CN103866551A - Fabric weft inclination rapid-detection method based on machine vision - Google Patents
Fabric weft inclination rapid-detection method based on machine vision Download PDFInfo
- Publication number
- CN103866551A CN103866551A CN201410124279.4A CN201410124279A CN103866551A CN 103866551 A CN103866551 A CN 103866551A CN 201410124279 A CN201410124279 A CN 201410124279A CN 103866551 A CN103866551 A CN 103866551A
- Authority
- CN
- China
- Prior art keywords
- skew
- striped
- pixel
- image
- coordinate
- 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)
Abstract
The invention discloses a fabric weft inclination rapid-detection method based on machine vision. The method comprises the following steps that a CCD video camera is adopted to collect texture images of fabrics, the best segmentation threshold value T is automatically found by using the Otsu method, image target and background segmentation is conducted on the texture images of the fabrics according to the best segmentation threshold value T, and binary texture images of the fabrics are obtained; the stripe trend in the binary texture images of the fabrics is tracked in a window pixel searching method, isolated point noise is removed, a stripe tracking starting point and end point are determined, and finally the stripe weft inclination angle is obtained; a plurality of stripes are tracked by adopting a stripe space domain retraining method, a set of stripe weft inclination angle values are obtained, selective averaging is conducted on the stripe weft inclination angle values, and final fabric stripe image weft inclination angle values are obtained. The method effectively avoids influence on weft inclination angle detection of environment light, is high in fabric weft inclination angle detection accuracy, and can rapidly detect the fabric weft inclination angle in real time.
Description
Technical field
The present invention relates to the technical field of digital picture feature extraction, particularly a kind of fabric skew quick detecting method based on machine vision.
Background technology
Knitwear, textiles are in process, and the techniques such as dyeing, washing, stamp make fabric be subject to inhomogeneous internal stress, and the fabric that yarn count is thicker, thread count is larger, weight is larger more easily produces skew of weft in process.Skew of weft can badly influence outward appearance and the quality of clothes.The whole latitude instrument of traditional photoelectricity exists and detects dead angle, shortcoming that accuracy of detection is lower, and the application in textile industry of digital picture Feature Extraction Technology based on CCD is more and more wider, as the Density based on image processing detects, fabric defects point detection etc.The scheme of Machine oriented vision-based detection fabric skew has two kinds at present:
A kind of scheme is by utilize morphological transformation to carry out smoothing processing to textile image, thereby utilizes morphological erosion to eliminate noise pixel point, utilize morphology edge extracting method to extract weft yarn edge pixel, finally detected and calculated skew of weft angle by the straight line of Hough transfer pair edge pixel composition.These class methods have been applied a large amount of morphological method smoothed images and extraction picture edge characteristic, have been utilized Hough transfer pair straight line to detect, at fabric because surround lighting, the shadow are walked about while causing local lines disappearance, accuracy in detection can decline to a great extent, and algorithm complex is higher, can not rapidly and efficiently detect fabric skew angle.
Another kind of scheme is to utilize the self registration character of Fourier transformation amplitude spectrum to survey the texture of fabric, sample patterns Fourier modulus spectrum is carried out to binaryzation calculating, and the grain direction coefficient of the sample patterns under all angles after Fourier transformation under storage; Textured pattern to be measured and Sample Storehouse are made comparisons, draw final skew of weft angle.Although the method self adaptation degree is higher, the same with the first scheme, algorithm complex is higher, and Fourier transformation can consume a large amount of time in the time that image slices vegetarian refreshments is more.In addition, there is uncontrollable factor in the method, reduce the most at last accuracy of detection in the time choosing sample.
Summary of the invention
The object of the present invention is to provide a kind of efficiently, the fabric skew quick detecting method based on machine vision accurately, for the control system of whole latitude instrument provides real-time reference data, thereby the skew of weft of fabric is corrected in time.
The technical solution that realizes the object of the invention is: a kind of fabric skew quick detecting method based on machine vision, comprises the following steps:
Step 1, adopts ccd video camera to gather the texture image of fabric, uses large Tianjin method Automatic-searching optimal segmenting threshold T, according to optimal segmenting threshold T, cloth textured image is carried out to cutting apart of image object and background, obtains the cloth textured image of binaryzation;
Step 2, by striped trend in the cloth textured image of window pixel searching method tracking binaryzation, and removes isolated point noise, determines starting point and the terminal of the striped of following the tracks of, and finally draws striped skew of weft angle;
Step 3, adopts the method for striped spatial domain constraint to follow the tracks of multichannel striped, and obtains one group of striped skew of weft angle value, and this group striped skew of weft angle value is carried out selectively on average obtaining final cloth textured image skew of weft angle value.
Compared with prior art, its remarkable advantage is in the present invention: (1) adopts large Tianjin method to carry out self-adaption binaryzation to image, all can extract rational binaryzation weft yarn information for the textile image under different illumination conditions; (2) adopt window pixel search method to judge that striped starting and terminal point finally draws stripe angle, and in search procedure, remove isolated point noise, the simple and effective noise pixel point of having eliminated, and large degree has retained the effective information in target texture; (3) produce spatial domain effect of contraction by survey many lines and mark simultaneously on target weft yarn texture, adopt selective average method to screen these skew of weft angles that detect, select the most rational grain angle, time space complexity is low, in the effective situation of texture information, can detect accurately stripe angle.
Brief description of the drawings
Fig. 1 is the flow chart that the present invention is based on the fabric skew quick detecting method of machine vision.
Fig. 2 is the textile image after binaryzation in the inventive method, isolated point remove, angular dimension.
Fig. 3 is the fabric sample of different skew of weft angles in the embodiment of the present invention 1 design sketch after binaryzation, denoising, stripe angle mark, wherein 2 ° of (a) skew of weft angles, (b) 10 ° of skew of weft angles, (c) 20 ° of skew of weft angles, (d) skew of weft angle-7 °, (e) skew of weft angle-13 °, (f) skew of weft angle-19 °.
Fig. 4 is the fabric skew angle that measures in the embodiment of the present invention 1 and the departure degree of actual skew of weft angle.
Detailed description of the invention
Below in conjunction with the drawings and the specific embodiments, the present invention is described in further detail.
The present invention is based on the fabric skew quick detecting method of machine vision, utilize high speed camera to gather the motion video sequence of fabric in production process, extract the image of associated frame, large Tianjin method by automatic search optimal threshold is carried out binaryzation to this two field picture, and determine skew of weft angle by window pixel search method, and in search procedure, remove isolated point noise, by to many stripe search angles and adopt spatial domain constraint and selective average method reduces the influence of noise being produced by local light, in conjunction with Fig. 1, concrete steps are as follows:
Step 1, adopt ccd video camera to gather the texture image of fabric, use large Tianjin method Automatic-searching optimal segmenting threshold T, according to optimal segmenting threshold T, cloth textured image is carried out to cutting apart of image object and background, all become white pixel by the highlights pixel in light and dark striped, and the pixel of dark portion all becomes black pixel, obtain the cloth textured image of binaryzation.Specific as follows:
(1.1) according to segmentation threshold T ', in statistics texture image, impact point accounts for the ratio of the total pixel of image, and the ratio that impact point accounts for the total pixel of image is p
t, impact point average gray is designated as g
t, the background ratio that accounts for image total pixel number of counting is p
b, background dot average gray is designated as g
b, total pixel average gray g of texture image is:
g=p
t×g
t+p
b×g
b (1)
Inter-class variance Dev is expressed as:
Dev=p
t×(g
t-g)
2+p
b×(g
b-g)
2 (2)
Inter-class variance Dev is expressed equivalently as dev:
dev=p
t×p
b×(g
t-g
b)
2 (3)
(1.2) traverse maximum gradation value successively as the segmentation threshold T ' of image from the minimum gradation value of texture image, in the time that dev is maximum, segmentation threshold is now the optimal segmenting threshold T of this width texture image, with this optimal segmenting threshold, T carries out binaryzation to texture image, all become white pixel by the highlights pixel in light and dark striped, and the pixel of dark portion all becomes black pixel.
Inter-class variance Dev has weighed the uniformity of the gray space territory distribution of entire image, gray scale difference maximum between two classes that when variance is maximum, explanation goes out with this Threshold segmentation, misjudged as target or part target and misjudged as background all can cause the gray scale difference between two classes and diminish when part background, therefore when inter-class variance is maximum, corresponding threshold value T is optimal segmenting threshold, can be reasonably by target and background segment out.
Step 2, by striped trend in the cloth textured image of window pixel searching method tracking binaryzation, and removes isolated point noise, determines starting point and the terminal of the striped of following the tracks of, and finally draws striped skew of weft angle;
Cloth textured image is after binaryzation, and the highlights pixel of every travel permit line all becomes white pixel, dark portion pixel all becomes black pixel.Such travel permit line is just similar to the effect that a hollow pipe is longitudinally cut open by center line, to some extent difference be striped dark portion often due to the longitude and latitude feature of fabric itself after binaryzation, be not seal, continuous.For the time being the highlights after fringes thresholding is analogized to die here, dark portion analogizes to tube wall.The noise containing in die is often isolated point noise, the feature of these isolated point noises shows as it and surrounding pixel point has notable difference, conventionally can adopt the method that morphological image expands to remove these noises, but this method be to entire image expand, amount of calculation is slightly large, and easily erodes useful tube wall information.In search procedure, suppose that search is correct, the often travel permit line just of that tracking need be destroyed just passable by the die isolated point noise in this travel permit line like this in search procedure.
By striped trend in the cloth textured image of window pixel searching method tracking binaryzation, and remove isolated point noise, specific as follows:
(2.1) establish window size for for S × S, S is more than or equal to 3 odd number, starts from window center coordinate points (x, y), add up respectively with (x, y), (x, y ± 1) ...,
centered by window in the sum of white pixel, in the time of statistics, remove isolated point noise according to following formula:
Wherein, g (i, j) denotation coordination (i, j) pixel value of locating, C (x, y) denotation coordination (x, y) is located the number that upper and lower, left and right four direction neighbor is white pixel, current search point (x, y) place pixel grey scale is zero while being black pixel, if C (x, y) >=3, (x, y) coordinate place pixel is noise, gray value that should (x, y) coordinate place pixel is revised as 255, changes white pixel into by black pixel;
(2.2) relatively previous step with (x, y), (x, y ± 1) ...,
centered by window in the size of population of white pixel, coordinate corresponding to the maximum window center of sum is as starting point coordinate (x
0, y
0);
(2.3) abscissa adds 1 continuation search, ordinate remains unchanged, carry out white pixel number in isolated point denoising statistical window with the method in (2.1)~(2.2), select the ordinate value that has a window center coordinate that the window of maximum white pixels is corresponding and upgrade current ordinate value, abscissa successively adds 1 iteration to be upgraded N time, record last (x+N, y
t) coordinate is as terminal point coordinate;
(2.4) according to terminal point coordinate (x+N, y
t) and starting point coordinate (x
0, y
0) determine skew of weft angle.
Step 3, adopts the method for striped spatial domain constraint to follow the tracks of multichannel striped, and obtains one group of striped skew of weft angle value, and this group striped skew of weft angle value is carried out selectively on average obtaining final cloth textured image skew of weft angle value;
What obtain due to large Tianjin method is the optimal segmenting threshold of entire image, under factory's complicated production condition, due to local light or the shadow is walked about and the impact of fabric face tidiness on fabric binaryzation striped, they tend to make tube wall to have large-area disappearance at some some place, cause the lines of following the trail of to form path with the lines adjacent with it.The error causing in order to avoid as far as possible this impact, can make full use of textile image, take to follow the trail of the method for multichannel striped, because desirable fabric lines often shows as a rule and have the light and dark striped of fixed intervals, so in the time determining starting point coordinate, x coordinate is motionless, only gets one group of y coordinate figure with fixed intervals just passable.By specifying certain order, lines is followed the trail of, track out starting point coordinate and the terminal point coordinate of lines, two coordinates are connected to thick line, and be drawn on binary image, like this, the thicker marking line being drawn by the lines first tracking out can produce space constraint effect to the window search process of the lines of rear tracking.Can specify order from inside to outside or from outside to inside to follow the tracks of such multichannel striped, finally draw one group of angle value.
Adopt the method for striped spatial domain constraint to follow the tracks of multichannel striped, and obtain one group of striped skew of weft angle value, this group striped skew of weft angle value is carried out selectively on average obtaining final cloth textured image skew of weft angle value, be specially:
(3.1) adopt the method for striped spatial domain constraint to follow the tracks of M travel permit line, in the time determining the starting point coordinate of each travel permit line, x coordinate figure is consistent, get one group of y coordinate figure with fixed intervals, follow the trail of in turn starting point coordinate and the terminal point coordinate of every travel permit line, and the starting point coordinate tracking and terminal point coordinate are linked and are drawn on binary image by black marking line, the marking line that first tracks out striped produces space constraint effect to the window pixel search procedure of rear tracking striped, finally obtains one group of striped skew of weft angle value;
(3.2) this group striped skew of weft angle value is carried out selectively on average, establishing i travel permit line skew of weft angle value is θ
i, i=1,2 ..., M, L is threshold value and the L > 0 setting, and works as
time, by θ
ifrom this group striped skew of weft angle value, reject, and upgrade the mean value of this group striped skew of weft angle value, otherwise, retain this striped skew of weft angle value, the binaryzation obtaining by said process, the textile image after isolated point noise remove, angular dimension are as Fig. 2, using the final mean value of gained as cloth textured image skew of weft angle value.
The embodiment surveying below in conjunction with sample weft patterns is described in further detail the present invention.
Embodiment 1
In order to verify accuracy, speed and the feasibility of this algorithm, employing denim is laboratory sample, gathers the sample pictures that given skew of weft angle changes within the scope of-20 °~+ 20 °.Sample size is 500 × 500(pixel), part samples pictures as shown in Figure 3: adopt the more as shown in table 1 of the inventive method angle value that the fabric image detect of given skew of weft angle is drawn and actual angle value.
The skew of weft angle that table 1 the inventive method detects and actual skew of weft angle contrast table
Can find out by Fig. 4 and table 1, the present invention is based on the fabric skew quick detecting method of machine vision, following the trail of in indivedual stripeds because this travel permit line is because of surround lighting, the shadow is walked about, woven design surfacing, the impact of tidiness etc., after binaryzation with adjacent lines UNICOM somewhere, thereby can produce larger deviation while adopting window pixel search method, spatial domain effect of contraction by angular dimension can reduce this deviation, fall this accidental error by selective average filtration simultaneously, result shows that error can reasonably be controlled within the scope of 0.5 °, and because algorithm time complexity is low, actual recording at S gets 5, 10 lines are followed the tracks of, gathering image size is 500 × 500(pixel) time, whole window pixel search procedure is only 3 milliseconds together with the entirety of isolated point noise remove process and angular dimension part is consuming time, confirm that this algorithm can be at a high speed, detect accurately the skew of weft information of fabric, using the parameter as whole latitude instrument automatic calibration, be convenient to fabric to make in time correction.
In sum, it is higher that the present invention detects fabric skew angle precision, can effectively avoid the impact of surround lighting on skew of weft angle detection, and processing speed is fast, can be at a high speed, real-time detection fabric skew angle.The present invention's high-speed camera head of can arranging in pairs or groups erects a set of detection system, for whole latitude instrument control system provides real-time reference data, ensures the high quality of production of textiles.
Claims (4)
1. the fabric skew quick detecting method based on machine vision, is characterized in that, comprises the following steps:
Step 1, adopts ccd video camera to gather the texture image of fabric, uses large Tianjin method Automatic-searching optimal segmenting threshold T, according to optimal segmenting threshold T, cloth textured image is carried out to cutting apart of image object and background, obtains the cloth textured image of binaryzation;
Step 2 by striped trend in the cloth textured image of window pixel searching method tracking binaryzation, and is removed isolated point noise in this process, determines starting point and the terminal of the striped of following the tracks of, and finally draws striped skew of weft angle;
Step 3, adopts the method for striped spatial domain constraint to follow the tracks of multichannel striped, and obtains one group of striped skew of weft angle value, and this group striped skew of weft angle value is carried out selectively on average obtaining final cloth textured image skew of weft angle value.
2. the fabric skew quick detecting method based on machine vision according to claim 1, it is characterized in that, described in step 1, use large Tianjin method Automatic-searching segmentation threshold T, according to segmentation threshold T, cloth textured image is carried out to cutting apart of image object and background, specific as follows:
(1.1) according to segmentation threshold T ', in statistics texture image, impact point accounts for the ratio of the total pixel of image, and the ratio that impact point accounts for the total pixel of image is p
t, impact point average gray is designated as g
t, the background ratio that accounts for image total pixel number of counting is p
b, background dot average gray is designated as g
b, total pixel average gray g of texture image is:
g=p
t×g
t+p
b×g
b (1)
Inter-class variance Dev is expressed as:
Dev=p
t×(g
t-g)
2+p
b×(g
b-g)
2 (2)
Inter-class variance Dev is expressed equivalently as dev:
dev=p
t×p
b×(g
t-g
b)
2 (3)
(1.2) traverse maximum gradation value successively as the segmentation threshold T ' of image from the minimum gradation value of texture image, in the time that dev is maximum, segmentation threshold is now the optimal segmenting threshold T of this width texture image, with this optimal segmenting threshold, T carries out binaryzation to texture image, all become white pixel by the highlights pixel in light and dark striped, and the pixel of dark portion all becomes black pixel.
3. the fabric skew quick detecting method based on machine vision according to claim 1, is characterized in that, follows the tracks of striped trend in the cloth textured image of binaryzation described in step 2, and remove isolated point noise by window pixel searching method, specific as follows:
(2.1) establishing window size is S × S, and S is more than or equal to 3 odd number, from window center coordinate points (x, y) start, add up respectively with (x, y), (x, y ± 1) ...,
centered by window in the sum of white pixel, in the time of statistics, remove isolated point noise according to following formula:
Wherein, g (i, j) denotation coordination (i, j) pixel value of locating, C (x, y) denotation coordination (x, y) is located the number that upper and lower, left and right four direction neighbor is white pixel, current search point (x, y) place pixel grey scale is zero to be that current coordinate points is while being black pixel, if C (x, y) >=3, (x, y) coordinate place pixel is noise, gray value that should (x, y) coordinate place pixel is revised as 255, changes white pixel into by black pixel;
(2.2) relatively previous step with (x, y), (x, y ± 1) ...,
centered by window in the size of population of white pixel, coordinate corresponding to the maximum window center of sum is as starting point coordinate (x
0, y
0);
(2.3) abscissa adds 1 continuation search, ordinate remains unchanged, carry out white pixel number in isolated point denoising statistical window with the method in (2.1)~(2.2), select the ordinate value that has a window center coordinate that the window of maximum white pixels is corresponding and upgrade current ordinate value, abscissa successively adds 1 iteration to be upgraded N time, record last (x+N, y
t) coordinate is as terminal point coordinate;
(2.4) according to terminal point coordinate (x+N, y
t) and starting point coordinate (x
0, y
0) determine skew of weft angle.
4. the fabric skew quick detecting method based on machine vision according to claim 1, it is characterized in that, described in step 3, adopt the method for striped spatial domain constraint to follow the tracks of multichannel striped, and obtain one group of striped skew of weft angle value, this group striped skew of weft angle value is carried out selectively on average obtaining final cloth textured image skew of weft angle value, is specially:
(3.1) adopt the method for striped spatial domain constraint to follow the tracks of M travel permit line, in the time determining the starting point coordinate of each travel permit line, x coordinate figure is consistent, get one group of y coordinate figure with fixed intervals, follow the trail of in turn starting point coordinate and the terminal point coordinate of every travel permit line, and the starting point coordinate tracking and terminal point coordinate are linked and are drawn on binary image by black marking line, the marking line that first tracks out striped produces space constraint effect to the window pixel search procedure of rear tracking striped, finally obtains one group of striped skew of weft angle value;
(3.2) this group striped skew of weft angle value is carried out selectively on average, establishing i travel permit line skew of weft angle value is θ
i, i=1,2 ..., M, L is threshold value and the L > 0 setting, and works as
time, by θ
ifrom this group striped skew of weft angle value, reject, and upgrade the mean value of this group striped skew of weft angle value, using the final mean value of gained as cloth textured image skew of weft angle value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410124279.4A CN103866551B (en) | 2014-03-28 | 2014-03-28 | Based on the fabric skew quick detecting method of machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410124279.4A CN103866551B (en) | 2014-03-28 | 2014-03-28 | Based on the fabric skew quick detecting method of machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103866551A true CN103866551A (en) | 2014-06-18 |
CN103866551B CN103866551B (en) | 2016-04-20 |
Family
ID=50905535
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410124279.4A Expired - Fee Related CN103866551B (en) | 2014-03-28 | 2014-03-28 | Based on the fabric skew quick detecting method of machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103866551B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104766327A (en) * | 2015-04-15 | 2015-07-08 | 华侨大学 | Fabric deviation detection method and system based on image |
CN106023169A (en) * | 2016-05-13 | 2016-10-12 | 西安工程大学 | Garment-making cutting piece cross stripe alignment method based on image matching |
CN107248154A (en) * | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
CN109211918A (en) * | 2018-08-28 | 2019-01-15 | 河海大学常州校区 | A kind of fabric weft bow detection method based on weft yarn trend |
CN109741302A (en) * | 2018-12-20 | 2019-05-10 | 江南大学 | SD card form recognition system and method based on machine vision |
CN113780185A (en) * | 2021-09-13 | 2021-12-10 | 常州市宏发纵横新材料科技股份有限公司 | Weft angle detection method and device based on carbon fibers and storage medium |
CN116200932A (en) * | 2023-03-03 | 2023-06-02 | 常州宏大智慧科技有限公司 | Fabric weft straightening method based on machine vision |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US2355465A (en) * | 1943-10-02 | 1944-08-08 | United Merchants & Mfg | Device to detect skew in fabrics |
CN1715551A (en) * | 2004-06-28 | 2006-01-04 | 宫元九 | Detecting method and device for textile bias filling |
CN1958934A (en) * | 2006-11-24 | 2007-05-09 | 东华大学 | Intelligent control method for tentering of weft through embedding type camera shot |
CN103234969A (en) * | 2013-04-12 | 2013-08-07 | 江苏大学 | Method for measuring fabric weft density based on machine vision |
-
2014
- 2014-03-28 CN CN201410124279.4A patent/CN103866551B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US2355465A (en) * | 1943-10-02 | 1944-08-08 | United Merchants & Mfg | Device to detect skew in fabrics |
CN1715551A (en) * | 2004-06-28 | 2006-01-04 | 宫元九 | Detecting method and device for textile bias filling |
CN1958934A (en) * | 2006-11-24 | 2007-05-09 | 东华大学 | Intelligent control method for tentering of weft through embedding type camera shot |
CN103234969A (en) * | 2013-04-12 | 2013-08-07 | 江苏大学 | Method for measuring fabric weft density based on machine vision |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104766327A (en) * | 2015-04-15 | 2015-07-08 | 华侨大学 | Fabric deviation detection method and system based on image |
CN104766327B (en) * | 2015-04-15 | 2017-06-09 | 华侨大学 | Fabric bias detecting method and system based on image |
CN106023169A (en) * | 2016-05-13 | 2016-10-12 | 西安工程大学 | Garment-making cutting piece cross stripe alignment method based on image matching |
CN106023169B (en) * | 2016-05-13 | 2019-02-19 | 西安工程大学 | A method of the clothing cut-parts based on images match are to horizontal stripe |
CN107248154A (en) * | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
CN109211918A (en) * | 2018-08-28 | 2019-01-15 | 河海大学常州校区 | A kind of fabric weft bow detection method based on weft yarn trend |
CN109741302A (en) * | 2018-12-20 | 2019-05-10 | 江南大学 | SD card form recognition system and method based on machine vision |
CN113780185A (en) * | 2021-09-13 | 2021-12-10 | 常州市宏发纵横新材料科技股份有限公司 | Weft angle detection method and device based on carbon fibers and storage medium |
CN116200932A (en) * | 2023-03-03 | 2023-06-02 | 常州宏大智慧科技有限公司 | Fabric weft straightening method based on machine vision |
Also Published As
Publication number | Publication date |
---|---|
CN103866551B (en) | 2016-04-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103866551B (en) | Based on the fabric skew quick detecting method of machine vision | |
CN113781402B (en) | Method and device for detecting scratch defects on chip surface and computer equipment | |
CN113643289B (en) | Fabric surface defect detection method and system based on image processing | |
CN109325935B (en) | Power transmission line detection method based on unmanned aerial vehicle image | |
CN109345554A (en) | A kind of adhesion mushroom vision in-situ measuring method based on RGB-D camera | |
CN102236784A (en) | Screen area detection method and system | |
CN108364291A (en) | Grey cloth rapid detection method based on computer vision technique | |
CN109211918B (en) | Fabric bow weft detection method based on weft trend | |
CN107392849B (en) | Target identification and positioning method based on image subdivision | |
CN109540925B (en) | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator | |
CN107358628B (en) | Linear array image processing method based on target | |
CN109580652A (en) | A kind of quality of battery pole piece detection method, electronic equipment and storage medium | |
CN104021561A (en) | Fabric fuzzing and pilling image segmentation method based on wavelet transformation and morphological algorithm | |
CN110047063B (en) | Material drop detection method, device, equipment and storage medium | |
CN115100206B (en) | Printing defect identification method for textile with periodic pattern | |
US20180268538A1 (en) | Facilitating anomaly detection for a product having a pattern | |
CN115187602A (en) | Injection molding part defect detection method and system based on image processing | |
CN109754440A (en) | A kind of shadow region detection method based on full convolutional network and average drifting | |
CN110111322A (en) | A kind of tunnel defect identifying system based on image | |
CN104778710A (en) | Morphological image edge detecting method based on quantum theory | |
CN108196729A (en) | A kind of finger tip point rapid detection method based on infrared video | |
CN111652844B (en) | X-ray defect detection method and system based on digital image region growing | |
CN109961065A (en) | A kind of surface vessel object detection method | |
CN108256385A (en) | The front vehicles detection method of view-based access control model | |
CN108492306A (en) | A kind of X-type Angular Point Extracting Method based on image outline |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160420 Termination date: 20200328 |