CN107274409A - woven fabric defect segmentation method - Google Patents
woven fabric defect segmentation method Download PDFInfo
- Publication number
- CN107274409A CN107274409A CN201710475910.9A CN201710475910A CN107274409A CN 107274409 A CN107274409 A CN 107274409A CN 201710475910 A CN201710475910 A CN 201710475910A CN 107274409 A CN107274409 A CN 107274409A
- Authority
- CN
- China
- Prior art keywords
- defect
- image
- noise reduction
- woven fabric
- defect image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000011218 segmentation Effects 0.000 title claims abstract description 23
- 239000002759 woven fabric Substances 0.000 title claims abstract description 21
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 238000006243 chemical reaction Methods 0.000 claims abstract description 8
- 238000005070 sampling Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30124—Fabrics; Textile; Paper
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Treatment Of Fiber Materials (AREA)
Abstract
The invention discloses a kind of woven fabric defect segmentation method, it comprises the following steps：S1, the defect image for obtaining woven fabric of the surface with fault；S2, using fourth order PDEs algorithm and relaxation median filtering algorithm to the defect image carry out mixing noise reduction；S3, binary conversion treatment is carried out to the defect image after noise reduction to obtain a binary image.The present invention carries out mixing noise reduction using fourth order PDEs algorithm and relaxation median filtering algorithm to defect image, preferably remains fault information while carrying out effectively smooth to noise, improves segmentation effect.
Description
Technical field
The present invention relates to woven fabric technical field of quality detection, more particularly to a kind of woven fabric defect segmentation method.
Background technology
Usually required in the application of woven fabric surface-defect detection method with defect segmentation step, it is intended to judge material
Expect whether surface has fault and display defect position and shape.However, generally there is the interference of noise in image to be split, it is
The erroneous judgement to normal picture is avoided, it is existing typically directly to use threshold method, such as maximum between-cluster variance threshold value, iteration threshold, heredity
Algorithm threshold value etc., defect regions that can not be in Accurate Segmentation image when carrying out binaryzation to image.On the other hand, influence segmentation effect
Another key factor of fruit is the interference of material surface texture and picture noise, and the method for exclusive PCR typically uses image
Noise reduction technology.But most of Image Denoisings have equivalent weakening effect to fault information and interference information, are unfavorable for figure
The defect segmentation of picture.
The content of the invention
The problem of present invention exists for prior art and deficiency are there is provided a kind of woven fabric defect segmentation method, according to machine
The characteristics of picture noise of fabric is disturbed, is entered using fourth order PDEs and relaxation median filtering algorithm mixing to defect image
Row noise reduction, defect segmentation is carried out after carrying out noise reduction to image again.
The present invention is to solve above-mentioned technical problem by following technical proposals：
The present invention provides a kind of woven fabric defect segmentation method, and its feature is that it comprises the following steps：
S1, the defect image for obtaining woven fabric of the surface with fault；
S2, using fourth order PDEs algorithm and relaxation median filtering algorithm to the defect image carry out mixing noise reduction；
S3, binary conversion treatment is carried out to the defect image after noise reduction to obtain a binary image.
For the difference between fault information and interference information, this programme is used has preferable retention to fault information
Image denoising method.
Fourth order PDEs noise-reduction method has edge gradient discrimination function and suppresses smooth work at edge gradient
With, but it is poor to high peak noise smoothing effect, relaxation medium filtering can make up the defect.At present, mixing quadravalence is partially micro-
The noise reduction technology of equation and relaxation medium filtering is divided to be applied to yet in the segmentation step that planar materials surface-defect is detected.
It is preferred that in step s 2, circulation is calculated using the quadravalence Laplce in the fourth order PDEs algorithm successively
Son and the relaxation median filtering algorithm carry out noise reduction to the defect image after down-sampling, maximize and protect while eliminating noise jamming
Stay defect regions feature.
It is preferred that including between step S1 and S2：Down-sampling is carried out to the defect image；
Include between step S2 and S3：Defect image after the noise reduction is carried out interpolation processing to obtain and original defect
The size identical defect image of dot image；
In step s3, the defect image after interpolation processing is carried out binary conversion treatment to obtain the binary image.
It is preferred that in step s3, binary conversion treatment is carried out to the defect image after the noise reduction using threshold method.
It is preferred that the threshold method carries out the operation of binaryzation：Calculate all elements in the defect image after the noise reduction
Mean μ and standard deviation sigma, when the grey of pixel in the defect image after the noise reduction is between the σ of μ ± 3, then the pixel
It is entered as 1；Otherwise the pixel is entered as 0.
On the basis of common sense in the field is met, above-mentioned each optimum condition can be combined, and produce each preferable reality of the present invention
Example.
The positive effect of the present invention is：
1st, the present invention carries out mixing drop using fourth order PDEs algorithm and relaxation median filtering algorithm to defect image
Make an uproar, preferably remain fault information while carrying out effectively smooth to noise, improve segmentation effect；
2nd, dividing method of the invention is applicable to polytype woven face defect detection.
Brief description of the drawings
Fig. 1 is the flow chart of the woven fabric defect segmentation method of present pre-ferred embodiments.
Fig. 2 is the schematic diagram of the defect image of the woven fabric of present pre-ferred embodiments.
Fig. 3 is the schematic diagram that image obtained by two-dimensional empirical mode decomposition is carried out to Fig. 1.
Fig. 4 is the schematic diagram that gained image after mixing noise reduction is carried out to Fig. 2.
Fig. 5 is the schematic diagram that binarization operation acquired results are directly carried out to Fig. 3.
Fig. 6 is the schematic diagram that binarization operation acquired results are carried out to Fig. 4.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained on the premise of creative work is not made, belongs to the scope of protection of the invention.
As shown in figure 1, the present embodiment provides a kind of woven fabric defect segmentation method, it comprises the following steps：
Step 101, the defect image for obtaining woven fabric of the surface with fault；
Step 102, to the defect image carry out down-sampling；
Step 103, circulation use quadravalence Laplace operator and the relaxation in the fourth order PDEs algorithm successively
Median filtering algorithm carries out noise reduction to the defect image after down-sampling, is maximized while eliminating noise jamming and retains defect regions
Feature；
Step 104, interpolation processing is carried out to the defect image after the noise reduction to obtain the size with original defect image
Identical defect image；
Step 105, threshold method is used to carry out binary conversion treatment to the defect image after interpolation processing to obtain a binaryzation
Image.
Wherein, the threshold method carries out the operation of binaryzation：
The mean μ and standard deviation sigma of all elements in the defect image after the interpolation processing are calculated, after the interpolation processing
Defect image in pixel grey between the σ of μ ± 3 when, then the pixel is entered as 1；Otherwise the pixel assignment
For 0, so as to obtain the binary image.
Name a specific example to illustrate technical scheme, to enable those skilled in the art more
Understand technical scheme well：
See Fig. 2, obtain the defect image of woven fabric of the surface with fault, see Fig. 3, it is that defect image carries out two dimension
Image obtained by empirical mode decomposition, to the image cycle in Fig. 3 successively using fourth order PDEs algorithm and relaxation intermediate value filter
Ripple algorithm carries out noise reduction, and the image of acquisition is as shown in Figure 4.
Directly to Fig. 3 using acquired results after threshold method progress binarization operation as shown in figure 5, using threshold method to Fig. 4
Acquired results are as shown in fig. 6, pass through comparison diagram 5 and Fig. 6 after progress binarization operation, it is clear that the segmentation knot of the gained after noise reduction
Fruit wants the excellent segmentation result without gained after noise reduction.
Although the foregoing describing the embodiment of the present invention, it will be appreciated by those of skill in the art that these
It is merely illustrative of, protection scope of the present invention is defined by the appended claims.Those skilled in the art is not carrying on the back
On the premise of principle and essence from the present invention, various changes or modifications can be made to these embodiments, but these are changed
Protection scope of the present invention is each fallen within modification.
Claims (5)
1. a kind of woven fabric defect segmentation method, it is characterised in that it comprises the following steps：
S1, the defect image for obtaining woven fabric of the surface with fault；
S2, using fourth order PDEs algorithm and relaxation median filtering algorithm to the defect image carry out mixing noise reduction；
S3, binary conversion treatment is carried out to the defect image after noise reduction to obtain a binary image.
2. woven fabric defect segmentation method as claimed in claim 1, it is characterised in that in step s 2, circulation is used successively
Quadravalence Laplace operator and the relaxation median filtering algorithm in the fourth order PDEs algorithm is to the fault after down-sampling
Image carries out noise reduction.
3. woven fabric defect segmentation method as claimed in claim 2, it is characterised in that include between step S1 and S2：It is right
The defect image carries out down-sampling；
Include between step S2 and S3：Defect image after the noise reduction is carried out interpolation processing to obtain and original fault figure
The size identical defect image of picture；
In step s3, the defect image after interpolation processing is carried out binary conversion treatment to obtain the binary image.
4. woven fabric defect segmentation method as claimed in claim 1, it is characterised in that in step s3, using threshold method pair
Defect image after the noise reduction carries out binary conversion treatment.
5. woven fabric defect segmentation method as claimed in claim 4, it is characterised in that the threshold method carries out the behaviour of binaryzation
Make：Calculate in the mean μ and standard deviation sigma of all elements in the defect image after the noise reduction, the defect image after noise reduction
When the grey of pixel is between the σ of μ ± 3, then the pixel is entered as 1；Otherwise the pixel is entered as 0.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710475910.9A CN107274409A (en) | 2017-06-21 | 2017-06-21 | woven fabric defect segmentation method |
| PCT/CN2017/107924 WO2018233167A1 (en) | 2017-06-21 | 2017-10-27 | Woven fabric defect segmentation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710475910.9A CN107274409A (en) | 2017-06-21 | 2017-06-21 | woven fabric defect segmentation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107274409A true CN107274409A (en) | 2017-10-20 |
Family
ID=60069542
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710475910.9A Pending CN107274409A (en) | 2017-06-21 | 2017-06-21 | woven fabric defect segmentation method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107274409A (en) |
| WO (1) | WO2018233167A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018233167A1 (en) * | 2017-06-21 | 2018-12-27 | 江阴芗菲服饰有限公司 | Woven fabric defect segmentation method |
| CN109858485A (en) * | 2019-01-25 | 2019-06-07 | 东华大学 | A kind of fabric defects detection method based on LBP and GLCM |
| CN115661113A (en) * | 2022-11-09 | 2023-01-31 | 浙江酷趣智能科技有限公司 | Moisture-absorbing and sweat-releasing fabric and preparation process thereof |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113706521A (en) * | 2021-09-08 | 2021-11-26 | 常州市新创智能科技有限公司 | Carbon fiber surface hairiness detection method and device, storage medium and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070133895A1 (en) * | 2005-12-08 | 2007-06-14 | Samsung Electronics Co.; Ltd | Method for filtering image noise using pattern information |
| CN104657983A (en) * | 2015-01-20 | 2015-05-27 | 浙江理工大学 | Method for detecting densities of fabric hairballs based on Gabor filtering |
| CN105261003A (en) * | 2015-09-10 | 2016-01-20 | 西安工程大学 | Defect point detection method on basis of self structure of fabric |
| CN106780464A (en) * | 2016-12-15 | 2017-05-31 | 东华大学 | A kind of fabric defect detection method based on improvement Threshold segmentation |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105701477B (en) * | 2016-02-19 | 2017-07-14 | 中原工学院 | A kind of fabric defect detection method based on Stationary Wavelet Transform vision significance |
| CN107274409A (en) * | 2017-06-21 | 2017-10-20 | 江阴芗菲服饰有限公司 | woven fabric defect segmentation method |
-
2017
- 2017-06-21 CN CN201710475910.9A patent/CN107274409A/en active Pending
- 2017-10-27 WO PCT/CN2017/107924 patent/WO2018233167A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070133895A1 (en) * | 2005-12-08 | 2007-06-14 | Samsung Electronics Co.; Ltd | Method for filtering image noise using pattern information |
| CN104657983A (en) * | 2015-01-20 | 2015-05-27 | 浙江理工大学 | Method for detecting densities of fabric hairballs based on Gabor filtering |
| CN105261003A (en) * | 2015-09-10 | 2016-01-20 | 西安工程大学 | Defect point detection method on basis of self structure of fabric |
| CN106780464A (en) * | 2016-12-15 | 2017-05-31 | 东华大学 | A kind of fabric defect detection method based on improvement Threshold segmentation |
Non-Patent Citations (1)
| Title |
|---|
| 芦碧波，李阳，王永茂: "结合松弛中值滤波的高阶彩色图像迭代去噪算法", 《应用光学》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018233167A1 (en) * | 2017-06-21 | 2018-12-27 | 江阴芗菲服饰有限公司 | Woven fabric defect segmentation method |
| CN109858485A (en) * | 2019-01-25 | 2019-06-07 | 东华大学 | A kind of fabric defects detection method based on LBP and GLCM |
| CN115661113A (en) * | 2022-11-09 | 2023-01-31 | 浙江酷趣智能科技有限公司 | Moisture-absorbing and sweat-releasing fabric and preparation process thereof |
| CN115661113B (en) * | 2022-11-09 | 2023-05-09 | 浙江酷趣智能科技有限公司 | Moisture-absorbing sweat-releasing fabric and preparation process thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2018233167A1 (en) | 2018-12-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107274409A (en) | woven fabric defect segmentation method | |
| CN103208104A (en) | Non-local theory-based image denoising method | |
| Kumar et al. | Edge detection and denoising medical image using morphology | |
| Molinara et al. | Automatic segmentation of the pectoral muscle in mediolateral oblique mammograms | |
| Taha et al. | Partial Differential Equations and Digital Image Processing: A Review | |
| US20190019272A1 (en) | Noise reduction for digital images | |
| Xiansheng | An edge detection new algorithm based on laplacian operator | |
| Siddique et al. | Multi-focus image fusion using block-wise color-principal component analysis | |
| Li et al. | Soft clustering guided image smoothing | |
| CN105678704B (en) | A kind of non local intermediate value blind landing method for de-noising of view-based access control model perception | |
| Suresha et al. | Kumaraswamy distribution based bi-histogram equalization for enhancement of microscopic images | |
| Kim et al. | Automated hedcut illustration using isophotes | |
| Qiegen et al. | Adaptive image decomposition by improved bilateral filter | |
| CN109035171B (en) | Reticulate pattern face image restoration method | |
| CN105931192A (en) | Image texture filtering method based on weighted median filtering | |
| CN113487496B (en) | Image denoising method, system and device based on pixel type inference | |
| CN110378907A (en) | The processing method and computer equipment of image, storage medium in intelligent refrigerator | |
| Chudasama et al. | Survey on Various Edge Detection Techniques on Noisy Images | |
| Ao et al. | Edge saliency map detection with texture suppression | |
| Huang et al. | Hyperspectral image denoising with multiscale low-rank matrix recovery | |
| Tasdizen et al. | Boundary estimation from intensity/color images with algebraic curve models | |
| Aishwarya et al. | A Comparative study of edge detection in noisy images using BM3D filter | |
| Jain et al. | Spatially localized implementation of SSR and DSR for image denoising | |
| Maiti et al. | A novel image inpainting framework using regression | |
| Kumar et al. | Adaptive patch based texture synthesis using wavelet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171020 |