CN105844675A - Color cluster analysis method of yarn-dyed fabric - Google Patents
Color cluster analysis method of yarn-dyed fabric Download PDFInfo
- Publication number
- CN105844675A CN105844675A CN201610172666.4A CN201610172666A CN105844675A CN 105844675 A CN105844675 A CN 105844675A CN 201610172666 A CN201610172666 A CN 201610172666A CN 105844675 A CN105844675 A CN 105844675A
- Authority
- CN
- China
- Prior art keywords
- cluster
- color
- dyed fabric
- yarn dyed
- yarn
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Spectrometry And Color Measurement (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a color cluster analysis method of a yarn-dyed fabric. The color cluster analysis method of the yarn-dyed fabric comprises the steps of utilizing median filtering (3*3 and 5*5) with different template dimensions for performing filtering de-noising processing on the original image of the yarn-dyed fabric, converting yarn dyed fabric image to an RGB color space to an Lab color space, and finally dividing the color cluster of the yarn-dyed fabric by means of a K means clustering algorithm, thereby obtaining the number and kinds of color yarns of the yarn-dyed fabric.
Description
Technical field
The present invention relates to automatic identification technology field, particularly to the color cluster analysis side of a kind of yarn dyed fabric
Method.
Background technology
Color cluster is analyzed and is produced and an indispensable step during fabric analysis as fabric, and it can
To be used for division and the classification of dyed yarn in fabric.Although digital image processing techniques are divided at textile image
Analysis field quickly grows, but energy is still concentrated on dividing of fabric gray level image by previous many research worker
Analysis, but the visually-perceptible of coloured image is more more meaningful than gray level image, therefore a kind of yarn dyed fabric color of exploitation is gathered
The method that class divides automatically is necessary.As a rule, the method that these color clusters divide automatically is permissible
It is divided into three classes: method based on fuzzy C-means clustering, method based on neutral net, based on rectangular histogram threshold
The method of value segmentation.
Although the method that color cluster divides is numerous, but said method is applied to PRINTED FABRIC mostly and monochrome is knitted
The analysis of thing, the research to yarn dyed fabric is less.Additionally, carry out clustering at RGB color, to bright
Change obvious textile image on degree and have preferable classifying quality, and to there being the most close brightness value, at tone
Changing obvious textile image in value, its classifying quality is bad.Clustering is carried out at HSL color space, right
There are the most close brightness value and tone value, intensity value changes obvious textile image, its classifying quality
The best.
Summary of the invention
The color cluster that it is an object of the invention to provide a kind of yarn dyed fabric analyzes method, by image procossing skill
Art, utilizes the medium filtering of different templates size to remove the noise on textile image, in Lab space by means of K
Means clustering algorithm textile image of checking colors carries out the division of color cluster, thus obtains kind and the number of dyed yarn.
For realizing object above, the color cluster that the invention provides a kind of yarn dyed fabric analyzes method, including following
Step:
S1, it is the subimage in tri-Color Channels of R, G, B by the yarn dyed fabric picture breakdown under rgb space；
S2, the noise on removal subimage surface；
S3, three sub-image reconstructions after noise will be removed become piece image；
S4, the subimage after noise will be removed it will be transformed into Lab color space from RGB color；
S5, color cluster to yarn dyed fabric divide.
As preferably, in step S2, by the way of medium filtering, remove picture noise.
As preferably, the mode of medium filtering uses the medium filtering template of 3 × 3 or 5 × 5.
As preferably, step 4 specifically includes following steps:
S4.1, image is transformed into XYZ color space from RGB color；
S4.2, textile image according to XYZ color space are converted to the textile image of Lab color space.
As preferably, step S4.1 specifically includes following steps:
S4.11, by the value of R, G, B divided by 255, respectively obtain r0、g0、b0Value；
S4.12, by r0、g0、b0Value compare with 0.04045 respectively, if r0> 0.04045, then OtherwiseIf g0> 0.04045, thenOtherwise
If b0> 0.04045, thenOtherwise
S4.13, pass throughObtain the fabric figure of XYZ color space
Picture.
As preferably, step S4.2 specifically includes following steps:
S4.21, by X, Y, Z respectively compared with 0.008856, if X > 0.008856, thenNo
ThenIf Y > 0.008856, thenOtherwiseIf
Z > 0.008856, thenOtherwise
S4.22, pass throughObtain the textile image of Lab color space.
As preferably, the color cluster of yarn dyed fabric is divided by step S5 by K mean cluster partitioning.
As preferably, K mean cluster partitioning comprises the following steps:
S5.1, determine classification number k；
S5.2, from set { x} arbitrarily chooses k object as initial cluster center: Z1(1),Z2(1),…,
Zk(1)；
S5.3, according to set the distance of the element in x} and these cluster centres, and assign these to respectively with
Its most like cluster；
S5.4, calculate the cluster centre z of each obtained new clusterj(k+1), j=1,2 ..., k, directly
To cluster CjK in (), all elements to the square distance of new cluster centre and minimizes value；
If S5.5 is zj(k+1)=zj(k), then the process of iteration will stop, otherwise repeating step S5.3 and
S5.4, until each cluster no longer changes.
Owing to tri-Color Channels of R, G, the B under RGB color are linear relationship, under its space
The color cluster of yarn dyed fabric being carried out division and can produce the classification of some mistakes, it is suitable only on luminance component
There is the clustering of the yarn dyed fabric image of larger difference, and Lab color space is to most yarn dyed fabric figure
As there being preferable clustering effect, even if having similar brightness and tone value between yarn dyed fabric image, and full
With in angle value distinguishing in the case of, the Lab color space textile image that still is able to check colors has preferable clustering
Effect.
Accompanying drawing explanation
Fig. 1 is the flow chart that in the present invention, yarn dyed fabric color cluster divides.
Fig. 2 is 3 × 3 medium filtering schematic diagram in the present invention.
Fig. 3 is the schematic diagram of Lab color space in the present invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to the present invention's
Each embodiment is explained in detail.
The first embodiment of the present invention proposes a kind of method dividing the color cluster of yarn dyed fabric,
The method mainly includes five parts, as it is shown in figure 1, comprise the following steps:
1) picture breakdown.In yarn dyed fabric picture breakdown under rgb space is tri-Color Channels of R, G, B
Subimage；
2) medium filtering.Utilize medium filtering that the subimage in three Color Channels is processed, remove figure
The noise of image surface；
3) image reconstruction.Subimage in three Color Channels after medium filtering is reconstructed piece image；
4) color space conversion.Filtered image is transformed into Lab color space from RGB color；
5) clustering.The color cluster of yarn dyed fabric is entered in Lab space by means of K means clustering algorithm
Row divides, thus obtains number and the kind of dyed yarn.
In digital image processing field, a coloured image is seen as a three-dimensional character matrix, the most often
One pixel is all formed by stacking by R (red), G (green), three color value of B (blue), Mei Geyan
The span of colour is all 0～255.So picture can be decomposed into three respectively containing only R value, G-value,
The subimage of B value, is known respectively as red channel subimage, green channel subimage, blue channel subimage.
By the subimage that picture breakdown is three Color Channels, it is the basic handling method in image procossing, application
Extensively, no longer describe at this.
Under normal circumstances, uneven, when textile image can be caused to gather due to the filoplume of fabric face and illumination
The appearance of noise, the present invention uses the method for two dimension median filter to remove picture noise, reaches to strengthen picture quality
Effect, in order to improve color analysis accuracy rate.
Two dimension median filter is a kind of can effectively suppression at the nonlinear properties of noise based on sequencing statistical theory
Reason technology, its ultimate principle is in each point value in a neighborhood of this point of the value of any in digital picture
Value replaces, and allows the actual value that the pixel value of surrounding is close, thus eliminates the noise spot isolated.Choose one 3 ×
As a example by the medium filtering of 3 template sizes, as illustrated in fig. 2, it is assumed that f (x, y) represents original-gray image, and g (x, y)
Representing the image after medium filtering, in original-gray image, the gray value of some pixel is 192, and it is adjacent
The gray value of eight pixels be 0,32,64,96,128,160,225 and 255, utilize intermediate value to filter
After ripple processes, the gray value of this pixel becomes 128.
Medium filtering is used to remove the noise in picture, and its used filtering size template size has 3 × 3,
5 × 5,7 × 7,9 × 9 etc..Template size is the biggest, and the effect removing noise is the best, but relative denoising
Picture clarity afterwards also can be the lowest, and picture quality required in comprehensive subsequent experimental considers, chooses 3
The medium filtering of × 3 and 5 × 5 two different templates sizes is in tri-Color Channels of yarn dyed fabric R, G, B
Subimage processes, and enters subimage with the medium filtering of 3 d effect graph two kinds of different templates sizes of reflection
After row convolution algorithm, the change that image occurs.The yarn dyed fabric image of sample 1 through medium filtering process after, its
Filoplume and the fiber fines on surface are removed, and the colouring information of yarn is effectively preserved.
Traditional color space is to be made up of tone, saturation and brightness, but people are difficult at RGB and HSL
Under space, color is identified exactly.In the present invention, a kind of Lab color space is proposed (such as Fig. 3 institute
Show), compared with traditional color space, the colour gamut of Lab color space is bigger, closer to the vision of the mankind
Sensing, compared with the uneven distribution of RGB color, it is devoted to the uniformity of perception.In Lab space
L * component represent brightness, the opposition dimension of a and b representation in components color, therefore, it can using L * component as
The standard that brightness is distinguished, using the component of a and b as the standard of color evaluation.
Because the yarn dyed fabric image collected is RGB image, therefore to be turned from RGB color by textile image
Change to Lab color space and carry out color cluster division, the most first image is turned from RGB color
Changing to XYZ color space, its concrete transformation process is as follows:
(1) by the value of R, G, B divided by 255, r is respectively obtained0、g0、b0Value；
(2) by r0、g0、b0Value compare with 0.04045 respectively, if r0> 0.04045, then
OtherwiseIf g0> 0.04045, thenOtherwiseIf b0> 0.04045,
ThenOtherwise
(3) formula is passed throughObtain the fabric of XYZ color space
Image.
After obtaining the textile image of XYZ color space, can proceed textile image to change, final
To the textile image of Lab color space, its detailed process is as follows:
(1) if X > 0.008856, soOtherwiseSimilarly, utilize
Y and Z component are processed by the method, thus obtain y, z；
(2) pass throughObtain the textile image of Lab color space.
After achieving above-mentioned transformation process, textile image is transformed into Lab color space, L from RGB color
In the range of the value of component is distributed in [0,100], in the range of the value of a, b component is distributed in [-128 ,+127].
Cluster analysis refers to the set of physics or abstract object is grouped into the multiple classes being made up of the object being similar to
Analysis process, and K means clustering algorithm is as a conventional hard clustering algorithm, can be used for image
The division of color cluster.It is the representative of typical object function clustering method based on prototype, data point is arrived
The Euclidean distance of prototype, as the object function optimized, utilizes function to ask the method for extreme value to obtain interative computation
Regulation rule.
The present invention chooses and comprises the yarn dyed fabric of three kinds of different colours yarns as object of study, utilizes K average to gather
The method of class is divided into different clusters, and concrete step is as follows:
(1) classification number k is determined；
(2) { x} arbitrarily chooses k object as initial cluster center: Z from set1(1),Z2(1),…,
Zk(1)；
(3) in the iterative algorithm step of kth rank, according to set the element in x} and these cluster centres away from
From, assign these to the cluster most like with it respectively, if | | x-zj(k) | | ＜ | | x-zi(k) | |, then x ∈ Cj(k),
To all of i=1,2 ..., k, i ≠ j sets up, wherein zjK () is cluster CjThe cluster centre of (k)；
(4) the cluster centre z of each obtained new cluster is calculatedj(k+1), j=1,2 ..., k, until
Cluster CjK in (), all elements to the square distance of new cluster centre and minimizes value, new cluster centre
Can be expressed asWherein, NjIt is cluster CjThe number of element in (k)
Mesh；
(5) to j=1,2 ..., for k, if zj(k+1)=zj(k), then the process of iteration will be stopped
Only, the most constantly circulation (3) is to (4) process, until each cluster no longer changes.
Under Lab space, utilize K means clustering algorithm that the color cluster of sample is divided, the method
The color cluster of yarn dyed fabric can be divided, but the dendrogram picture marked off has a large amount of isolated making an uproar
Sound point, therefore, before cluster analysis, can remove partial noise by means of the medium filtering of 3 × 3,
Improve the clustering effect of yarn dyed fabric image.
Under Lab color space, yarn dyed fabric image to sample carries out the effect of color division and is better than RGB face
Under the colour space, the color cluster of yarn dyed fabric divides, and has similar brightness value, different colors especially between yarn
During tone pitch, such as green yarn and white yarn, they are difficult to divided under RGB color, but
There is under Lab space preferable clustering effect.Meanwhile, yarn dyed fabric image is through 5 × 5 template sizes
Color cluster after medium filtering processes divides effect, and ratio is after the medium filtering of 3 × 3 template sizes processes
Color cluster to divide effect more preferable, more edge noise can be removed.
The respective embodiments described above are to realize the specific embodiment of the present invention, and those of ordinary skill in the art is permissible
Understand, and in actual applications, can to it, various changes can be made in the form and details, without departing from this
Bright spirit and scope.
Claims (8)
1. the color cluster of a yarn dyed fabric analyzes method, it is characterised in that comprise the following steps:
S1, it is the subimage in tri-Color Channels of R, G, B by the yarn dyed fabric picture breakdown under rgb space；
S2, the noise on removal subimage surface；
S3, three sub-image reconstructions after noise will be removed become piece image；
S4, the subimage after noise will be removed it will be transformed into Lab color space from RGB color；
S5, color cluster to yarn dyed fabric divide.
The color cluster of yarn dyed fabric the most according to claim 1 analyzes method, it is characterised in that described
In step S2, by the way of medium filtering, remove picture noise.
The color cluster of yarn dyed fabric the most according to claim 2 analyzes method, it is characterised in that described
The mode of medium filtering uses the medium filtering template of 3 × 3 or 5 × 5.
The color cluster of yarn dyed fabric the most according to claim 1 analyzes method, it is characterised in that described
Step 4 specifically includes following steps:
S4.1, image is transformed into XYZ color space from RGB color；
S4.2, textile image according to XYZ color space are converted to the textile image of Lab color space.
The color cluster of yarn dyed fabric the most according to claim 4 analyzes method, it is characterised in that described
Step S4.1 specifically includes following steps:
S4.11, by the value of R, G, B divided by 255, respectively obtain r0、g0、b0Value；
S4.12, by r0、g0、b0Value compare with 0.04045 respectively, if r0> 0.04045, then OtherwiseIf g0> 0.04045, thenOtherwise
If b0> 0.04045, thenOtherwise
S4.13, pass throughObtain the fabric figure of XYZ color space
Picture.
The color cluster of yarn dyed fabric the most according to claim 4 analyzes method, it is characterised in that described
Step S4.2 specifically includes following steps:
S4.21, by X, Y, Z respectively compared with 0.008856, if X > 0.008856, thenNo
ThenIf Y > 0.008856, thenOtherwiseIf
Z > 0.008856, thenOtherwise
S4.22, pass throughObtain the textile image of Lab color space.
The color cluster of yarn dyed fabric the most according to claim 1 analyzes method, it is characterised in that described
The color cluster of yarn dyed fabric is divided by step S5 by K mean cluster partitioning.
The color cluster of yarn dyed fabric the most according to claim 7 analyzes method, it is characterised in that described
K mean cluster partitioning comprises the following steps:
S5.1, determine classification number k；
S5.2, from set { x} arbitrarily chooses k object as initial cluster center: Z1(1),Z2(1),…,
Zk(1)；
S5.3, according to set the distance of the element in x} and these cluster centres, and assign these to respectively with
Its most like cluster；
S5.4, calculate the cluster centre z of each obtained new clusterj(k+1), j=1,2 ..., k, directly
To cluster CjK in (), all elements to the square distance of new cluster centre and minimizes value；
If S5.5 is zj(k+1)=zj(k), then the process of iteration will stop, otherwise repeating said steps S5.3
And S5.4, until each cluster no longer changes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610172666.4A CN105844675A (en) | 2016-03-24 | 2016-03-24 | Color cluster analysis method of yarn-dyed fabric |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610172666.4A CN105844675A (en) | 2016-03-24 | 2016-03-24 | Color cluster analysis method of yarn-dyed fabric |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105844675A true CN105844675A (en) | 2016-08-10 |
Family
ID=56583117
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610172666.4A Pending CN105844675A (en) | 2016-03-24 | 2016-03-24 | Color cluster analysis method of yarn-dyed fabric |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105844675A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106878618A (en) * | 2017-03-07 | 2017-06-20 | 武汉华之洋科技有限公司 | A kind of image processing system and method based on FPGA |
| CN107886549A (en) * | 2017-10-18 | 2018-04-06 | 浙江工业大学之江学院 | A kind of dermatoglyphic pattern of the fabric color transfer method based on braiding grain details enhancing |
| CN109063781A (en) * | 2018-08-14 | 2018-12-21 | 浙江理工大学 | A kind of fuzzy image Fabric Design method of imitative natural colour function and form |
| CN109443542A (en) * | 2018-11-06 | 2019-03-08 | 中国矿业大学 | A kind of pressure fan on-Line Monitor Device and monitoring method based on infrared thermal imaging technique |
| CN112581432A (en) * | 2020-12-08 | 2021-03-30 | 中国纺织科学研究院有限公司 | Method and device for measuring mixing proportion of colored spun yarn, computer readable storage medium and electronic equipment |
| CN113344047A (en) * | 2021-05-24 | 2021-09-03 | 广西电网有限责任公司电力科学研究院 | Platen state identification method based on improved K-means algorithm |
| CN114324189A (en) * | 2021-12-22 | 2022-04-12 | 江苏恒力化纤股份有限公司 | Method for evaluating color uniformity of warp and weft yarns of woven fabric |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036529A (en) * | 2014-06-10 | 2014-09-10 | 浙江工业大学之江学院 | Image analysis method for embroidery fabric design colors |
| CN105354864A (en) * | 2015-09-25 | 2016-02-24 | 浙江大学 | Textile tissue color replacement simulation method with relatively high truth |
-
2016
- 2016-03-24 CN CN201610172666.4A patent/CN105844675A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036529A (en) * | 2014-06-10 | 2014-09-10 | 浙江工业大学之江学院 | Image analysis method for embroidery fabric design colors |
| CN105354864A (en) * | 2015-09-25 | 2016-02-24 | 浙江大学 | Textile tissue color replacement simulation method with relatively high truth |
Non-Patent Citations (3)
| Title |
|---|
| 张杰: "基于双面成像技术的织物纹理与颜色特征分析", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王九各: "基于图像分割的织物疵点检测与识别算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王林吉: "基于CIELAB均匀颜色空间和聚类算法的混纺测色研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106878618A (en) * | 2017-03-07 | 2017-06-20 | 武汉华之洋科技有限公司 | A kind of image processing system and method based on FPGA |
| CN106878618B (en) * | 2017-03-07 | 2020-02-21 | 武汉华之洋科技有限公司 | Image processing system and method based on FPGA |
| CN107886549A (en) * | 2017-10-18 | 2018-04-06 | 浙江工业大学之江学院 | A kind of dermatoglyphic pattern of the fabric color transfer method based on braiding grain details enhancing |
| CN107886549B (en) * | 2017-10-18 | 2021-07-30 | 浙江工业大学之江学院 | Fabric pattern color transfer method based on weaving texture detail enhancement |
| CN109063781A (en) * | 2018-08-14 | 2018-12-21 | 浙江理工大学 | A kind of fuzzy image Fabric Design method of imitative natural colour function and form |
| CN109443542A (en) * | 2018-11-06 | 2019-03-08 | 中国矿业大学 | A kind of pressure fan on-Line Monitor Device and monitoring method based on infrared thermal imaging technique |
| CN109443542B (en) * | 2018-11-06 | 2020-01-21 | 中国矿业大学 | On-line monitoring device and monitoring method for forced draught fan based on infrared thermal imaging technology |
| CN112581432A (en) * | 2020-12-08 | 2021-03-30 | 中国纺织科学研究院有限公司 | Method and device for measuring mixing proportion of colored spun yarn, computer readable storage medium and electronic equipment |
| CN113344047A (en) * | 2021-05-24 | 2021-09-03 | 广西电网有限责任公司电力科学研究院 | Platen state identification method based on improved K-means algorithm |
| CN114324189A (en) * | 2021-12-22 | 2022-04-12 | 江苏恒力化纤股份有限公司 | Method for evaluating color uniformity of warp and weft yarns of woven fabric |
| CN114324189B (en) * | 2021-12-22 | 2023-06-02 | 江苏恒力化纤股份有限公司 | Method for evaluating color uniformity of warp and weft yarns of woven fabric |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105844675A (en) | Color cluster analysis method of yarn-dyed fabric | |
| Greenfield et al. | Image recoloring induced by palette color associations | |
| CN1655583B (en) | Systems and methods for generating high compression image data files having multiple foreground planes | |
| JP2008546990A (en) | How to split white blood cells | |
| Niu et al. | Image segmentation algorithm for disease detection of wheat leaves | |
| CN110687121B (en) | Intelligent online detection and automatic grading method and system for ceramic tiles | |
| US8121401B2 (en) | Method for reducing enhancement of artifacts and noise in image color enhancement | |
| Chen et al. | Color feature extraction of Hainan Li brocade image based on RGB and HSV | |
| CN108711160B (en) | Target segmentation method based on HSI (high speed input/output) enhanced model | |
| CN114299051A (en) | Leather material surface defect detection method based on feature modeling significance detection | |
| CN107886549B (en) | Fabric pattern color transfer method based on weaving texture detail enhancement | |
| CN109816629B (en) | Method and device for separating moss based on k-means clustering | |
| CN105844676A (en) | Color cluster analysis device and color cluster analysis method for printed fabric | |
| ITTO990996A1 (en) | CLASSIFICATION METHOD OF DIGITAL IMAGES BASED ON THEIR CONTENT. | |
| CN117152159B (en) | Method and system for detecting printing flaws of complex cloth | |
| CN102496139A (en) | Image processing-based method for transforming photo to digital oil painting | |
| CN107578379A (en) | A kind of processing method of chess robot to checkerboard image | |
| CN103136729B (en) | Fuzzy vector morphological filtering method based on hypercomplex description | |
| CN110807747B (en) | Document image noise reduction method based on foreground mask | |
| CN115082741B (en) | Waste textile classification method based on image processing | |
| WO2010128683A1 (en) | Blue sky color detection technique | |
| CN103325101A (en) | Extraction method and device of color characteristics | |
| CN104778432A (en) | Image recognition method | |
| Merciol et al. | Fast image and video segmentation based on alpha-tree multiscale representation | |
| CN114155384A (en) | Method for calculating pattern clipping effect similarity of colored woven fabric |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160810 |