CN106503681B - A kind of digital picture stage division of wool scale layer - Google Patents
A kind of digital picture stage division of wool scale layer Download PDFInfo
- Publication number
- CN106503681B CN106503681B CN201610966643.0A CN201610966643A CN106503681B CN 106503681 B CN106503681 B CN 106503681B CN 201610966643 A CN201610966643 A CN 201610966643A CN 106503681 B CN106503681 B CN 106503681B
- Authority
- CN
- China
- Prior art keywords
- wool
- digital picture
- image
- peak point
- binaryzation
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/695—Preprocessing, e.g. image segmentation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Treatment Of Fiber Materials (AREA)
Abstract
The present invention relates to a kind of digital picture stage divisions of wool scale layer, comprising the following steps: carries out denoising to the digital picture got；Digital picture after denoising is subjected to binaryzation；The straight-line detection at wool edge is carried out to the digital picture after binaryzation, and chooses suitable region；Wool fiber in the appropriate area of selection is projected in horizontal direction and obtains the waveform of reaction wool scale present position；The first derivative and second dervative for the wavy curve being calculated determine peak point according to the value of first derivative and second dervative；The spacing for calculating adjacent peak point, is classified according to the spacing of adjacent peak point.The present invention can quickly equitably evaluate wool modifying effect.
Description
Technical field
The present invention relates to wool scale layer classification technique fields, more particularly to a kind of digital picture point of wool scale layer
Grade method.
Background technique
Wool is a kind of natural good protein fiber, is one of the important raw material of modern textile industry.Wool has
Many excellent characteristics, for example, such as warmth retention property is good, elasticity is good, hygroscopicity is strong, soft touch, be not easy it is dirty.These property
The fabric of wool can be made to have the advantage of oneself, various advanced fabric for clothes can be woven with wool: such as cassimere, there is feel
Cunning is glutinous, elasticity is good, quality is abundant, the clothing of the natural season in spring and autumn of gloss；Such as medium weight fancy suiting, it is textured it is abundant, warmth retention property is strong,
The winter clothing of comfortable feel.Such as the woollen sweater etc. that winter often wears.Wool can also weave industrial woollen goods, woollen blanket, liner
Etc. important materials.
But wool is also not perfect, and there is also some disadvantages for it.As wool fabric is using washing of drum type washing machine
When be easy to so that fabric size occur significant change, this is because direction friction effect be easy to occur felting deformation；Secondly,
The presence of wool scale layer makes wool fabric that can have prodding and itching feeling when wearing next to the skin, this can largely effect on wearing for wool fabric
Experience, to limit its purposes；Meanwhile wool product needs to have can be only achieved under the higher temperature compared with other materials
Dyeing effect, which results in a large amount of costs on the energy in dyeing process；In addition, wool product need to be by softness in final finishing
The last phase tidying ups process such as change, antibacterial, scale layer reduces these auxiliary agents to the adsorption capacity of wool fabric, so that final finishing becomes
It is more difficult.
With progress of the epoch, requirement of the people to wool fabric performance also becomes higher and higher.People are not only intended merely to
Wool dress ornament can also require wool dress ornament being capable of fashion, easy cleaning, comfortable and easy to wear, price parent in the warming effect that winter has had
People etc..This just to the dyeability of wool fabric, anti-felting performance, permeability, wear when feeling etc. propose it is very high
Requirement.Simultaneously as the reinforcement of social environment consciousness, people more pay attention to environment.Textile dyeing and finishing industry is due to right
Environment is easy to have an impact, and should more do everything possible and protect environment.For having for the wool of many good characteristics, people
Wish that the processing of the dyeing and finishing to this excellent natural fiber can also go on clean and environmental protection, the new approach of low energy.Therefore, using pair
Environmental-friendly method is modified wool, eliminates and negatively affects caused by scale layer, is an imperative research.
Why wool possesses above-mentioned many characteristics, has special morphosis mainly due to it.Wherein wool fiber
The Microstructure of epidermal scale on surface has very big influence to the physical and chemical properties of wool fabric.These laminar cells are just
It is overlapped covering the scale covered as reptile body surface, is coated on the outside of hair shaft.Root is attached to hair shaft, and taper stretches out
Hair shaft surface and it is directed toward high-quality green tea, by protruding from fiber surface in various degree and being flared out, forms a ladder-like knot in steep face
Structure.
Scale layer is one of distinctive structure of wool fiber, it is easy to happen when resulting in woolen dyed difficulty, washing
The reason of prodding and itching feeling of felting deformation and dress.Scale layer is to be glued by the flat cell of keratinization by cell membrane complexes
It connects.The squamous cell of wool consists of three parts: being respectively epicuticula, scale outer layer and scale internal layer.
It is bad that the Microstructure of epidermal scale of wool surface experiences the dress of wool fabric, also adds to the dyeing and finishing in production
Work process brings many troubles.Therefore, the modified research of stripping squama is carried out to wool just to come into being.About grinding for wool stripping squama
Studying carefully has had quite long history, is initially the shrinkproof performance in order to improve wool fabric, improves the dimensional stability of product,
Woolen is set to obtain the performance of " machine washable ".With people to the structure of scale layer and it wool processing take during not
The understanding of same-action is more deep, to wool stripping squama technology research there has also been be more widely applied.Wool strips scale layer
Benefit, mainly have improve wool dyeability, improve wool antistatic property, improve wool snugness of fit, change
The chemisorption performance of kind wool improves five aspects of preventing felting and shrinking of wood fabric performance.
China is needed every year to a large amount of fine wool of the ground import such as Australia, and the country mainly produces low-grade wool fiber,
And supply falls short of demand in China large market for medium and high-quality wool fiber, the state in extremely shortage.If there is method
Existing common and braid wool or half common and braid wool can be modified, keep it similar to the performance of good fine wool, it in this way can be with
The lower common and braid wool raw material of relative mass are used, carry out textile manufacturing instead of fine wool, so that it may create bigger
Social value creates significant economic benefit and social benefit.It is stripped so wool fiber of poor quality is carried out surface scale,
High-quality wool fiber needed for being modified in the market has great potential commercial value, and China will also go into overdrive to carry out this side
The research work in face.
However, it is not uniform that existing modified technique, which optimizes modified effect to same a collection of wool fiber,
, it is difficult that all wool fibers is accomplished all perfectly to strip scale.Such as in the wool fiber of same batch processing, have
Scale strips the fabulous wool fiber of effect, also has almost without the wool fiber of scale is stripped, might have therebetween larger
Gap.Therefore, it is necessary to assess modified effect, the preferable wool fiber of wherein treatment effect is screened out, it is found out
Account for the percentage of total amount.If stripping scale in the wool of this batch processed, good wool fiber is in the great majority, it is believed that changes
The effect of property is good；, whereas if it is most of all almost without scale is peelled off in this batch wool fiber, then it is considered that modified
Effect it is bad.
The existing evaluation process to modified effect mainly or by manpower carries out, not only time-consuming and laborious, and assess
As a result it is affected by evaluator's subjective factor, causes the credible result degree of assessment not high.
Computer digital image processing is exactly to be received using computer to image, then carries out the extraction of information, adds
The process of work, pattern-recognition and storage and display.Computer has obtained after the information of image, so that it may carry out image to image
Enhancing, compression, recovery, segmentation, the various processing of identification etc..
Computer digital image processing is a comparatively emerging subject, with the rapid development of information technology, letter
There is the extensive use of Digital Image Processing in the fields such as breath processing, measuring technique and engineering technology.With computer section
The fast development of technology, digital image processing techniques achieve the progress for the formula of leaping.At the same time, at computer digital image
Reason with contacting between Other subjects and intersect also increasingly extensive.As a kind of effective help mode, computer graphic image
Processing technique has gradually been penetrated into the research work of Other subjects, such as arrives health care, aerospace, industry, military affairs,
Its application field is more and more extensive, and increasingly plays its powerful effect.
Computer digital image processing is to develop a very quick branch in the information processing technology.Since computer is hard
The hardware condition of rapid development in terms of part, image processing techniques development is strengthened.Within nearly 30 years time, calculate
Machine digital image processing techniques provide huge help for the research and development in each field, intersecting between Other subjects
Increasingly deepen.
With the fast development of production level, people are to the requirement in the requirement and level of comfort of the aesthetic aspect of dress ornament
It is higher and higher；And as production scale is increasing, the level of socializations of production process improves, requirement of the dress ornament to raw material
It is higher and higher.Therefore, the quality and grade of purely evaluating and identify product by people can not meet the requirement of production again.
This is because human eye often has very big subjective factor to the judgement of product, and the evaluation and identification of mistake will lead to huge damage
It becomes estranged risk.It would therefore be desirable to a kind of new appraisal procedure, so that accuracy, science and reliability that assessment has more.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of digital picture stage divisions of wool scale layer, can be fast
Speed equitably evaluates wool modifying effect.
The technical solution adopted by the present invention to solve the technical problems is: providing a kind of digital picture point of wool scale layer
Grade method, comprising the following steps:
(1) denoising is carried out to the digital picture got；
(2) digital picture after denoising is subjected to binaryzation；
(3) straight-line detection at wool edge is carried out to the digital picture after binaryzation, and chooses suitable region；
(4) wool fiber in the appropriate area of selection is projected in horizontal direction and obtains position locating for reaction wool scale
The waveform set；
(5) first derivative and second dervative for the wavy curve being calculated, according to the value of first derivative and second dervative
Determine peak point, wherein it is 0 that peak point, which should meet first derivative, and second dervative is negative；
(6) spacing for calculating adjacent peak point, is classified according to the spacing of adjacent peak point.
The step (1) carries out denoising using Gaussian smoothing filter method.
It matches by human eye and edge line detection when binaryzation in the step (2) and carrys out threshold value.
Straight line fitting is carried out by Hough transformation to the digital picture after binaryzation in the step (3), it is fine to extract wool
The straight-line detection to wool edge is completed on the both sides of dimension.
The step (4) specifically includes: by the wool fiber Image Reversal in the appropriate area of selection to horizontal direction, time
Each pixel on image is gone through, integral projection is carried out to image and obtains drop shadow curve.
Further include the steps that screening peak point in the step (5), specifically: remove some height and entire song
The wave crest of the average value of line relatively；Remove one in adjacent two close wave crests.
Beneficial effect
Due to the adoption of the above technical solution, compared with prior art, the present invention having the following advantages that and actively imitating
Fruit: the present invention establishes a set of analysis of clearly sentencing for stripping effect about wool scale and is classified, and cooperation digital image analysis handles skill
The algorithm that wool scale density is identified and is classified is researched and developed in the use of art, is realized to wool fiber electron micrograph
Automatic classification function, make its in actual operation can the treatment effect to treated wool fiber carry out classification identification.?
On the basis of being classified, accurate, scientific, reliable assessment thus is made to the effect of modification, and then judge at this batch
Which kind of purposes the wool managed can make in practical.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is that Hough transformation detects straight line schematic diagram in embodiment；
Fig. 3 is the schematic diagram that detection zone is selected in embodiment；
Fig. 4 is the wool fiber digitized map that horizontal direction is turned in embodiment；
Fig. 5 is that wool fiber drop shadow curve schemes in embodiment.
Specific embodiment
Present invention will be further explained below with reference to specific examples.It should be understood that these embodiments are merely to illustrate the present invention
Rather than it limits the scope of the invention.In addition, it should also be understood that, after reading the content taught by the present invention, those skilled in the art
Member can make various changes or modifications the present invention, and such equivalent forms equally fall within the application the appended claims and limited
Range.
Embodiments of the present invention are related to a kind of digital picture stage division of wool scale layer, specifically include following step
It is rapid:
(1) denoising is carried out to the digital picture got.
A most important condition in digital picture application is exactly to search out image object under noise background.And eliminate image
The method of noise is image smoothing (smoothing).Main purpose that there are two image smoothings, one is improving image quality,
Second is that the feature of extraction object.It is therefore desirable to which effective filtering mechanism completes the promotion and object to picture quality
The extraction of feature.
During shooting electromicroscopic photograph, impurity always can be inevitably discharged into.Or in fiber micro-image
Acquisition procedure in, fiber edge will appear pseudo-side.For these phenomenons that fiber picture occurs, " noise should be used first
Removal " algorithm, carries out the smoothing processing of image, is found through experiments that the smooth effect of which kind of algorithm is preferable.Common filtering is calculated
Method has simple smooth filtering, Gaussian smoothing filter, median filtering etc..
It is the simple smooth processing of image first.In simple smooth processing, algorithm is using convolution mask in image
Each pixel is handled one by one, then generates the filter effect of image.This process can be likened to as one by one mistake
Each of filter processing source images pixel, and relevant data operation is carried out, finally obtain data result, i.e., it is required
Smoothed image.
The Gaussian smoothing of image is to be smoothed one of method of image, is also based on the average basic think of in field
Road.And the different place of simple smooth of itself and image is, in the Gaussian smoothing of image, is carrying out the average image field
When different weights is given to the pixel of different location, then carry out data operation again, then obtain required smoothed image.With
The simple smooth of image compares, and the treatment effeciency of Gaussian smoothing is not high when being smoothed to high-contrast image, smoothly
Effect is also not very ideal.But in the case where possessing the overall gray level characteristic distributions of source images in needing smooth operation,
Gaussian smoothing has very big advantage, available preferable smooth effect.
Median filtering is a kind of nonlinear filtering, relative to other two kinds of filters, its advantage is that: firstly, intermediate value
The effect that filtering reduces noise is obvious；Secondly, median filtering is available very in the case where gray-value variation is smaller
Good smoothing processing result；Directly filtering can preferably protect complete edge details in finally, rather than as mean filter floods
No details.
Present embodiment uses Gaussian smoothing filter.Show through many experiments in Gaussian smoothing filter, simple average
In filtering and median filtering, the noise reduction effect of Gaussian smoothing filter is best.
(2) digital picture after denoising is subjected to binaryzation.
The binaryzation of image is exactly to set 0 or 255 for the gray value of pixel each on image, i.e., non-black i.e. white, figure
As upper there are two types of color.In Digital Image Processing, the binary conversion treatment of image is a highly important step, this is because figure
The binaryzation of picture is greatly reduced data volume in image, convenient to carry out to clarification of objective so as to highlight the profile of target
It is further to extract.
During image binaryzation, a threshold value is needed, to determine the binaryzation degree of image.Threshold value is to determine
A range out, then the pixel on the image in this numerical value is arranged to 0, and the pixel except this numerical value is set
It is set to 255.If some pixel value on image is greater than threshold value, then the gray value of the pixel is 255 after converting, as black
Color；It is on the contrary then be 0, i.e., it is white.
For wool fiber image, required target is the density of wool surfaces scale, so what is desired is that squama
The edge of piece, so being black white image by the image binaryzation obtained after edge detection, to protrude the surface scale of fibre image
Feature.For image, the maximum difference of sundries is exactly gray level in object and background.The binaryzation of image is exactly logical
It crosses and the gray scale of image is divided into different brackets, then background and object are distinguished by the way that the method for threshold value is arranged, thus
Obtain the surface scale feature of the i.e. fiber of object needed for us.
In binarization, the size of threshold value can directly decide the effect of segmentation, when fixed excessively high of threshold value, accidentally go out
Existing object point will be taken as background, so that entire picture is disorderly and unsystematic, be hard to tell object and background sundries；If threshold value
It is too low, it will there is a situation where opposite.Object is dissolved into background, is not separated.Therefore, for the two-value of image
For change, it is mostly important for selecting a suitable threshold value, so that the error for dividing rear backdrop and target is minimum.However, by
Inevitably there is sundries influence in wool electromicroscopic photograph, and have the variation of the conditions such as light, gray scale in different images, it is difficult to make
Computer directly determines threshold value.Therefore, the straight line searched out by human eye cooperation Hough transformation is needed in present embodiment, to determine
A fixed optimal threshold value, to guarantee the accuracy of wool fiber classification.
(3) straight-line detection at wool edge is carried out to the digital picture after binaryzation, and chooses suitable region.
The edge both sides of wool fiber are close to straight line, in the research process for carrying out wool identification, can choose using logical
The method that Hough transformation (Hough) carries out straight line fitting is crossed, to extract the both sides of wool fiber, so that it is determined that a suitable threshold
Value parameter, and wool fiber is carried out to rotate to horizontal operation convenient for the later period.
Hough transformation is to extract objective contour using a kind of transform domain, by the way that the coordinate transform of the point on straight line was arrived
The coefficient domain of the straight line of point, and the collinearly relationship with straight line intersection is cleverly utilized, it is that the extraction problem of straight line is converted into meter
Number problem.Therefore, the extraction problem of image border can be by connecting group without several Hough straight lines by edge pixel point
At closed area boundary.
Assuming that interior straight line, equation in O-xy plane are as follows: y=ux+v, wherein u and v be respectively straight line slope and
Intercept, for given straight line, corresponding one several to (u, v), whereas if given one several to (u, v), then correspond to one
Several on straight line y=ux+v, i.e. straight line y=ux+v in O-xy plane and O-uv plane correspond (u, v), this relationship
Referred to as Hough transformation.Therefore, each point in O-xy plane corresponds to straight line, these straight lines all intersect at a bit (u,
V), but the slope of this straight line may be infinitely great, therefore this linear equation is changed to polar equation: xcos α+ysin α=ρ,
In formula, ρ is distance of the straight line to coordinate origin, and α is the angle of straight line normal and X-axis, then, one in O-xy plane coordinates
Point in straight line and coordinate O- ρ α plane corresponds.
Fig. 2 is Hough transformation detection straight line schematic diagram.According to the difference of the threshold value manually selected, Hough transformation is found straight
Line can also change.As a result, when more clearly detecting the edge line of wool fiber, the scale feature of wool also obtain compared with
Good embodiment.It is possible thereby to determine a more accurately threshold value, after a threshold value has been determined, need manually to wool fibre
Dimension word picture carries out a regional choice, the region that edge should be selected clearly to detected at this time (see Fig. 3).
(4) wool fiber in the appropriate area of selection is projected in horizontal direction and obtains position locating for reaction wool scale
The waveform set.
After the binary image for obtaining can reacting wool scale feature, as long as wool fiber is projected to horizontal direction
On, so that it may reacted the waveform of wool scale present position.And the position of its medium wave peak, often exactly where scale
Position.And if directly projected, the problem of due to angle, it is difficult really to reflect true position locating for scale edge
It sets.It is turned to horizontal direction (see Fig. 4) for this purpose, first having to a wool fiber digital picture, traverses each pixel on image
Point carries out integral projection to image and obtains drop shadow curve (see Fig. 5).
(5) first derivative and second dervative for the wavy curve being calculated, according to the value of first derivative and second dervative
Determine peak point, wherein it is 0 that peak point, which should meet first derivative, and second dervative is negative.
The core ideas of wave crest detection algorithm is that should to meet first derivative be 0 to the peak point of curve, and meets second order
Derivative is negative.The algorithm that present embodiment uses calculates the first derivative diff_v [i] of curve first, is then converted
For -1 and 1, this is because the specific value of first derivative is not significant, what is needed is only its symbol, so that it may find out former song
The monotonicity of line.Then in the point for being 0 for a level number, a part therein is not real wave crest, and only one
Order derivative is 0.So their first derivative is set as identical as the gradient direction where them.
Then when calculating second dervative, if the second dervative of the point is -2, the direction of illustrative graph is at this
Point occurs by just becoming negative, that is, the wave crest point in virgin curve.And position corresponding to wave crest point, it just contains corresponding in original image
Wool scale corresponding to position.
However due to the scrambling of wool scale, the wave crest that is found through the above steps is simultaneously not all wool fiber squama
Position where the edge of piece, it is therefore desirable to which all wave crests found are screened.First have to reject some height with it is entire
The wave crest of the average value of curve relatively.This is because the edge projection that these wave crests are only wool fiber obtains, and
It is not the reaction to the scale edge of wool surface.Second, to reject one in adjacent two close wave crests, this is
Since there are a scale edges due to its pattern is irregular etc., may projection there are two adjacent very close waves
Peak.Thus, it is possible to obtain more accurately, can react scale edge wave crest quantity.
(6) spacing for calculating adjacent peak point, is classified according to the spacing of adjacent peak point.
According to the quantity of the available wave crest of the above method, the spacing of wave crest, the i.e. spacing of scale so can be obtained by
d.After obtaining the spacing between scale, on the basis of a large amount of scale spacing datas detected early period, it is concluded
And analysis, degree thus can be stripped to the scale of wool fiber to be gone out to be classified according to the value work done in the manner of a certain author of d.The mode of classification is specific as follows:
It is the first order if d≤40；It is the second level if 40 d≤50 <；It is the third level if 50 d≤60 <；If 60 < d≤
70, it is the fourth stage；It is level V if d > 70.Wherein level V is the superlative degree, that is, represents wool fiber scale and strip effect
Best grade；The first order is lowermost level, i.e., it is poor to strip effect for wool fiber scale.
Wherein it is especially noted that if wool surface almost strips scale, or only it is left
It can be used as a kind of special circumstances when minimal amount of scale vestiges, should directly define the level as level V in this case.This be by
Drop shadow curve at this time does not have apparent wave crest, this, which may cause, finds wave crest result inaccuracy, and then leads to the knot of classification
Fruit inaccuracy.Therefore, hereinto in the case of It is not necessary to be classified using this method, level V of directly defining the level.
It is classified it is not difficult to find that the present invention establishes a set of analysis of clearly sentencing for stripping effect about wool scale, cooperation number
The algorithm that wool scale density is identified and is classified is researched and developed in the use of image analyzing and processing technology, is realized to wool fiber
Electron micrograph automatic classification function, make its in actual operation can treatment effect to treated wool fiber into
Row classification identification.On the basis of obtaining classification, accurate, scientific, reliable assessment thus is made to the effect of modification, into
And judge which kind of purposes the wool of the batch processing can make in practical.
Claims (3)
1. a kind of digital picture stage division of wool scale layer, which comprises the following steps:
(1) denoising is carried out to the digital picture got；
(2) digital picture after denoising is subjected to binaryzation, wherein match when binaryzation by human eye and edge line detection
Carry out threshold value；
(3) straight-line detection at wool edge is carried out to the digital picture after binaryzation, and chooses suitable region；Specifically: it is right
Digital picture after binaryzation carries out straight line fitting by Hough transformation, extracts the both sides of wool fiber, completes to wool edge
Straight-line detection；
(4) wool fiber in the appropriate area of selection is projected in horizontal direction and obtains reaction wool scale present position
Waveform；It specifically includes: the wool fiber Image Reversal in the appropriate area of selection to horizontal direction traverses each on image
A pixel carries out integral projection to image and obtains drop shadow curve；
(5) first derivative and second dervative for the wavy curve being calculated are determined according to the value of first derivative and second dervative
Peak point, wherein it is 0 that peak point, which should meet first derivative, and second dervative is negative；
(6) spacing for calculating adjacent peak point, is classified according to the spacing of adjacent peak point.
2. the digital picture stage division of wool scale layer according to claim 1, which is characterized in that the step (1)
Denoising is carried out using Gaussian smoothing filter method.
3. the digital picture stage division of wool scale layer according to claim 1, which is characterized in that the step (5)
In further include the steps that screening peak point, specifically: remove the average value of some height and entire curve relatively
Wave crest；Remove one in adjacent two close wave crests.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610966643.0A CN106503681B (en) | 2016-10-28 | 2016-10-28 | A kind of digital picture stage division of wool scale layer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610966643.0A CN106503681B (en) | 2016-10-28 | 2016-10-28 | A kind of digital picture stage division of wool scale layer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106503681A CN106503681A (en) | 2017-03-15 |
| CN106503681B true CN106503681B (en) | 2019-05-03 |
Family
ID=58321799
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610966643.0A Expired - Fee Related CN106503681B (en) | 2016-10-28 | 2016-10-28 | A kind of digital picture stage division of wool scale layer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106503681B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112469994A (en) * | 2018-08-17 | 2021-03-09 | 日本制纸株式会社 | Method for analyzing structure containing fibers and inorganic particles |
| CN113640445B (en) * | 2021-08-11 | 2024-06-11 | 贵州中烟工业有限责任公司 | Characteristic peak identification method based on image processing, computing device and storage medium |
| CN115684737B (en) * | 2022-10-26 | 2023-11-17 | 常州同惠电子股份有限公司 | Algorithm for calculating waveform glitch |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8023742B2 (en) * | 2007-10-09 | 2011-09-20 | Microsoft Corporation | Local image descriptors using linear discriminant embedding |
| CN103149210A (en) * | 2013-02-25 | 2013-06-12 | 东华大学 | System and method for detecting fabric cashmere content based on scale graphic features |
| CN105678341A (en) * | 2016-02-19 | 2016-06-15 | 天纺标检测科技有限公司 | Wool cashmere recognition algorithm based on Gabor wavelet analysis |
-
2016
- 2016-10-28 CN CN201610966643.0A patent/CN106503681B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8023742B2 (en) * | 2007-10-09 | 2011-09-20 | Microsoft Corporation | Local image descriptors using linear discriminant embedding |
| CN103149210A (en) * | 2013-02-25 | 2013-06-12 | 东华大学 | System and method for detecting fabric cashmere content based on scale graphic features |
| CN105678341A (en) * | 2016-02-19 | 2016-06-15 | 天纺标检测科技有限公司 | Wool cashmere recognition algorithm based on Gabor wavelet analysis |
Non-Patent Citations (3)
| Title |
|---|
| Fuzzy Classification of Animal Fibers Using Neuro-Fuzzy Classifier;Xian-Jun Shi et al.;《2008 First International Conference on Intelligent Networks and Intelligent Systems》;20081103;第119-122页 |
| 基于数字图像处理的羊绒和羊毛识别模型的研究;高丽;《中国优秀硕士学位论文全文数据库 信息科技辑》;20160215;第I138-1912页 |
| 基于贝叶斯方法的山羊绒与细羊毛的鉴别;石先军 等;《纺织学报》;20080131;第29卷(第1期);第26-29页，第33页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106503681A (en) | 2017-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106503681B (en) | A kind of digital picture stage division of wool scale layer | |
| CN101840511B (en) | Method for extracting, matching and recognizing characteristics of finger veins | |
| CN101424680B (en) | Computer automatic recognition apparatus and method for profile fiber | |
| CN103279745B (en) | A kind of face identification method based on half face multiple features fusion | |
| CN113554080A (en) | Non-woven fabric defect detection and classification method and system based on machine vision | |
| CN105740829A (en) | Scanning line processing based automatic reading method for pointer instrument | |
| CN103454276B (en) | A kind of fabric shape style evaluation method based on dynamic sequence image | |
| Zhang et al. | A review of fabric identification based on image analysis technology | |
| CN102096827A (en) | Profiled fiber automatic recognition method based on scale invariability and support vector machine classification | |
| Li et al. | TireNet: A high recall rate method for practical application of tire defect type classification | |
| Sun et al. | Body shape classification and block optimization based on space vector length | |
| Citraro et al. | Extended three-dimensional rotation invariant local binary patterns | |
| CN109977909A (en) | Finger vein identification method and system based on minutiae point Region Matching | |
| Lee et al. | Classification of body shape characteristics of women's torsos using angles | |
| CN104657983B (en) | A kind of fabric ball top Density Detection method filtered based on Gabor | |
| CN1971274A (en) | Method for assessing smoothness level of dress material based on point model | |
| CN106770372B (en) | White glass detection method of surface flaw based on micro-vision | |
| CN101430789A (en) | Image edge detection method based on Fast Slant Stack transformation | |
| CN105844618A (en) | Image processing and characteristic extraction method of evaluating clothes wearing wrinkling degree | |
| Wu et al. | Automatic texture exemplar extraction based on global and local textureness measures | |
| CN113627531B (en) | Method for determining pear ring rot resistance based on support vector machine classification algorithm | |
| Guo et al. | Multi-criteria decision making under uncertainty for Personality-driven Modern design | |
| Yeh et al. | A reference standard of defect compensation for leather transactions | |
| Lee et al. | A study on representation of 3D virtual fabric simulation with drape image analysis II-focus on the comparison between real clothing and 3D virtual clothing | |
| Tsai et al. | User-guided line abstraction using coherence and structure analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190503 Termination date: 20211028 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |