CN106127781A - The measuring method of a kind of yarn dyed fabric density and device - Google Patents

Abstract

The invention provides measuring method and the device of a kind of yarn dyed fabric density.Front-back two-sided image capturing system gathers the front-back two-sided image of yarn dyed fabric；Utilize affine transformation to realize the one_to_one corresponding of front-back two-sided image pixel-class, complete the para-position coupling of front-back two-sided image；By Image Fusion based on double-deck wavelet transformation, the gray-scale map of front-back two-sided image is merged；The spectrogram of fusion image is obtained by fast Fourier transform technology；Complete yarn dyed fabric density finally by the characteristic frequency point of correspondence yarn cycle information in extraction spectrogram automatically to measure.

Description

The measuring method of a kind of yarn dyed fabric density and device

Technical field

The present invention relates to Automatic Measurement Technique field, particularly to a kind of side measuring yarn dyed fabric density based on spectrum analysis Method and device.

Background technology

Yarn dyed fabric density is the critical specifications parameter of woven fabric, and fabric and the outward appearance of clothing and physical property are played decision Property effect, the measurement of yarn dyed fabric density is the most also part indispensable in textile product quality testing link.Traditional measurement The method of yarn dyed fabric density be specialty testing staff with the help of pick glass, observe by the naked eye analysis and complete.Use pick glass Although it is simple and practical to measure Density, but there is detection time length, inefficient shortcoming, also it is vulnerable to the skilled of testing staff Degree and the impact of individual's subjective sensation.

Along with the fast development of image processing and artificial intelligence, the research worker of weaving neighborhood start by computer technology by Step is applied to quality of textile products detection, the most just includes the device automatically measuring Density with digitized.Can be from existing skill Art (CN201110174101.7；CN201210094860.7) learn the automatic measurement of Density, be mostly based on one-sided image It is digitized measurement to complete.But yarn dyed fabric is to be interweaved by warp thread and weft yarn to form, filling yarn is the one of fabric Face presents periodically sink-float, so only cannot obtain complete yarn information from yarn dyed fabric one-sided image, thus affects yarn The accuracy of density measure.

Summary of the invention

The measuring method that it is an object of the invention to provide a kind of yarn dyed fabric density and the measurement utilizing the method to design thereof The device of yarn dyed fabric density.The measuring method of a kind of yarn dyed fabric density, utilizes affine transformation to complete the front-back two-sided image of yarn dyed fabric Coupling, utilize Wavelet Fusion algorithm to be merged by front-back two-sided image, then obtain fusion image by Fourier transformation Spectrogram, utilize the characteristic frequency peak dot in spectrogram obtain reconstruct filling yarn image and calculate thread density.One Kind utilize the measurement apparatus of the yarn dyed fabric density that this yarn dyed fabric density measuring method designs, mainly include five parts: CCD phase Machine, standard light source, with standard color block and the sample holder of characteristic matching point, close camera bellows, data handling machine.For realizing Object above, the invention provides the density measuring method of a kind of yarn dyed fabric, comprises the following steps:

1) front-back two-sided for yarn dyed fabric image is carried out image co-registration and obtain fusion image；

2) fusion image is obtained spectrogram by Fourier transform；Spectrogram is processed the filling yarn figure obtaining reconstruct Picture, then calculates yarn dyed fabric density according to the filling yarn image of reconstruct.

Step 1) front-back two-sided for yarn dyed fabric image carried out image co-registration obtain fusion image:

Front-back two-sided for yarn dyed fabric image is completed front-back two-sided images match by affine transformation, then utilizes double-deck small echo Conversion blending algorithm processes and obtains fusion image.

As preferably, complete front-back two-sided images match process by affine transformation and specifically include following steps:

A, image segmentation and Sobel edge edge detective operators is utilized to extract place, three limits of feature triangle straight line；

B, Radon transformation calculations is utilized to go out the linear equation on three limits；

C, calculating triangular apex coordinate obtain its focus point coordinate；

D, using center of gravity as characteristic matching point, obtained the front-back two-sided image mated by affine transformation.

As preferably, in utilizing Wavelet Fusion algorithm to be merged by front-back two-sided image, front-back two-sided image enters The process that row merges specifically includes step:

If the textile image that S is input, SA is the low frequency subgraph picture that ground floor wavelet decomposition obtains, and SH is horizontal high-frequent Image, SV is vertical high frequency subimage, and SD is diagonal angle high frequency subimage；SSA, SSH, SSV and SSD are respectively the little wavelength-division of the second layer Solve low frequency subgraph as corresponding four subimages of SA；

A) utilize double-deck wavelet decomposition fabric front-back two-sided image S1 and S2, obtain the subimage after it decomposes；

B) the horizontal high-frequent subimage that direct picture S1 and verso images S2 obtains after double-deck wavelet decomposition is extracted respectively SSH1 and SSH2；In horizontal high-frequent subimage, calculate centered by each pixel, in 3x3 region, amount to 9 pixels Local variance value, if D_SSH1And D_SSH2Represent the local variance value of calculated SSH1 and SSH2, then fusion image the 3rd respectively In layer horizontal high-frequent subimage SSH, the value of respective pixel point can obtain by below equation:

$SSH(m,n)=\left\{\begin{array}{cc}SSH1(m,n),& D_{SSH1}(m,n)\&GreaterEqual;D_{SSH2}(m,n)\\ SSH2(m,n),& D_{SSH1}(m,n)<D_{SSH2}(m,n)\end{array}\right.$

Wherein, m, n are respectively picture level direction pixel number and vertical direction pixel number；

C) use and B) identical computational methods, utilize SSV1 and SSV2, SSA1 and SSA2, SSD1 and SSD2 can obtain Fusion image third layer vertical high frequency subimage SSV and diagonal angle high frequency subimage SSD and front low frequency subgraph are as SSA；

D) utilize the SSH, SSV and the SSD that calculate and SSA, obtain fusion image the by 2-d discrete wavelet inverse transformation Two layers of low frequency subgraph are as SA；

E) use and step B) to D) the horizontal high-frequent subimage SH of the identical method calculating fusion image second layer, vertically High frequency subimage SV and diagonal angle high frequency subimage SD；

F) utilize SH, SV, SD and the SA calculated, finally give fusion image S by 2-d discrete wavelet inverse transformation.

Step 2) fusion image is obtained spectrogram by Fourier transform, spectrogram is processed and obtains filling yarn reconstruct Figure, then calculates yarn dyed fabric density according to the filling yarn image of reconstruct.

As preferably, fusion image is obtained spectrogram by Fourier transform, spectrogram is processed and obtains through weft yarn Line reconstruct figure, then calculates yarn dyed fabric density according to the filling yarn image of reconstruct, specifically includes following steps:

1, fast Fourier transform technology is utilized to obtain the spectrogram of fusion image；

2, utilize peak dot filtering to extract in spectrogram and be positioned at the characteristic frequency point both horizontally and vertically gone up, represent warp respectively Yarn and weft yarn cycle；

3, utilize the peak dot information extracted, obtain filling yarn reconstruct figure by Fourier inversion and calculate yarn Density, as it is furthermore preferred that the computing formula of thread density is:

$\left\{\begin{array}{c}\frac{R}{1}=\frac{x}{s_w}\\ \frac{R}{1}=\frac{y}{s_h}\end{array}\right.\&DoubleRightArrow;\left\{\begin{array}{c}{s}_w=\frac{x}{R}\\ s_h=\frac{y}{R}\end{array}\right.$

$\left\{\begin{array}{c}{d}_j=\frac{n_j}{s_w}\\ d_w=\frac{n_w}{s_h}\end{array}\right.$

Wherein, in formula: R represents the resolution gathering image；S_w, S_hBeing width and the height of original image respectively, unit is Inch；X, y are respectively original image pixel count both horizontally and vertically；d_j, d_wIt is respectively warp count and weft count.

A kind of yarn dyed fabric density measuring equipment utilizing the method to design, mainly includes five parts: CCD camera, standard Light source, with standard color block and the sample holder of characteristic matching point, close camera bellows, data handling machine.CCD camera, standard Light source, sample holder with standard color block and characteristic matching point are placed in closing camera bellows；Sample holder is vertically placed on camera bellows On bottom centerline；Yarn dyed fabric sample to be measured is positioned in sample holder；Standard light source is symmetrically disposed at yarn dyed fabric both sides；At color The symmetrically placed CCD camera in fabric both sides；CCD camera is connected with data handling machine；

After opening standard light source, the CCD camera shooting front-back two-sided image of yarn dyed fabric, image information transmission is processed to data Computer, data handling machine utilizes the described corresponding algorithm of yarn dyed fabric density measuring method to complete yarn dyed fabric density measure.

Preferably, sample holder includes the two pieces of grip blocks mutually clamped, and the surface of grip block is provided with 24 color calibration color lumps Triangle is mated with para-position；Between grip block, the magnet adsorption by being mutually matched clamps.Preferably, yarn dyed fabric density measure dress Putting Plays light source is four, and symmetry is placed in the right and left of sample holder.

In the present invention, the front-back two-sided image of yarn dyed fabric is gathered by front-back two-sided image capturing system；Utilize affine change Change the one_to_one corresponding realizing front-back two-sided image pixel-class, complete two-sided para-position coupling；By figure based on double-deck wavelet transformation As blending algorithm, the gray-scale map of front-back two-sided image is merged；Fusion image is obtained by fast Fourier transform technology Spectrogram；The survey of yarn dyed fabric density is completed finally by extracting the characteristic frequency point of corresponding yarn cycle information in spectrogram Amount.

Accompanying drawing explanation

Fig. 1 is the yarn dyed fabric density measuring equipment in the present invention；1-camera bellows in figure, 2-computer, 3-CCD camera, 4-marks Quasi-optical source, 5-sample holder；

Fig. 2 is the sample holder schematic diagram in yarn dyed fabric density measuring equipment；7-match point, 8-standard color block, 9-rectangle Magnetic stripe；

Fig. 3 is the measuring method flow chart of yarn dyed fabric density in the present invention；FFT is two-dimensional fast fourier transform, IFFT For two dimension Fast Fourier Transform Inverse；

Fig. 4 is double-deck wavelet decomposition tree schematic diagram in the present invention；

Fig. 5 is example sample fabric face gradation of image figure in the present invention；

Fig. 6 is example sample fabric backing gradation of image figure in the present invention；

Fig. 7 is fusion image in the present invention；

Fig. 8 is small echo blending algorithm flow chart in the present invention；

Fig. 9 is spectrogram and the filling yarn reconstruct figure of fusion image in the present invention；A-fusion image spectrogram, b-level Two peak dots on direction, two peak dots in c-vertical direction, d-warp thread reconstruct figure, e-weft yarn reconstruct figure.

Detailed description of the invention

Below in conjunction with specific embodiments and the drawings, the present invention is expanded on further.

Fig. 1 is yarn dyed fabric density measuring equipment, and it mainly includes 5 parts: 1, two symmetrically placed high resolution CCD Camera, for gathering the front-back two-sided image of yarn dyed fabric；2, the sample holder of a set of autonomous Design, as in figure 2 it is shown, sample holder Surface comprises four characteristic matching points for carrying out the para-position of front-back two-sided image and mating, and 24 colour standard color lumps be used for into Row color of image is calibrated；Inside sample holder, symmetry fixes four rectangle magnetic stripes, keeps certain when making fabric sample held Power.3, four standard light source symmetries are placed in the right and left of sample holder, are used for providing bilateral uniform irradiation.4, a closing Camera bellows, for get rid of CCD camera gather image time ambient light impact.5, a data handling machine, for collection Image process and analyze.

In the present embodiment, the resolution of CCD camera is 14,000,000 pixels.During use, utilize sample holder to clamp and knit Thing is vertically placed on two camera centre positions, the most corresponding camera of fabric tow sides.Guarantee camera bellows does not has the external world After light enters, opening four standard light sources, two cameras shoot simultaneously, can collect the image of fabric.

By arranging the calibration color lump of 24 colors in sample holder so that the present invention after collecting the image of fabric, energy Enough according to calibration color lump, color is proofreaded, make the fabric color collected closer to real color, reduce and gather image Aberration.

Present embodiment additionally provides the measuring method of a kind of yarn dyed fabric density, first with front-back two-sided image acquisition system System gathers the front-back two-sided image of yarn dyed fabric, after extracting the subimage containing characteristic matching point, is mated by affine transformation The front-back two-sided image completed, utilizes Wavelet Fusion algorithm to be merged by front-back two-sided image, then passes through fast Fourier Conversion obtains the spectrogram of fusion image, utilizes the characteristic frequency peak dot in spectrogram to obtain the filling yarn image of reconstruct, After calculate thread density.Concrete algorithm performs flow chart as shown in Figure 3.

After collecting the front-back two-sided image of fabric, extract the subimage containing matching characteristic point for front-back two-sided image Para-position with mate,

Extract place, three limits of feature triangle straight line first with image segmentation and Sobel edge edge detective operators, utilize After Radon transformation calculations goes out the linear equation on three limits, obtain its focus point coordinate by calculating triangular apex coordinate, will weight The heart, as characteristic matching point, obtains the front-back two-sided image mated, it is achieved front-back two-sided image finally by affine transformation The one_to_one corresponding of Pixel-level.After fabric positive and negative images match completes, Wavelet Fusion algorithm is utilized to be carried out by front-back two-sided image Merge, strengthen the periodic structure information of yarn in image, measure for thread density.The mathematic(al) representation of wavelet transformation is such as Under:

$W_a(a,b)={\&Integral;}_{-\mathrm{\&infin;}}^{+\mathrm{\&infin;}}f\left(t\right){\mathrm{\&Psi;}}_{a,b}\left(t\right)dt={\&Integral;}_{-\mathrm{\&infin;}}^{+\mathrm{\&infin;}}f\left(t\right)a^{-\frac{1}{2}}\mathrm{\&Psi;}\left(\frac{t-b}{a}\right)\left(t\right)dt$

In formula: a is scale factor, b is translation vector.

Wavelet transformation is divided into continuous wavelet transform and wavelet transform.Owing to the digital signal of image is discrete type, Therefore use discrete wavelet that image is carried out conversion process.Two dimension off-line wavelet transformation and inverse transformation expression formula thereof are as follows:

$C_{a,b}={\&Integral;}_{-\mathrm{\&infin;}}^{+\mathrm{\&infin;}}f\left(t\right){\mathrm{\&Psi;}}_{a,b}\left(t\right)dt$

$f\left(t\right)=c\underset{-\mathrm{\&infin;}}{\overset{+\mathrm{\&infin;}}{\&Sigma;}}\underset{-\mathrm{\&infin;}}{\overset{+\mathrm{\&infin;}}{\&Sigma;}}{C}_{a,b}{\mathrm{\&Psi;}}_{a,b}\left(t\right)$

In formula: a is scale factor, b is translation vector.

Select to use double-deck wavelet transformation respectively fabric positive and negative image to be decomposed, its wavelet decomposition tree such as Fig. 4 institute Show.Wherein, S is the textile image of input, and SA is the low frequency subgraph picture that ground floor wavelet decomposition obtains, and SH is horizontal high-frequent subgraph Picture, SV is vertical high frequency subimage, and SD is diagonal angle high frequency subimage；SSA, SSH, SSV and SSD are respectively second layer wavelet decomposition Low frequency subgraph is as corresponding four subimages of SA.

Assume that S1 represents the fabric face image mated, as shown in Figure 5；S2 represents the gray-scale map of verso images, as Shown in Fig. 6.S represents fusion image, then front-back two-sided Image Fusion based on wavelet transformation is specific as follows, its flow process such as figure Shown in 8:

(1) utilize double-deck wavelet transform to decompose fabric front-back two-sided image S1 and S2, obtain the son after it decomposes Image.

(2) the horizontal high-frequent subimage that direct picture S1 and verso images S2 obtains after double-deck wavelet decomposition is extracted respectively SSH1 and SSH2.In subimage, calculate the local variance amounting to 9 pixels centered by each pixel in 3x3 region Value, it is assumed that D_SSH1And D_SSH2Represent the local variance value of calculated SSH1 and SSH2, then fusion image third layer level respectively In high frequency subimage, the value of respective pixel point can obtain by below equation:

$SSH(m,n)=\left\{\begin{array}{cc}SSH1(m,n),& D_{SSH1}(m,n)\&GreaterEqual;D_{SSH2}(m,n)\\ SSH2(m,n),& D_{SSH1}(m,n)<D_{SSH2}(m,n)\end{array}\right.$

Wherein m, n represent the pixel count of horizontal direction and vertical direction successively.

(3) utilize the SSH, SSV and the SSD that calculate and SSA, obtain fusion image the by 2-d discrete wavelet inverse transformation Two layers of low frequency subgraph are as SA.

(4) method identical with (2nd) step is used to calculate the horizontal high-frequent subimage of the fusion image second layer, vertical high frequency Subimage SV and diagonal angle high frequency subimage SD.

(5) utilize SH, SV, SD and the SA calculated, finally give fusion image S by 2-d discrete wavelet inverse transformation, As shown in Figure 7.

After being merged by front-back two-sided for fabric image by double-deck wavelet transformation, utilize two-dimensional fast fourier transform (FFT) and inverse transformation (IFFT) technology obtains the spectrogram of fusion image, extract the cyclic component in spectrogram, finally record Yarn dyed fabric filling yarn density.

Assume two-dimensional discrete function f (x, y) represents fusion image, then its Fourier transformation and inverse transformation expression formula are as follows:

$F(u,v)={\&Integral;}_{\mathrm{\&infin;}}^{+\mathrm{\&infin;}}{\&Integral;}_{-\mathrm{\&infin;}}^{+\mathrm{\&infin;}}f(x,y)e^{-i2\mathrm{\&pi;}(ux+vy)}dxdy$

$f(x,y)={\&Integral;}_{-\mathrm{\&infin;}}^{+\mathrm{\&infin;}}{\&Integral;}_{-\mathrm{\&infin;}}^{+\mathrm{\&infin;}}F(u,v)e^{i2\mathrm{\&pi;}(ux+vy)}dudv$

In formula: u, v are frequency variable, u=0,1 ..., M-1, v=0,1 ..., N-1.Due to the digital signal of image be from Dissipate type, therefore use discrete Fourier transform that image is processed.Two dimensional discrete Fourier transform and inverse transformation thereof are expressed Formula is as follows:

$F(u,v)=\frac{1}{MN}\underset{x=0}{\overset{M-1}{\&Sigma;}}\underset{y=0}{\overset{N-1}{\&Sigma;}}f(x,y)\exp \&lsqb;-i2\mathrm{\&pi;}(\frac{ux}{M}+\frac{vy}{N})\&rsqb;$

$f(x,y)=\underset{u=0}{\overset{M-1}{\&Sigma;}}\underset{v=0}{\overset{N-1}{\&Sigma;}}F(u,v)\exp \&lsqb;i2\mathrm{\&pi;}(\frac{ux}{M}+\frac{vy}{N})\&rsqb;$

In formula: u=0,1 ..., M-1, v=0,1 ..., N-1.When the data volume of picture is the biggest, discrete fourier becomes The treatment effeciency of scaling method is relatively low.Therefore, the fast Fourier grown up on the basis of two dimensional discrete Fourier transform Mapping algorithm is widely used in image processing field.

Utilize fast Fourier transform technology, woven fabric picture can be changed to frequency domain from transform of spatial domain, and obtain phase The spectrogram answered.Peak dot in spectrogram represents the periodic structure information of fabric, two peaks on the horizontal line of initial point Point correspond to warp thread periodical information, and two peak dots on the vertical line of initial point correspond to weft yarn periodical information.Elected After determining the peak dot in both direction, filling yarn original image can be reconstructed by Fast Fourier Transform Inverse, such as Fig. 9 institute Show: original fusion image S obtains fusion image spectrogram a, fusion image spectrogram a through Fourier transform and extracts level side Peak dot figure c on peak dot figure b upwards and vertical direction, then peak dot figure reconstructs warp thread reconstruct figure by inverse fourier transform D and weft yarn reconstruct figure e.Utilizing filling yarn reconstruct figure, can accurately calculate the thread density of fabric, computing formula is as follows:

$\left\{\begin{array}{c}\frac{R}{1}=\frac{x}{s_w}\\ \frac{R}{1}=\frac{y}{s_h}\end{array}\right.\&DoubleRightArrow;\left\{\begin{array}{c}{s}_w=\frac{x}{R}\\ s_h=\frac{y}{R}\end{array}\right.$

$\left\{\begin{array}{c}{d}_j=\frac{n_j}{s_w}\\ d_w=\frac{n_w}{s_h}\end{array}\right.$

In formula: R represents the resolution gathering image；S_w, S_hBeing width and the height of original image respectively, unit is inch； X, y are respectively original image pixel count both horizontally and vertically；d_j, d_wIt is respectively warp count and weft count.

The fabric warp density that this example finally records is 70.3 pieces/inch, and weft count is 51.6 pieces/inch, manually What pick glass recorded is 70.1/ inch through weft count, and 50.9/ inch, both errors are only 0.28% and 1.37% respectively.

The respective embodiments described above are to realize the specific embodiment of the present invention, it will be understood by those skilled in the art that And in actual applications, can to it, various changes can be made in the form and details, without departing from the spirit and scope of the present invention.

Claims (8)

1. the measuring method of a yarn dyed fabric density, it is characterised in that comprise the following steps: 1) by front-back two-sided for yarn dyed fabric image Carry out image co-registration and obtain fusion image；2) fusion image is obtained spectrogram by Fourier transform, spectrogram is processed To the filling yarn image of reconstruct, then calculate yarn dyed fabric density according to the filling yarn image of reconstruct.

The measuring method of yarn dyed fabric density the most according to claim 1, it is characterised in that: by front-back two-sided for yarn dyed fabric image Complete images match by affine transformation, then utilize double-deck Wavelet Transform Fusion algorithm to obtain fusion image.

The measuring method of yarn dyed fabric density the most according to claim 2, it is characterised in that: described completed by affine transformation Images match comprises the following steps:

A, image segmentation and Sobel edge edge detective operators is utilized to extract place, three limits of feature triangle straight line；

B, Radon transformation calculations is utilized to go out the linear equation on three limits；

C, calculating triangular apex coordinate obtain its focus point coordinate；

D, using center of gravity as characteristic matching point, obtained the front-back two-sided image mated by affine transformation.

4. according to the measuring method of the yarn dyed fabric density described in Claims 2 or 3, it is characterised in that:

Described double-deck wavelet image blending algorithm comprises the following steps: setting the S textile image as input, SA is that ground floor is little The low frequency subgraph picture that Wave Decomposition obtains, SH is horizontal high-frequent subimage, and SV is vertical high frequency subimage, and SD is diagonal angle high frequency subgraph Picture；SSA, SSH, SSV and SSD are respectively second layer wavelet decomposition low frequency subgraph as corresponding four subimages of SA；

A) double-deck wavelet decomposition fabric front-back two-sided image S1 and S2, the subimage after being decomposed are utilized；

B) the horizontal high-frequent subimage SSH1 that direct picture S1 and verso images S2 obtains after double-deck wavelet decomposition is extracted respectively And SSH2；In horizontal high-frequent subimage, calculate the local amounting to 9 pixels centered by each pixel in 3x3 region Variance yields, if D_SSH1And D_SSH2Represent the local variance value of calculated SSH1 and SSH2, then fusion image third layer water respectively In flat high frequency subimage SSH, the value of respective pixel point can obtain by below equation:

$SSH(m,n)=\left\{\begin{array}{cc}SSH1(m,n),& D_{SSH1}(m,n)\&GreaterEqual;D_{SSH2}(m,n)\\ SSH2(m,n),& D_{SSH1}(m,n)<D_{SSH2}(m,n)\end{array}\right.$

Wherein, m representative picture horizontal direction pixel number, n represents vertical direction pixel number；

C) use and B) identical computational methods, utilize SSV1 and SSV2, SSA1 and SSA2, SSD1 and SSD2 can be merged Image third layer vertical high frequency subimage SSV and diagonal angle high frequency subimage SSD and front low frequency subgraph are as SSA；

D) utilize SSH, SSV and the SSD calculated and SSA, obtain the fusion image second layer by 2-d discrete wavelet inverse transformation Low frequency subgraph is as SA；

E) use and step B) to D) the horizontal high-frequent subimage SH of the identical method calculating fusion image second layer, vertical high frequency Subimage SV and diagonal angle high frequency subimage SD；

F) utilize SH, SV, SD and the SA calculated, finally give fusion image S by 2-d discrete wavelet inverse transformation.

5. according to the measuring method of the yarn dyed fabric density described in claim 1 or 2 or 3, it is characterised in that: measure yarn dyed fabric density Data handling procedure in, image obtains spectrogram by fast Fourier transform,

The characteristic frequency peak dot in spectrogram is utilized to obtain filling yarn reconstruct figure through Fourier inversion and calculate yarn Density；The computing formula of thread density is:

$\left\{\begin{array}{c}\frac{R}{1}=\frac{x}{s_w}\\ \frac{R}{1}=\frac{y}{s_h}\end{array}\right.\&DoubleRightArrow;\left\{\begin{array}{c}{s}_w=\frac{x}{R}\\ s_h=\frac{y}{R}\end{array}\right.$

$\left\{\begin{array}{c}{d}_j=\frac{n_j}{s_w}\\ d_w=\frac{n_w}{s_h}\end{array}\right.$

Wherein, in formula: R represents the resolution gathering image；S_w, S_hBeing width and the height of original image respectively, unit is English Very little；X, y are respectively original image pixel count both horizontally and vertically；d_j, d_wIt is respectively warp count and weft count.

6. a yarn dyed fabric density measuring equipment for the measuring method of yarn dyed fabric density according to claim 1, its feature Be: mainly include five parts: CCD camera, standard light source, with standard color block and the sample holder of characteristic matching point, envelope Close camera bellows, data handling machine；

CCD camera, standard light source, sample holder with standard color block and characteristic matching point are placed in closing camera bellows；Sample presss from both sides Tool is vertically placed on camera bellows bottom centerline；Yarn dyed fabric sample to be measured is positioned in sample holder；Standard light source is symmetrically placed On yarn dyed fabric both sides；In the symmetrically placed CCD camera in yarn dyed fabric both sides；CCD camera is connected with data handling machine；

After opening standard light source, the CCD camera shooting front-back two-sided image of yarn dyed fabric, image information transmission is processed to data and calculates Machine, computer completes yarn dyed fabric density measure.

Yarn dyed fabric density measuring equipment the most according to claim 6, it is characterised in that: sample holder includes mutually clamping Two pieces of grip blocks, the surface of grip block is provided with 24 color calibration color lumps and para-position coupling triangle；By mutual between grip block The magnet adsorption clamping joined.

Yarn dyed fabric density measuring equipment the most according to claim 6, it is characterised in that: yarn dyed fabric density measuring equipment is got the bid Quasi-optical source is four, and symmetry is placed in the right and left of sample holder.