retrieval method based on vision saliency point characteristics |
CN103544709A (en) * | 2013-11-07 | 2014-01-29 | 东北大学 | Method for detecting micro defects on surface of silicon steel plate based on significant active contour model |
Non-Patent Citations (3)
Title |
---|
HAREL J 等: "Graph-based visual saliency", 《ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS》, 28 February 2006 (2006-02-28), pages 545 - 552 * |
吴家伟: "基于图像显著性特征的铸坯表面缺陷检测", 《智能***学报》, vol. 7, no. 1, 28 February 2012 (2012-02-28), pages 75 - 80 * |
屈博 等: "一种改进的多通道Gabor滤波器布匹瑕疵检测方法", 《纺织学报》, vol. 30, no. 12, 31 December 2009 (2009-12-31), pages 37 - 40 * |
Cited By (48)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104198497B (en) * | 2014-09-12 | 2017-01-25 | 苏州大学 | Surface defect detection method based on visual saliency map and support vector machine |
CN104574353A (en) * | 2014-09-12 | 2015-04-29 | 苏州大学 | Surface defect judgment method based on visual saliency |
CN104198497A (en) * | 2014-09-12 | 2014-12-10 | 苏州大学 | Surface defect detection method based on visual saliency map and support vector machine |
CN104574353B (en) * | 2014-09-12 | 2017-08-01 | 苏州大学 | The surface defect decision method of view-based access control model conspicuousness |
CN104700412A (en) * | 2015-03-17 | 2015-06-10 | 苏州大学 | Calculating method of visual salience drawing |
CN104700412B (en) * | 2015-03-17 | 2019-01-22 | 苏州大学 | A kind of calculation method of visual saliency map |
CN105021628A (en) * | 2015-07-19 | 2015-11-04 | 中北大学 | Detection method for surface defects of optical fiber image inverter |
CN106157314A (en) * | 2015-12-23 | 2016-11-23 | 河海大学 | A kind of real-time fault diagnosis method of paper making equipment |
CN106157314B (en) * | 2015-12-23 | 2018-12-14 | 河海大学 | A kind of real-time fault diagnosis method of paper making equipment |
CN105678767A (en) * | 2016-01-07 | 2016-06-15 | 无锡信捷电气股份有限公司 | SoC software and hardware collaborative design-based cloth surface blemish detection method |
CN105678767B (en) * | 2016-01-07 | 2019-02-19 | 无锡信捷电气股份有限公司 | A kind of cloth surface flaw detection method based on SoC Hardware/Software Collaborative Design |
CN108268866B (en) * | 2016-12-30 | 2020-07-10 | 纵目科技(上海)股份有限公司 | Vehicle detection method and system |
CN108268866A (en) * | 2016-12-30 | 2018-07-10 | 纵目科技(上海)股份有限公司 | A kind of vehicle checking method and system |
CN108629225B (en) * | 2017-03-15 | 2022-02-25 | 纵目科技(上海)股份有限公司 | Vehicle detection method based on multiple sub-images and image significance analysis |
CN108629806A (en) * | 2017-03-15 | 2018-10-09 | 纵目科技(上海)股份有限公司 | A kind of image significance object detection method and system based on picture contrast |
CN108629778B (en) * | 2017-03-15 | 2022-02-25 | 纵目科技(上海)股份有限公司 | Image salient object detection method and system based on local feature weighting enhancement |
CN108629778A (en) * | 2017-03-15 | 2018-10-09 | 纵目科技(上海)股份有限公司 | Image significance object detection method and system based on local feature weighting enhancing |
CN108629806B (en) * | 2017-03-15 | 2022-02-25 | 纵目科技(上海)股份有限公司 | Image salient object detection method and system based on image contrast |
CN108629225A (en) * | 2017-03-15 | 2018-10-09 | 纵目科技(上海)股份有限公司 | A kind of vehicle checking method based on several subgraphs and saliency analysis |
CN107145824A (en) * | 2017-03-29 | 2017-09-08 | 纵目科技(上海)股份有限公司 | A kind of lane line dividing method and system, car-mounted terminal based on significance analysis |
CN107220962B (en) * | 2017-04-07 | 2020-04-21 | 北京工业大学 | Image detection method and device for tunnel cracks |
CN107220962A (en) * | 2017-04-07 | 2017-09-29 | 北京工业大学 | A kind of image detecting method and device of tunnel crackle |
CN107316295A (en) * | 2017-07-02 | 2017-11-03 | 苏州大学 | A kind of fabric defects detection method based on deep neural network |
CN107895363A (en) * | 2017-10-31 | 2018-04-10 | 常州大学 | Textile flaw detection method based on template characteristic |
CN109960984A (en) * | 2017-12-25 | 2019-07-02 | 大连楼兰科技股份有限公司 | Vehicle checking method based on contrast and significance analysis |
CN108960255A (en) * | 2018-06-28 | 2018-12-07 | 西安工程大学 | Conspicuousness fabric defect detection method based on color similarity and position aggregation |
CN109816631A (en) * | 2018-12-25 | 2019-05-28 | 河海大学 | A kind of image partition method based on new cost function |
CN110031461A (en) * | 2019-02-14 | 2019-07-19 | 江苏恒力化纤股份有限公司 | A kind of polyester filament dye uniformity test method |
CN110031461B (en) * | 2019-02-14 | 2022-03-18 | 江苏恒力化纤股份有限公司 | Polyester filament yarn dyeing uniformity test method |
CN109961437B (en) * | 2019-04-04 | 2021-06-25 | 江南大学 | Method for detecting significant fabric defects based on machine teaching mode |
CN109961437A (en) * | 2019-04-04 | 2019-07-02 | 江南大学 | A kind of conspicuousness fabric defect detection method under the mode based on machine teaching |
CN110298835B (en) * | 2019-07-02 | 2023-07-18 | 广东工业大学 | Leather surface damage detection method, system and related device |
CN110298835A (en) * | 2019-07-02 | 2019-10-01 | 广东工业大学 | A kind of leather surface damage testing method, system and relevant apparatus |
CN110706205A (en) * | 2019-09-07 | 2020-01-17 | 创新奇智(重庆)科技有限公司 | Method for detecting cloth hole-breaking defect by using computer vision technology |
CN110751135A (en) * | 2019-12-24 | 2020-02-04 | 广东博智林机器人有限公司 | Drawing checking method and device, electronic equipment and storage medium |
CN111402225A (en) * | 2020-03-13 | 2020-07-10 | 浙江工业大学 | Cloth folding false detection defect discrimination method |
CN111402225B (en) * | 2020-03-13 | 2023-05-26 | 浙江工业大学 | Cloth folding false-detection defect discriminating method |
CN111667476B (en) * | 2020-06-09 | 2022-12-06 | 创新奇智(广州)科技有限公司 | Cloth flaw detection method and device, electronic equipment and readable storage medium |
CN111667476A (en) * | 2020-06-09 | 2020-09-15 | 创新奇智(广州)科技有限公司 | Cloth flaw detection method and device, electronic equipment and readable storage medium |
CN111861996A (en) * | 2020-06-23 | 2020-10-30 | 西安工程大学 | Printed fabric defect detection method |
CN111861996B (en) * | 2020-06-23 | 2023-11-03 | 西安工程大学 | Printed fabric defect detection method |
CN112669306A (en) * | 2021-01-06 | 2021-04-16 | 北京信息科技大学 | Solar cell defect detection method and system based on saliency map |
CN113638104B (en) * | 2021-10-14 | 2022-02-08 | 海门市恒昌织带有限公司 | Intelligent yarn cleaning control method and system for bobbin winder |
CN113638104A (en) * | 2021-10-14 | 2021-11-12 | 海门市恒昌织带有限公司 | Intelligent yarn cleaning control method and system for bobbin winder |
CN114486903A (en) * | 2021-12-06 | 2022-05-13 | 浙江大学台州研究院 | Gray-scale self-adaptive coiled material detection system, device and algorithm |
CN114486903B (en) * | 2021-12-06 | 2024-05-14 | 浙江大学台州研究院 | Gray-scale self-adaptive coiled material detection system, device and algorithm |
CN114511851A (en) * | 2022-01-30 | 2022-05-17 | 南水北调中线干线工程建设管理局 | Hairspring algae cell statistical method based on microscope image |
CN115598138A (en) * | 2022-11-23 | 2023-01-13 | 惠州威尔高电子有限公司(Cn) | Power control circuit board flaw detection method and system based on significance detection |
Also Published As
Publication number | Publication date |
---|---|
CN103871053B (en) | 2017-02-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103871053A (en) | Vision conspicuousness-based cloth flaw detection method | |
CN104008399B (en) | The recognition methodss of the gauge pointer shake based on support vector machine in a kind of instrument detection | |
CN103604809B (en) | A kind of online visible detection method of pattern cloth flaw | |
CN104077594B (en) | A kind of image-recognizing method and device | |
CN104732536A (en) | Sub-pixel edge detection method based on improved morphology | |
CN107767387B (en) | Contour detection method based on variable receptive field scale global modulation | |
CN104198497B (en) | Surface defect detection method based on visual saliency map and support vector machine | |
CN104091157A (en) | Pedestrian detection method based on feature fusion | |
CN101852768A (en) | Workpiece flaw identification method based on compound characteristics in magnaflux powder inspection environment | |
CN103234976A (en) | Warp knitting machine cloth flaw on-line visual inspection method based on Gabor transformation | |
CN104574353A (en) | Surface defect judgment method based on visual saliency | |
CN107093175B (en) | A kind of power battery welding quality detection method | |
CN112381870B (en) | Binocular vision-based ship identification and navigational speed measurement system and method | |
CN103955922A (en) | Method for detecting flaws of printed fabric based on Gabor filter | |
CN106846313A (en) | Surface Flaw Detection method and apparatus | |
CN104751474A (en) | Cascade quick image defect segmentation method | |
CN103778627A (en) | Sea oil spill detection method based on SAR image | |
Ji et al. | Apple grading method based on features of color and defect | |
CN105225216A (en) | Based on the Iris preprocessing algorithm of space apart from circle mark rim detection | |
CN114972356A (en) | Plastic product surface defect detection and identification method and system | |
Zhao et al. | The edge detection of river model based on self-adaptive Canny Algorithm and connected domain segmentation | |
CN103729856A (en) | Method for cloth defect detection by utilizing S transformation signal extraction | |
CN103258218B (en) | Masking-out detects skeleton generating method, device, defect inspection method and device | |
CN108009574A (en) | A kind of rail clip detection method | |
CN113435460A (en) | Method for identifying brilliant particle limestone image |
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 |