CN101996322A - Method for extracting fractal detail feature for representing fabric texture - Google Patents
Method for extracting fractal detail feature for representing fabric texture Download PDFInfo
- Publication number
- CN101996322A CN101996322A CN 201010536853 CN201010536853A CN101996322A CN 101996322 A CN101996322 A CN 101996322A CN 201010536853 CN201010536853 CN 201010536853 CN 201010536853 A CN201010536853 A CN 201010536853A CN 101996322 A CN101996322 A CN 101996322A
- Authority
- CN
- China
- Prior art keywords
- fractal
- dimension
- fractal dimension
- basic cycle
- subwindow
- 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 belongs to the field of digital image processing and pattern recognition, particularly relates to a method for extracting fractal detail feature for representing a fabric texture, comprising the following steps of: solving basic transverse and longitudinal cycle periods of a fabric texture image by adopting fourier transform; calculating a fractal dimenstion of each sub-window containing the transverse basic cycle period or the longitudinal basic cycle period in the original image according to a traversing method, wherein the fractal dimenstion of each sub-window is calculated on the basis of a corresponding one-dimensional time sequence formed by transversely or longitudinally accumulating image pixel gray values; and finally, selecting two fractal dimenstion extreme values of reflecting transverse detail information and two fractal dimenstion extreme values of reflecting longitudinal detail information as detail features used for representing the fabric texture. Every two of the four extreme value fractal features have obvious complementarity, and a feature vector formed from the four extreme value fractal features can realize rapid, overall and deep representation on fabric texture details.
Description
Technical field
The invention belongs to Digital Image Processing and area of pattern recognition, particularly a kind ofly be used to characterize cloth textured fractal minutia extracting method.
Background technology
Can realize cloth textured parameter estimation, Texture classification, appearance of fabrics evaluation, Defect Detection or the like purpose by cloth textured characterization technique.Want careful and profoundly characterize the cloth textured detailed information that just must deeply excavate the texture each side.For the fast characterizing texture, must implement necessary dimensionality reduction pre-service simultaneously to texture image to be analyzed.The present invention is intended to discuss the cloth textured details characterizing method based on fractal characteristic, this method has been implemented certain dimension-reduction treatment according to the characteristics of fabric longitude and latitude orientation to the original texture image, calculates four extreme value fractal dimensions as minutia according to cloth textured basic cycle period and traversal method principle then.
Than euclidean geometry, fractal geometry can provide better method when describing or generating natural things with self-similarity or class natural things, thereby extensively are used in the simulation of pattern-recognition, image and emulation or the like numerous areas.Self-similarity is one of central concept of fractal theory, and the notion of it and dimension is closely related.The object that fractal geometry are described has the self similarity on the statistical significance, and it is one of characteristic parameter of the most normal use when carrying out graphical analysis with fractal theory that self-similarity characterizes FRACTAL DIMENSION with FRACTAL DIMENSION.Fractal characteristic particularly fractal dimension can be portrayed coarse texture degree and complexity preferably, thus in practices such as Texture classification, identification as tolerance feature be rational.Wherein box counting dimension simple owing to notion, calculate the easy the most general a kind of fractal dimension of use that becomes.
For ease of explanation invention main points, be necessary the ultimate principle and the evaluation method of box counting dimension are done simple the introduction.
If
Be non-NULL bounded aggregate arbitrarily, with N (δ, F) expression cover collection F required diameter be to the maximum δ the minimal number of collection, then the box dimension definition of F is
Note, used δ-covering is still a general collection class in the definition, in this patent collection F refer in particular to into textile image when the vertical, horizontal projection, by the gradation of image one dimension time series that each is capable, each row pixel grey scale adds up and get the average gained, it also is the curve of each each row pixel grey scale Change in Mean of row of a presentation video, (δ, F) the required length of side of expression covering F is the minimum number of squares of δ to N, notes by abridging into N (δ).D
B(F) brief note is D.
During seasonal effect in time series meter box dimension of actual estimation, because this sequence is a curve, horizontal ordinate is the position of each point in the sequence, and ordinate is the sequential value of each point correspondence, need remove to cover fully this curve and add up N (δ) with the grid that is of a size of δ * δ.From the definition of box counting dimension as can be known, logN (δ) ∞ Dlog (1/ δ), this shows, and some is straight line to (log (1/ δ), logN (δ)) asymptotic line in δ → 0 o'clock, and its slope is D.It is right that thereby change δ size can obtain a plurality of above-mentioned points, goes out respective straight by least square fitting then.The slope of this straight line is the box counting dimension of being asked.
Consider that woven fabric is by vertically being interwoven mutually through weft yarn, its image is a kind of typical texture image, and therefore available fractal characteristic characterizes cloth textured.People such as Conci (1998) adopt difference meter box method to extract cloth textured FRACTAL DIMENSION and standard deviation is used for characterizing cloth textured as characteristic parameter and detect fabric defects.People such as Xu Zengbo (2000) carry out at cloth textured image on the basis of Wold model analysis, with ask in the FRACTAL DIMENSION process whole fractal characteristic curve as characterizing cloth textured feature, carried out the fabric defects detection.People such as Wen (2002) adopt this fractal parameter of Hurst coefficient of estimating textile image based on the Fourier domain maximal possibility estimation operator of fractal Brown motion, detect fault with this as characterizing cloth textured characteristic parameter.People such as Yang Yan (2007) have extracted an overall FRACTAL DIMENSION feature and have realized objective evaluation to the fabric crape effect from the crepe fabric image.Go on foot the red firm people of grade (2007) in order to overcome the limitation of single fractal characteristic, a kind of many fractal characteristic vector extracting method have been proposed, the more relevant research on the defect detection effect of this method has had significantly improvement, but, thereby do not conform in detecting a lot of local faults because a plurality of fractal characteristic vectors that extracted all are general picture features of reflection global information.In the patent of " based on square and fractal texture classifying method " (2006), the researcher is the second moment of computed image at first, produce the moment characteristics image, again original image piece and moment characteristics image are estimated its fractal dimension, the fractal dimension with original image piece and six moment characteristics images forms proper vector, carries out cloth textured classification as the input of support vector machine.
Above-mentioned periodical literature or patent are in cloth textured fractal characterization, have following shortcoming: 1, fractal characteristic directly extracts on the two dimensional image basis, the global information that calculated amount is big by 2, fractal characteristic that extracted only can be portrayed texture is underused cloth textured intrinsic longitude and latitude orientation characteristics to improve the stability of feature in the time of can not carefully profoundly characterizing cloth textured detailed information 3, feature extraction.
Summary of the invention
The invention belongs to Digital Image Processing and area of pattern recognition, particularly a kind ofly be used to characterize cloth textured fractal minutia extracting method.At first adopt the one dimension Fast Fourier Transform (FFT) to obtain the horizontal and vertical substantially cycle period of cloth textured image, then according to each comprises the fractal dimension of the subwindow of a laterally basic cycle period or vertically basic cycle period in the traversal method principle computed image, wherein the fractal dimension of each subwindow is to calculate on laterally or vertically adding up the corresponding one dimension time series basis that forms on image pixel gray-scale value edge, therefrom chooses the fractal dimension extreme value (being maximum fractal dimension and minimum fractal dimension) of two horizontal detailed information of reflection and the cloth textured minutia of fractal dimension extreme values (being maximum fractal dimension and minimum fractal dimension) conduct sign of two vertical detailed information of reflection at last.Above-mentioned four fractal extremal features have tangible complementarity each other, and the proper vector of being made up of them can realize cloth textured details fast, comprehensively and is profoundly characterized.Since fractal dimension can characterization data complexity, four features of Ti Quing can characterize laterally greatly complexity, horizontal minimum complexity, vertically very big complexity and vertical minimum complexity of texture respectively thus.In addition, owing to above-mentioned Feature Extraction is all carried out on one dimension time series basis, thereby computation amount, will concentrate on owing to cloth textured information spinner simultaneously, thereby most useful information is kept through broadwise.
A kind of cloth textured fractal minutia extracting method that is used to characterize of the present invention, at first adopt the one dimension Fast Fourier Transform (FFT) to obtain the laterally basic cycle period and the vertically basic cycle period of cloth textured image, calculate in the cloth textured image fractal dimension of each subwindow that comprises a laterally basic cycle period according to traversal method principle then and each comprises the fractal dimension of the subwindow of a vertically basic cycle period;
Wherein said comprise one laterally the subwindow of basic cycle period be to serve as that long and cloth textured image wide is that wide rectangular window, fractal dimension of the subwindow of described each laterally basic cycle period are laterally to add up and calculate on the corresponding one dimension time series basis that forms in the image pixel gray-scale value edge in this subwindow with a laterally basic cycle period; Therefrom choosing two fractal dimension extreme values is horizontal maximum fractal dimension and horizontal minimum fractal dimension;
Described comprise one vertically the subwindow of basic cycle period be that length with a cloth textured image serves as that long and vertically basic cycle period is wide rectangular window, fractal dimension of the subwindow of described each vertically basic cycle period is that the image pixel gray-scale value in this subwindow longitudinally adds up and calculates on the corresponding one dimension time series basis that forms; Therefrom choosing two fractal dimension extreme values is vertical maximum fractal dimension and vertical minimum fractal dimension;
At last horizontal maximum fractal dimension, horizontal minimum fractal dimension, vertical maximum fractal dimension and vertical minimum fractal dimension as characterizing cloth textured minutia;
The leaching process of described horizontal maximum fractal dimension, horizontal minimum fractal dimension, vertical maximum fractal dimension and vertical minimum fractal dimension is as follows:
At first gather the cloth textured image of digitizing, be designated as W, W is a rectangle, its size is long * wide be L
1* L
2, promptly horizontal and vertical length is respectively L
1And L
2, promptly the row cycle is P and it is along the horizontal basic cycle
1Individual pixel, the basic cycle longitudinally is that line period is P
2, the pixel count after line period and row cycle all refer to round, P
1The basic cycle period of gathering by the arbitrary capable image pixel gray-scale value that calculates W obtains P
2The basic cycle period of gathering by the arbitrary row image pixel gray-scale value that calculates W obtains, and wherein the calculating of above-mentioned basic cycle realizes by the one dimension Fast Fourier Transform (FFT); For N point sequence x (n), its FFT transform definition is
Wherein,
Be called twiddle factor;
X (k) sequence that sequence of real numbers x (n) obtains after FFT handles is a sequence of complex numbers, first value respective frequencies of this sequence of complex numbers is 0, does not have practical significance, directly with its removal, remaining sequence is a structural symmetry sequence, and the data that only need get its preceding N/2 when carrying out analysis of spectrum get final product.The mould of X (k) is called amplitude, amplitude square be called power, be designated as W.The pairing frequency of peak power is the dominant frequency of sequence x (n), and the inverse of dominant frequency is the basic cycle of this sequence.If the sample frequency of sequence x (n) is f
s(Hz), then the k point is the pairing actual frequency of X (k)
Generally speaking, regulation f
s=1, therefore,
Thereby the actual cycle that the k point is corresponding
Extract the arbitrary capable pixel grey scale value set of cloth textured image to be analyzed, Fast Fourier Transform (FFT) is tried to achieve it and promptly is listed as basic cycle P along the horizontal basic cycle by one dimension
1Extract arbitrary row pixel grey scale value set of cloth textured image to be analyzed, its basic cycle longitudinally basic cycle P is at once tried to achieve in Fast Fourier Transform (FFT) by one dimension
2
In cloth textured image W, choose a laterally basic cycle period P
1Wide L for long and cloth textured image
2For wide rectangular window conduct comprises a horizontal subwindow of cycle period substantially, be designated as W
1Choose the long L of a cloth textured image
1Be length, vertically basic cycle period P
2For wide rectangular window conduct comprises a vertical subwindow of cycle period substantially, be designated as W
2
For a certain W
1, calculating the image pixel gray-scale value projection that it follows direction, the image pixel gray-scale value that is about to each row of this subwindow obtains an one dimension time series along laterally stack, can calculate a fractal dimension from this time series, then with W
1To travel through whole W, have L with fixed step size slippage flatly
1-P
1+ 1 W
1Thereby, can correspondingly try to achieve L
1-P
1+ 1 fractal dimension, note reckling and the maximum wherein is E respectively
1And E
2, being horizontal minimum fractal dimension and horizontal maximum fractal dimension, both reflect the lateral extreme detailed information of texture for this;
For a certain W
2, calculating its image pixel gray-scale value projection along column direction, the image pixel gray-scale value that is about to each row of this subwindow longitudinally superposes, and obtains an one dimension time series, can calculate a fractal dimension from this time series, then with W
2To travel through whole W, have L with fixed step size slippage vertically
2-P
2+ 1 W
2Thereby, can correspondingly try to achieve L
1-P
2+ 1 fractal dimension, note reckling and the maximum wherein is E respectively
3And E
4, being vertical minimum fractal dimension and vertical maximum fractal dimension, both reflect the vertically extreme detailed information of texture for this;
Finally obtain characterizing cloth textured proper vector [E
1E
2E
3E
4].
Wherein, aforesaid a kind of being used to characterizes cloth textured fractal minutia extracting method, and described fabric is a woven fabric.
Aforesaid a kind of being used to characterizes cloth textured fractal minutia extracting method, and described fractal dimension is meant box counting dimension, and its concrete computing method are referring to the relevant introduction in the background technology.
Aforesaid a kind of cloth textured fractal minutia extracting method that is used to characterize, used δ size sequence was 2～6 pixels when described box counting dimension was estimated.
Aforesaid a kind of being used to characterizes cloth textured fractal minutia extracting method, and described fixed step size refers to 1～3 pixel.
Aforesaid a kind of be used to characterize cloth textured fractal minutia extracting method, described rectangle subwindow W
1Each horizontal sliding fixed step size and W
2Each vertical slippage fixed step size needn't be identical.
Beneficial effect
Of the present inventionly a kind ofly be used to characterize that cloth textured fractal minutia extracting method extracted is used to characterize cloth textured proper vector and possesses following advantage:
(1) can characterize cloth textured detailed information, especially cloth textured very big and minimum complexity information comprehensively and meticulously;
(2) thus in leaching process, made full use of the cloth textured characteristics that its main information of warp-wise and broadwise orientation concentrates on warp-wise and broadwise that have, on the basis of the vertical or horizontal one dimension projection sequence of image, calculate, both keep most Useful Informations, reduced calculated amount again significantly;
(3) made full use of cloth textured distinctive cycline rule characteristics, thereby the feature that calculates is stable more and press close to true.
Description of drawings
Fig. 1 is a grain details feature extraction synoptic diagram of the present invention
Fig. 2 is the textile image of one 64 * 64 pixel size of embodiment 1
Fig. 3 is the textile image of one 64 * 64 pixel size of embodiment 2
Embodiment
Below in conjunction with embodiment, further set forth the present invention.Should be understood that these embodiment only to be used to the present invention is described and be not used in and limit the scope of the invention.Should be understood that in addition those skilled in the art can make various changes or modifications the present invention after the content of having read the present invention's instruction, these equivalent form of values fall within the application's appended claims institute restricted portion equally.
A kind of cloth textured fractal minutia extracting method that is used to characterize of the present invention, at first adopt the one dimension Fast Fourier Transform (FFT) to obtain the laterally basic cycle period and the vertically basic cycle period of cloth textured image, calculate in the cloth textured image fractal dimension of each subwindow that comprises a laterally basic cycle period according to traversal method principle then and each comprises the fractal dimension of the subwindow of a vertically basic cycle period;
Wherein said comprise one laterally the subwindow of basic cycle period be to serve as that long and cloth textured image wide is that wide rectangular window, fractal dimension of the subwindow of described each laterally basic cycle period are laterally to add up and calculate on the corresponding one dimension time series basis that forms in the image pixel gray-scale value edge in this subwindow with a laterally basic cycle period; Therefrom choosing two fractal dimension extreme values is horizontal maximum fractal dimension and horizontal minimum fractal dimension;
Described comprise one vertically the subwindow of basic cycle period be that length with a cloth textured image serves as that long and vertically basic cycle period is wide rectangular window, fractal dimension of the subwindow of described each vertically basic cycle period is that the image pixel gray-scale value in this subwindow longitudinally adds up and calculates on the corresponding one dimension time series basis that forms; Therefrom choosing two fractal dimension extreme values is vertical maximum fractal dimension and vertical minimum fractal dimension;
At last horizontal maximum fractal dimension, horizontal minimum fractal dimension, vertical maximum fractal dimension and vertical minimum fractal dimension as characterizing cloth textured minutia;
The leaching process of described horizontal maximum fractal dimension, horizontal minimum fractal dimension, vertical maximum fractal dimension and vertical minimum fractal dimension is as follows:
At first gather the cloth textured image of digitizing, be designated as W, W is a rectangle, its size is long * wide be L
1* L
2, promptly horizontal and vertical length is respectively L
1And L
2, promptly the row cycle is P and it is along the horizontal basic cycle
1Individual pixel, the basic cycle longitudinally is that line period is P
2, the pixel count after line period and row cycle all refer to round, P
1The basic cycle period of the arbitrary capable pixel grey scale value set by calculating W obtains P
2The basic cycle period of the arbitrary row pixel grey scale value set by calculating W obtains, and wherein the calculating of above-mentioned basic cycle realizes by the one dimension Fast Fourier Transform (FFT); For N point sequence x (n), its FFT transform definition is
Wherein,
Be called twiddle factor;
X (k) sequence that sequence of real numbers x (n) obtains after FFT handles is a sequence of complex numbers, first value respective frequencies of this sequence of complex numbers is 0, does not have practical significance, directly with its removal, remaining sequence is a structural symmetry sequence, and the data that only need get its preceding N/2 when carrying out analysis of spectrum get final product.The mould of X (k) is called amplitude, amplitude square be called power, be designated as W.The pairing frequency of peak power is the dominant frequency of sequence x (n), and the inverse of dominant frequency is the basic cycle of this sequence.If the sample frequency of sequence x (n) is f
s(Hz), then the k point is the pairing actual frequency of X (k)
Generally speaking, regulation f
s=1, therefore,
Thereby the actual cycle that the k point is corresponding
Extract the arbitrary capable pixel grey scale value set of cloth textured image to be analyzed, Fast Fourier Transform (FFT) is tried to achieve it and promptly is listed as basic cycle P along the horizontal basic cycle by one dimension
1Extract arbitrary row pixel grey scale value set of cloth textured image to be analyzed, its basic cycle longitudinally basic cycle P is at once tried to achieve in Fast Fourier Transform (FFT) by one dimension
2
In cloth textured image W, choose a laterally basic cycle period P
1Wide L for long and cloth textured image
2For wide rectangular window conduct comprises a horizontal subwindow of cycle period substantially, be designated as W
1Choose the long L of a cloth textured image
1Be length, vertically basic cycle period P
2For wide rectangular window conduct comprises a vertical subwindow of cycle period substantially, be designated as W
2
For a certain W
1, calculating the image pixel gray-scale value projection that it follows direction, the image pixel gray-scale value that is about to each row of this subwindow obtains an one dimension time series along laterally stack, can calculate a fractal dimension from this time series, then with W
1To travel through whole W, have L with fixed step size slippage flatly
1-P
1+ 1 W
1Thereby, can correspondingly try to achieve L
1-P
1+ 1 fractal dimension, note reckling and the maximum wherein is E respectively
1And E
2, being horizontal minimum fractal dimension and horizontal maximum fractal dimension, both reflect the lateral extreme detailed information of texture for this;
For a certain W
2, calculating its image pixel gray-scale value projection along column direction, the image pixel gray-scale value that is about to each row of this subwindow longitudinally superposes, and obtains an one dimension time series, can calculate a fractal dimension from this time series, then with W
2To travel through whole W, have L with fixed step size slippage vertically
2-P
2+ 1 W
2Thereby, can correspondingly try to achieve L
2-P
2+ 1 fractal dimension, note reckling and the maximum wherein is E respectively
3And E
4, being vertical minimum fractal dimension and vertical maximum fractal dimension, both reflect the vertically extreme detailed information of texture for this;
Finally obtain characterizing cloth textured proper vector [E
1E
2E
3E
4].
Wherein, aforesaid a kind of being used to characterizes cloth textured fractal minutia extracting method, and described fabric is a woven fabric.
Aforesaid a kind of being used to characterizes cloth textured fractal minutia extracting method, and described fractal dimension is meant box counting dimension, and its concrete computing method are referring to the relevant introduction in the background technology.
Aforesaid a kind of cloth textured fractal minutia extracting method that is used to characterize, used δ size sequence was 2～6 pixels when described box counting dimension was estimated.
Aforesaid a kind of being used to characterizes cloth textured fractal minutia extracting method, and described fixed step size refers to 1～3 pixel.
Aforesaid a kind of be used to characterize cloth textured fractal minutia extracting method, described rectangle subwindow W
1Each horizontal sliding fixed step size and W
2Each vertical slippage fixed step size needn't be identical.
Be further described below in conjunction with accompanying drawing:
(1) gather the cloth textured image of digitizing, be designated as W, shown in the rectangle ABCD among Fig. 1, it is of a size of L
1* L
2Pixel, promptly laterally (AD) and vertical (AB) length are respectively L
1Pixel and L
2Pixel, the textile image of collection should make the laterally consistent with the fabric weft direction of image as far as possible, and image is vertical consistent with warp thread direction;
(2) extract the arbitrary capable pixel grey scale value set of cloth textured image to be analyzed, Fast Fourier Transform (FFT) is tried to achieve it and promptly is listed as basic cycle P along the horizontal basic cycle by one dimension
1
(3) extract arbitrary row pixel grey scale value set of cloth textured image to be analyzed, its basic cycle longitudinally basic cycle P is at once tried to achieve in Fast Fourier Transform (FFT) by one dimension
2
(4) in W, appoint and get one as rectangle A
1B
1C
1D
1Shown subwindow is noted by abridging to being designated as W
1, this subwindow lateral length is P
1And longitudinal length is L
2, for this W
1, calculating the image pixel gray-scale value projection that it follows direction, the image pixel gray-scale value that is about to each row of this subwindow obtains an one dimension time series along laterally stack, can calculate a box counting dimension feature from this time series, then with W
1With the fixed step size slippage flatly of each 1～3 pixel to travel through whole W, total L
1-P
1+ 1 W
1Thereby, can correspondingly try to achieve L
1-P
1+ 1 box counting dimension feature, note reckling and the maximum wherein is E respectively
1And E
2
(5) in W, appoint and get one as rectangle A
2B
2C
2D
2Shown subwindow is noted by abridging to being designated as W
2, this subwindow lateral length is L
1And longitudinal length is P
2, for this W
2, calculating its image pixel gray-scale value projection along column direction, the image pixel gray-scale value that is about to each row of this subwindow longitudinally superposes, and obtains an one dimension time series, can calculate a box counting dimension feature from this time series, then with W
2With the fixed step size slippage vertically of each 1～3 pixel to travel through whole W, total L
2-P
2+ 1 W
2Thereby, can correspondingly try to achieve L
2-P
2+ 1 box counting dimension feature, note reckling and the maximum wherein is E respectively
3And E
4
(6) obtain characterizing cloth textured proper vector [E
1E
2E
3E
4].
Embodiment 1
(1) obtain textile image W, this image size is 64 * 64 pixels, as shown in Figure 2.
(2) adopt one dimension FFT that the arbitrary capable gradation data of former figure is carried out computation of Period, obtain being listed as basic cycle P
1=20 pixels.
(3) adopt one dimension FFT that arbitrary row gradation data of former figure is carried out computation of Period, obtain row basic cycle P
2=11 pixels.
(4) in W, get as rectangle A among Fig. 1
1B
1C
1D
1Shown subwindow W
1This subwindow lateral length is 20 pixels and longitudinal length is 64 pixels, each row image pixel gray-scale value of this subwindow is along the transverse projection one dimension time series that can to obtain a length be 64 pixels, then according to traversal method principle, elect as under the situation of 1 pixel at subwindow slippage fixed step size, can try to achieve a series of corresponding box counting dimensions on a series of one dimension time series basis, reckling wherein and the maximum are respectively 1.34 and 1.46.
(5) in W, get as rectangle A among Fig. 1
2B
2C
2D
2Shown subwindow W
2This subwindow lateral length is 64 pixels and longitudinal length is 11 pixels, the one dimension time series that it is 64 pixels that each the row image pixel gray-scale value projection longitudinally of this subwindow can obtain a length, then according to traversal method principle, elect as under the situation of 1 pixel at subwindow slippage fixed step size, can try to achieve a series of corresponding box counting dimensions on a series of one dimension time series basis, reckling wherein and the maximum are respectively 1.30 and 1.47.
(6) finally obtain characterizing cloth textured fractal minutia vector [1.341.461.301.47].
Embodiment 2
(1) obtain textile image W, this image size is 64 * 64 pixels, as shown in Figure 3.
(2) adopt one dimension FFT that the arbitrary capable gradation data of former figure is carried out computation of Period, obtain being listed as basic cycle P
1=8 pixels.
(3) adopt one dimension FFT that arbitrary row gradation data of former figure is carried out computation of Period, obtain row basic cycle P
2=15 pixels.
(4) in W, get as rectangle A among Fig. 1
1B
1C
1D
1Shown subwindow W
1This subwindow lateral length is 8 pixels and longitudinal length is 64 pixels, each row image pixel gray-scale value of this subwindow is along the transverse projection one dimension time series that can to obtain a length be 64 pixels, then according to traversal method principle, elect as under the situation of 2 pixels at subwindow slippage fixed step size, can try to achieve a series of corresponding box counting dimensions on a series of one dimension time series basis, reckling wherein and the maximum are respectively 1.21 and 1.29.
(5) in W, get as rectangle A among Fig. 1
2B
2C
2D
2Shown subwindow W
2This subwindow lateral length is 64 pixels and longitudinal length is 15 pixels, the one dimension time series that it is 64 pixels that each the row image pixel gray-scale value projection longitudinally of this subwindow can obtain a length, then according to traversal method principle, elect as under the situation of 3 pixels at subwindow slippage fixed step size, can try to achieve a series of corresponding box counting dimensions on a series of one dimension time series basis, reckling wherein and the maximum are respectively 1.25 and 1.33.
(6) finally obtain characterizing cloth textured fractal minutia vector [1.211.291.251.33].
Claims (6)
1. one kind is used to characterize cloth textured fractal minutia extracting method, it is characterized in that: at first adopt the one dimension Fast Fourier Transform (FFT) to obtain the laterally basic cycle period and the vertically basic cycle period of cloth textured image, calculate in the cloth textured image fractal dimension of each subwindow that comprises a laterally basic cycle period according to traversal method principle then and each comprises the fractal dimension of the subwindow of a vertically basic cycle period;
Wherein said comprise one laterally the subwindow of basic cycle period be to serve as that long and cloth textured image wide is that wide rectangular window, fractal dimension of the subwindow of described each laterally basic cycle period are laterally to add up and calculate on the corresponding one dimension time series basis that forms in the image pixel gray-scale value edge in this subwindow with a laterally basic cycle period; Therefrom choosing two fractal dimension extreme values is horizontal maximum fractal dimension and horizontal minimum fractal dimension;
Described comprise one vertically the subwindow of basic cycle period be that length with a cloth textured image serves as that long and vertically basic cycle period is wide rectangular window, fractal dimension of the subwindow of described each vertically basic cycle period is that the image pixel gray-scale value in this subwindow longitudinally adds up and calculates on the corresponding one dimension time series basis that forms; Therefrom choosing two fractal dimension extreme values is vertical maximum fractal dimension and vertical minimum fractal dimension;
At last horizontal maximum fractal dimension, horizontal minimum fractal dimension, vertical maximum fractal dimension and vertical minimum fractal dimension as characterizing cloth textured minutia;
The leaching process of described horizontal maximum fractal dimension, horizontal minimum fractal dimension, vertical maximum fractal dimension and vertical minimum fractal dimension is as follows:
At first gather the cloth textured image of digitizing, be designated as W, W is a rectangle, its size is long * wide be L
1* L
2, promptly horizontal and vertical length is respectively L
1And L
2, promptly the row cycle is P and it is along the horizontal basic cycle
1Individual pixel, the basic cycle longitudinally is that line period is P
2, the pixel count after line period and row cycle all refer to round, P
1The basic cycle period of the arbitrary capable pixel grey scale value set by calculating W obtains P
2The basic cycle period of the arbitrary row pixel grey scale value set by calculating W obtains, and wherein the calculating of above-mentioned basic cycle realizes by the one dimension Fast Fourier Transform (FFT);
In cloth textured image W, choose a laterally basic cycle period P
1Wide L for long and cloth textured image
2For wide rectangular window conduct comprises a horizontal subwindow of cycle period substantially, be designated as W
1Choose the long L of a cloth textured image
1Be length, vertically basic cycle period P
2For wide rectangular window conduct comprises a vertical subwindow of cycle period substantially, be designated as W
2
For a certain W
1, calculating the image pixel gray-scale value projection that it follows direction, the image pixel gray-scale value that is about to each row of this subwindow obtains an one dimension time series along laterally stack, can calculate a fractal dimension from this time series, then with W
1To travel through whole W, have L with fixed step size slippage flatly
1-P
1+ 1 W
1Thereby, can correspondingly try to achieve L
1-P
1+ 1 fractal dimension, note reckling and the maximum wherein is E respectively
1And E
2, being horizontal minimum fractal dimension and horizontal maximum fractal dimension, both reflect the lateral extreme detailed information of texture for this;
For a certain W
2, calculating its image pixel gray-scale value projection along column direction, the image pixel gray-scale value that is about to each row of this subwindow longitudinally superposes, and obtains an one dimension time series, can calculate a fractal dimension from this time series, then with W
2To travel through whole W, have L with fixed step size slippage vertically
2-P
2+ 1 W
2Thereby, can correspondingly try to achieve L
2-P
2+ 1 fractal dimension, note reckling and the maximum wherein is E respectively
3And E
4, being vertical minimum fractal dimension and vertical maximum fractal dimension, both reflect the vertically extreme detailed information of texture for this;
Finally obtain characterizing cloth textured proper vector [E
1E
2E
3E
4].
2. a kind of being used to as claimed in claim 1 characterizes cloth textured fractal minutia extracting method, it is characterized in that described fabric is a woven fabric.
3. a kind of being used to as claimed in claim 1 characterizes cloth textured fractal minutia extracting method, it is characterized in that described fractal dimension is meant box counting dimension.
4. a kind of being used to as claimed in claim 3 characterizes cloth textured fractal minutia extracting method, it is characterized in that, used δ size sequence was 2～6 pixels when described box counting dimension was estimated.
5. a kind of being used to as claimed in claim 1 characterizes cloth textured fractal minutia extracting method, it is characterized in that described fixed step size refers to 1～3 pixel.
6. a kind of being used to as claimed in claim 1 characterizes cloth textured fractal minutia extracting method, it is characterized in that described rectangle subwindow W
1Each horizontal sliding fixed step size and W
2Each vertical slippage fixed step size needn't be identical.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105368539A CN101996322B (en) | 2010-11-09 | 2010-11-09 | Method for extracting fractal detail feature for representing fabric texture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105368539A CN101996322B (en) | 2010-11-09 | 2010-11-09 | Method for extracting fractal detail feature for representing fabric texture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101996322A true CN101996322A (en) | 2011-03-30 |
| CN101996322B CN101996322B (en) | 2012-11-14 |
Family
ID=43786460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105368539A Expired - Fee Related CN101996322B (en) | 2010-11-09 | 2010-11-09 | Method for extracting fractal detail feature for representing fabric texture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101996322B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110223309A (en) * | 2019-05-20 | 2019-09-10 | 深圳新视智科技术有限公司 | Edge detection method, device, computer equipment and storage medium |
| CN110889888A (en) * | 2019-10-29 | 2020-03-17 | 桂林理工大学 | Three-dimensional model visualization method integrating texture simplification and fractal compression |
| CN111275111A (en) * | 2020-01-20 | 2020-06-12 | 陕西科技大学 | Classification method for pelts of same animal and same color |
| US10872429B2 (en) | 2016-11-24 | 2020-12-22 | Université D'aix-Marseille | Method for characterising the anisotropy of the texture of a digital image |
| CN114707301A (en) * | 2022-03-09 | 2022-07-05 | 东华大学 | Fractal theory-based riding suit fabric resistance crisis prediction method |
| CN115393303A (en) * | 2022-08-17 | 2022-11-25 | 上海精积微半导体技术有限公司 | Periodic extraction method of periodic structure and wafer defect detection method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6192150B1 (en) * | 1998-11-16 | 2001-02-20 | National University Of Singapore | Invariant texture matching method for image retrieval |
| CN1648321A (en) * | 2005-02-01 | 2005-08-03 | 西安工程科技学院 | Method and device for detecting fabric texture |
| CN101216435A (en) * | 2008-01-03 | 2008-07-09 | 东华大学 | Fabric flaw automatic detection method based on multi-fractal characteristic parameter |
-
2010
- 2010-11-09 CN CN2010105368539A patent/CN101996322B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6192150B1 (en) * | 1998-11-16 | 2001-02-20 | National University Of Singapore | Invariant texture matching method for image retrieval |
| CN1648321A (en) * | 2005-02-01 | 2005-08-03 | 西安工程科技学院 | Method and device for detecting fabric texture |
| CN101216435A (en) * | 2008-01-03 | 2008-07-09 | 东华大学 | Fabric flaw automatic detection method based on multi-fractal characteristic parameter |
Non-Patent Citations (2)
| Title |
|---|
| 《中国纺织大学学报》 20000229 徐增波，等。 基于Wold纹理模型和分形理论的织物疵点检测 第6～9，19页 1-6 第26卷, 第1期 * |
| 《武汉科技学院学报》 20100430 何方容，等。 分形图形在纹织CAD 中的应用 第13～17页 1-6 第23卷, 第2期 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10872429B2 (en) | 2016-11-24 | 2020-12-22 | Université D'aix-Marseille | Method for characterising the anisotropy of the texture of a digital image |
| CN110223309A (en) * | 2019-05-20 | 2019-09-10 | 深圳新视智科技术有限公司 | Edge detection method, device, computer equipment and storage medium |
| CN110889888A (en) * | 2019-10-29 | 2020-03-17 | 桂林理工大学 | Three-dimensional model visualization method integrating texture simplification and fractal compression |
| CN111275111A (en) * | 2020-01-20 | 2020-06-12 | 陕西科技大学 | Classification method for pelts of same animal and same color |
| CN111275111B (en) * | 2020-01-20 | 2023-10-31 | 陕西科技大学 | Method for classifying homozoon furs |
| CN114707301A (en) * | 2022-03-09 | 2022-07-05 | 东华大学 | Fractal theory-based riding suit fabric resistance crisis prediction method |
| CN115393303A (en) * | 2022-08-17 | 2022-11-25 | 上海精积微半导体技术有限公司 | Periodic extraction method of periodic structure and wafer defect detection method |
| CN115393303B (en) * | 2022-08-17 | 2023-11-17 | 上海精积微半导体技术有限公司 | Periodic structure period extraction method and wafer defect detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101996322B (en) | 2012-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101976442B (en) | Method for extracting fractal profile for representing fabric texture and Sobel operator filtering detail mixed characteristic vector | |
| CN101996322B (en) | Method for extracting fractal detail feature for representing fabric texture | |
| CN103729842B (en) | Based on the fabric defect detection method of partial statistics characteristic and overall significance analysis | |
| CN106770323B (en) | The textile flaw detection method filtered based on hierarchical clustering and Gabor | |
| Kilic et al. | Exploiting the synergy between fractal dimension and lacunarity for improved texture recognition | |
| CN103234969B (en) | Method for measuring fabric weft density based on machine vision | |
| CN101976440B (en) | Sobel operator-based extraction method of profile and detail composite characteristic vector used for representing fabric texture | |
| CN105277567B (en) | A kind of fabric defects detection method | |
| CN104751472B (en) | Fabric defect detection method based on B-spline small echo and deep neural network | |
| CN101976441B (en) | Method for extracting Sobel operator filtering profile for representing fabric texture and fractal detail mixed characteristic vector | |
| CN103108586A (en) | A method of generating nuclear magnetic resonance images using susceptibility weighted imaging and susceptibility mapping (swim) | |
| CN107516301A (en) | Construction optimization method for measuring matrix in image reconstruction based on compressed sensing | |
| CN104299232B (en) | SAR image segmentation method based on self-adaptive window directionlet domain and improved FCM | |
| Anandan et al. | Fabric defect detection using discrete curvelet transform | |
| CN101216435A (en) | Fabric flaw automatic detection method based on multi-fractal characteristic parameter | |
| CN108122221A (en) | The dividing method and device of diffusion-weighted imaging image midbrain ischemic area | |
| CN103218832A (en) | Visual saliency algorithm based on overall color contrast ratio and space distribution in image | |
| CN105335972A (en) | Warp knitting fabric defect detection method based on wavelet contourlet transformation and visual saliency | |
| CN104156736A (en) | Polarized SAR image classification method on basis of SAE and IDL | |
| CN108682007A (en) | Jpeg image resampling automatic testing method based on depth random forest | |
| Songdechakraiwut et al. | Dynamic topological data analysis for functional brain signals | |
| CN106023098A (en) | Image repairing method based on tensor structure multi-dictionary learning and sparse coding | |
| CN101996323B (en) | Method for extracting fractal general view and fractal detail mixed characteristic vector for representing fabric texture | |
| CN111862027B (en) | Textile flaw detection method based on low-rank sparse matrix decomposition | |
| CN106296649A (en) | A kind of texture image segmenting method based on Level Set Models |
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: 20121114 Termination date: 20191109 |