CN106198542B - A kind of knitted fabric industrial analysis method based on smart mobile phone - Google Patents
A kind of knitted fabric industrial analysis method based on smart mobile phone Download PDFInfo
- Publication number
- CN106198542B CN106198542B CN201610522846.0A CN201610522846A CN106198542B CN 106198542 B CN106198542 B CN 106198542B CN 201610522846 A CN201610522846 A CN 201610522846A CN 106198542 B CN106198542 B CN 106198542B
- Authority
- CN
- China
- Prior art keywords
- mrow
- image
- msub
- fabric
- textile
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N2021/1765—Method using an image detector and processing of image signal
- G01N2021/177—Detector of the video camera type
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Treatment Of Fiber Materials (AREA)
- Image Processing (AREA)
Abstract
Quickly analyzed the invention discloses a kind of knitted fabric technique based on smart mobile phone, including the server for being handled knitted fabric image, being analyzed, and for carrying out the smart mobile phone of fabric positive and negative IMAQ, smart mobile phone can be pre-processed to image and enter row information with server and mutually be passed；The invention also discloses a kind of knitted fabric technique rapid analysis method, smart mobile phone gathers the front-back two-sided image of knitted fabric and is sent to server, server receives the textile image that described smart mobile phone sends and is handled, analyzed, and analysis result then is sent into smart mobile phone；Technique is carried out to knitted fabric quickly to analyze, realize knitted fabric industrial analysis automation, digitized requirement using said apparatus and method, it is easy to use, it is easy to popularize；And user can obtain the technique information of fabric whenever and wherever possible, industrial analysis efficiency is improved, the technical threshold of industrial analysis is eliminated.
Description
Technical field
The invention belongs to fabric weaving field, more particularly to a kind of weaving image processing field.
Background technology
Knitted fabric institutional framework has extremely close relation with its performance, influences the strength, draftability, fabric wind of fabric
Lattice and feel etc., therefore, to the problem of the analysis of knitted fabric institutional framework to be one extremely have researching value.But traditional warp
Braid contextual analysis of organization method is all manual analysis, is completed using microscope with the magnifying glass with scale, analyzes process
Dull, dull, analysis efficiency is relatively low, and higher to analysis personnel's technical requirements.Increase and weaving row with labor cost
Industry automaticity lifted, automation, Weigh sensor knitted fabric institutional framework turn into one it is urgently to be resolved hurrily and with application
The problem of prospect.
With the development of computer technology, application of the Morden Image Processing Technology in textile manufacturing is gradually extensive, warp knit
The automatic analysis method of fabric construction starts to find application, and existing fabric construction autoanalyzer method is main by scanner or number
The computer software part composition of image collecting device and the fabric construction analysis of camera, this method still has following deficiency
Place：1) scanner or digital camera complex operation, not portable, and user's popularity is not high, it is impossible to meets user whenever and wherever possible
The purpose used；2) the Knitting Principle of woven fabric and knitted fabric is far from each other, and it is relatively simple woven that device is applied to structure
Thing, can not be applicable for baroque warp-knitted fabric.
The content of the invention
In view of the foregoing, quickly divide it is an object of the invention to provide a kind of knitted fabric technique based on smart mobile phone
Analysis apparatus, its simple and convenient, the art flower of being capable of automatic measurement and identification fabric is wide, art flower is high, the horizontal close, finished product of finished product is indulged
Close, institutional framework and pass circulation etc., improve knitted fabric industrial analysis efficiency, meet the requirement that user uses whenever and wherever possible.
Present invention also offers a kind of knitted fabric technique rapid analysis method based on smart mobile phone.
To achieve the above object, the present invention uses following technical scheme：
A kind of knitted fabric technique fast analyser based on smart mobile phone, including smart mobile phone and server, it is described
Smart mobile phone is used to gather the front-back two-sided image of fabric and sent to server, and the server is used to receive the smart mobile phone
The textile image that sends simultaneously is handled, analyzed, and analysis result then is sent into smart mobile phone.
Further, the smart mobile phone is provided with:
Camera, for shooting and gathering the front-back two-sided image of fabric；
The pretreatment module of image first, for judging picture quality and carrying out corresponding white balance processing and framing sanction
Cut；
Image sending module, for sending textile image to server.
Information receiving module, the fabric analysis result sent for the reception server.
Further, it is provided with the server：
Image receiver module, for receiving the textile image sent by smart mobile phone sending module；
The pretreatment module of image second, the textile image for being received to image receiver module carries out gray processing processing simultaneously
Detect whether containing fabric；
Image processing module, for by the second pretreatment module processing after textile image carry out threshold transformation, from
The processing such as adaptive filtering, Fast Fourier Transform (FFT), rim detection, obtains the periodic feature of textile processes positive and negative image respectively
Point；
Fabric analysis module, for being analyzed textile image Weave parameters, being calculated, draws fabric analysis knot
Really, including art flower it is wide, art flower is high, the horizontal close, finished product of finished product indulges close, institutional framework and pass circulation etc.；
As a result feedback module, for fabric analysis result to be sent into smart mobile phone client.
A kind of knitted fabric technique rapid analysis method based on smart mobile phone, comprises the following steps：
S1 smart mobile phones gather knitted fabric dual-side image and are sent to server；
S2 servers receive the textile image that described smart mobile phone sends and are handled, analyzed, then will analysis knot
Fruit is sent to smart mobile phone.
Further, the step S1 specifically include it is following step by step：
S11 detects ambient lighting, obtains the illumination value of surrounding environment by the light sensor of smart mobile phone, presets
Illumination value span, when ambient lighting value is not in default scope, image obtains failure；
S12 textile image previews, set fabric shooting preview frame, to be presented on fabric pre- by adjusting camera site by user
Look in frame and ensure the textile image in preview pane at least containing a complete cycle size；
S13 is pre-processed for the first time：By smart mobile phone by the image collected in step S12 be amplified 3-6 times it is laggard
Row location cutting and the AWB processing for carrying out dynamic threshold, to obtain the textile image for meeting fabric analysis requirement；
S14 textile images are sent, the image sending module of smart mobile phone by the textile image obtained in step S13 send to
Server.
Further, the step pre-processes white balancing treatment method in S13 and included for the first time：
Textile image is YC by RGB patten transformations by S131bCr, N number of part is divided the image into after converting, is counted respectively
Calculate the C of each partbAnd CrAverage value Mb、MrAnd both mean square deviation Db、Dr, computational methods are as follows：
The average value M in each region of S132 statistics textile imagesb、MrAnd mean square deviation Db、DrAs whole image value (most
After need to cast out Db<0.05MbAnd/or Dr<0.05MrWhen value)；
S133 determines the white reference point in textile image as follows,
|Cb(i,j)-(Mb+Db×sign(Mb)) | 1.5 × D of ＜b
|Cr(i,j)-(Mr+Dr×sign(Mr)) | 1.5 × D of ＜r
10% white reference point is used as final white reference point before brightness value；
S134 calculates each white reference point average brightness R of textile imageav, Gav, Bav, the gain R of each passagegain, Ggain,
BgainWith image in YCbCrY maximum Y under patternmax；
S135 calculates the final R ' G ' B ' of each passage of textile image：
R '=R × Rgain
G '=G × Ggain
B '=B × Bgain
Further, the step S2 include it is following step by step：
S21 textile images are received, and server receiving module receives the warp sent by smart mobile phone image sending module in advance
Textile image after reason；
S22 textile images are pre-processed, and the textile image that the pretreatment module of server second is received to step S21 carries out ash
Degreeization is handled and detected whether containing fabric；
The processing of S23 textile images, image processing module carries out threshold transformation to the image after step S22 processing, adaptive
The processing such as filtering, Fast Fourier Transform (FFT), rim detection, obtains the periodic feature point of textile processes positive and negative image respectively；
S24 textile images are analyzed, and fabric analysis module is analyzed the image in step S23, draws fabric analysis knot
Really, including art flower it is wide, art flower is high, the horizontal close, finished product of finished product indulges close, institutional framework and pass circulation etc.；
S25 analysis results are fed back, and as a result feedback module sends the fabric analysis result drawn in step S24 to intelligent hand
In machine.
Preferably, the image gray processing processing in the step S22 is converted by below equation：
Gray (i, j)=0.11R (i, j)+0.59G (i, j)+0.3B (i, j)
Preferably, the step S23 includes following sub-step：
S231 threshold transformations, using the method for threshold transformation by the fabric tow sides image binaryzation of extraction, image is grey
The threshold transformation method of degree is as follows：
S232 is adjusted according to the local variance of image and filtered by the noise in Adaptive noise cancellation fabric two-face image
Output, it is ensured that image border and detail of the high frequency.
The space area image of textile processes direct picture is converted to frequency spectrum by S233 by Fast Fourier Transform (FFT) (FFT)
Figure：
S234 carries out rim detection to the textile processes verso images after the processing of step S231, S232, obtains fabric
The edge graph of technical back.
Preferably, the step S24 includes following sub-step：
Textile processes front frequency domain figure picture after the processing of S241 server images processing module is in the horizontal and vertical directions
The bright band of rule is distributed, horizontal direction bright band represents fabric row information, and vertical direction bright band represents fabric stringer information, point
It is other that frequency domain figure picture is both horizontally and vertically projected；
The central point of S242 selected levels (vertical) direction projection image, along vertical (horizontal) direction point by point scanning image ash
Degree, until scanning is arrived the pixel when gray value is 255 as starting point, continues along vertical (horizontal) direction point by point scanning, until
Scanning next time stops when being 255 to gray value, the pixel count m (n) of writing scan, adjacent two row of m (n) extremely fabrics
Frequency between (stringer).
The image resolution ratio that S243 smart mobile phones 100 are shot is r (pixel/inch), the textile image size of location cutting
For M × N (pixel), the wide W of art flower of the fabricidth, the high H of art flowereight, the horizontal close W of finished productPCClose C is indulged with finished productPCCan be with
It is expressed as：
Width=n
Height=m
Textile processes verso images after the processing of the image processing module 230 of S244 servers 200, to the fabric edge of extraction
Multiple target area parameters for shape characteristic are identified in figure, are judged pixel-by-pixel with element marking method, obtain tissue and follow
Coil underlap edge parameters in ring, and then analyze, calculate the institutional framework of fabric and pass circulation.
After adopting the above technical scheme, the present invention has the following advantages that compared with background technology：
1st, the present invention obtains textile image information by smart mobile phone, using the online wireless technology of smart mobile phone by fabric
Image information is sent to long-range high-performance server, realizes knitted fabric industrial analysis automation, digitized requirement, letter
It is single easy-to-use, it is easy to popularize；
2nd, the present invention automatic identification knitted fabric information and can be analyzed, calculated, and user can whenever and wherever possible obtain and knit
The technique information of thing, including art flower is wide, art flower is high, the horizontal close, finished product of finished product indulges close, institutional framework and passes circulation etc., carries
High technology analysis efficiency, eliminates the technical threshold of industrial analysis.
Brief description of the drawings
Fig. 1 is a kind of knitted fabric technique fast analyser interface based on smart mobile phone of the present invention with reference to figure；
Fig. 2 is the schematic flow sheet of knitted fabric technique rapid analysis method of the present invention；
Fig. 3 a are fabric face image of the knitted fabric after cutting in a kind of embodiment；
Fig. 3 b are fabric backing image of the knitted fabric after cutting in a kind of embodiment；
Fig. 4 a are fabric face images of Fig. 3 a after threshold transformation；
Fig. 4 b are fabric backing images of Fig. 3 b after threshold transformation；
Fig. 5 is the textile processes front spectrogram after 4a is handled through processing module；
Fig. 6 is the textile processes opposite longitudinal side edge figure after 4a is handled through processing module.
Embodiment：
In order that the object of the invention, technical scheme and advantage are clearer, below in conjunction with drawings and Examples, to the present invention
It is described further.It should be appreciated that specific embodiment described herein is only to explain the present invention, but it is not used to limit this hair
It is bright.
Embodiment
The invention discloses a kind of knitted fabric technique fast analyser based on smart mobile phone, including smart mobile phone and
Server, wherein：
Smart mobile phone is used to gather the front-back two-sided image of fabric and be sent to server, is provided with camera, image
First pretreatment module and image sending module and information receiving module；Camera is used to shoot and gather textile image；Image
First pretreatment module is used to judge textile image quality and carries out white balance processing and framing cutting；Image sending module
For sending textile image to server；Information receiving module is used for the fabric analysis result that the reception server is sent.
Server is used to receiving the textile image that smart mobile phone sends and is handled, analyzed and calculated, and will analysis knot
Fruit is sent to smart mobile phone, and server sets image receiver module, the pretreatment module of image second, image processing module, fabric
Analysis module and result fabric analysis result feedback module.Image receiver module is used to receive to be sent by smart mobile phone sending module
Textile image；The pretreatment module of image second is used to the textile image that receiving module is received is smoothed and detected
It is no to have fabric；Image processing module is used for the textile image progress threshold transformation after the processing of the second pretreatment module, adaptively
The processing such as filtering, Fast Fourier Transform (FFT), rim detection, obtains the periodic feature point of textile processes positive and negative image respectively；
Fabric analysis module is used to carry out Weave parameters analysis to the textile image after image processing module processing, calculated, and draws
Fabric analysis result, including art flower is wide, art flower is high, the horizontal close, finished product of finished product indulges close, institutional framework and passes circulation etc.；Knot
Fruit feedback module, for the analysis result of fabric analysis module to be sent into smart mobile phone.
With reference to Fig. 1 and Fig. 2, a kind of knitted fabric technique rapid analysis method based on smart mobile phone is special with a kind of double combs
In exemplified by section's knitted fabric, it comprises the following steps：
S1 smart mobile phones gather the front-back two-sided image of knitted fabric and are sent to server, and the step is through following real step by step
It is existing：
S11 detects ambient lighting, and ambient lighting value is obtained by smart mobile phone light sensor, and this step needs to set in advance
Good illumination span, if ambient lighting value is not in default scope, image obtains failure；
S12 textile image previews, set fabric shooting preview frame, and user can be presented on fabric by adjusting picture-taking position
In preview pane and ensure the textile image in preview pane at least containing a complete cycle size；
S13 is pre-processed for the first time：By smart mobile phone by the image collected in step S12 be amplified 3-6 times it is laggard
Row location cutting and the AWB processing based on progress dynamic threshold, to obtain the fabric figure for meeting fabric analysis requirement
Picture；(with reference to Fig. 3 a and Fig. 3 b), white balance processing specific method is as follows：
By textile image by RGB patten transformations be YCbCr, N number of part is divided the image into after converting, calculates every respectively
The C of individual partbAnd CrAverage value Mb、MrAnd both mean square deviation Db、Dr, computational methods are as follows：
Count the average value M in each region of textile imageb、MrAnd mean square deviation Db、DrAs the value of whole image, if certain portion
The mean square deviation divided is less than normal, omits and (casts out Db<0.05MbAnd/or Dr<0.05MrWhen value), it is to avoid because distribution of color is uniform and shadow
Ring analysis result；
The white reference point in textile image is determined as follows, if the preliminary pixel for being set to white reference point takes its bright
10% is used as final white reference point before angle value：
|Cb(i,j)-(Mb+Db×sign(Mb)) | 1.5 × D of ＜b
|Cr(i,j)-(Mr+Dr×sign(Mr)) | 1.5 × D of ＜r
Calculate each white reference point average brightness R of textile imageav, Gav, Bav, the gain R of each passagegain, Ggain, Bgain
With image in YCbCrY maximum Y under patternmax：
Calculate the final R ' G ' B ' of each passage of textile image：
R '=R × Rgain
G '=G × Ggain
B '=B × Bgain
S14 textile images are sent, the image sending module of smart mobile phone by the textile image obtained in step S13 send to
Server.
S2 servers receive the textile image that described smart mobile phone sends and are handled, analyzed, then will analysis knot
Fruit is sent to smart mobile phone, and the step is realized step by step through following：
S21 servers receiving module receives the fabric figure after pretreatment sent by smart mobile phone image sending module
Picture；
The textile image that the pretreatment module of S22 servers second is received to step carries out gray processing processing and detected whether
Containing fabric, gray processing is converted by below equation：
Gray (i, j)=0.11R (i, j)+0.59G (i, j)+0.3B (i, j)
S23 textile images processing, image processing module carries out threshold value change to the textile processes image after step S22 processing
Change, adaptive-filtering, Fast Fourier Transform (FFT), the processing such as rim detection, the cycle of textile processes positive and negative image is obtained respectively
Property characteristic point, specific method is as follows：
S231 selectes fabric two-face gray level image threshold value T, if the grey scale pixel value of certain in textile image is less than T, pixel ash
Angle value is set to 0, if certain grey scale pixel value is more than T, grey scale pixel value is set to 255, by textile image binaryzation (with reference to Fig. 4 a and
Fig. 4 b), the threshold transformation method of gradation of image is as follows：
S232 is adjusted according to the local variance of image and filtered by the noise in Adaptive noise cancellation fabric two-face image
Output, it is ensured that image border and detail of the high frequency.
The space area image of textile processes direct picture is converted to spectrogram by S233 by Fast Fourier Transform (FFT) (FFT)
(referring to Fig. 5)：
S234 carries out rim detection to the textile processes verso images after the processing of step S231, S232, obtains fabric
The edge graph (referring to Fig. 6) of technical back.
S24 textile images are analyzed, and server fabric analysis module is analyzed the image after step S23 processing, is drawn
Fabric analysis result, including art flower is wide, art flower is high, the horizontal close, finished product of finished product indulges close, institutional framework and passes circulation etc., tool
Body method is as follows:
Textile processes front frequency domain figure picture after the processing of server images processing module has point in the horizontal and vertical directions
The bright band of cloth rule, horizontal direction bright band represents fabric row information, and vertical direction bright band represents fabric stringer information, right respectively
Frequency domain figure picture is both horizontally and vertically projected；
The central point of selected level (vertical) direction projection image, along vertical (horizontal) direction point by point scanning gradation of image,
Until scanning is arrived the pixel when gray value is 255 as starting point, continue along vertical (horizontal) direction point by point scanning, until next
Secondary scanning stops when being 255 to gray value, and the pixel count m (n) of writing scan, adjacent two row of m (n) extremely fabrics is (vertical
Frequency between OK).
The image resolution ratio that smart mobile phone is shot is r (pixel/inch), and the textile image size of location cutting is M × N
(pixel), then the art flower of fabric is wide, art flower is high, finished product is horizontal close and finished product is indulged and close can be expressed as：
Width=n
Height=m
Textile processes verso images after the processing of server images processing module, to multiple mesh in the fabric edge figure of extraction
Mark region shape characteristic parameter is identified, and is judged pixel-by-pixel with element marking method, obtains the coil in Weaving Cycle
Underlap edge parameters, and then analyze, calculate the institutional framework of fabric and pass circulation.
S25 analysis results are fed back, and as a result feedback module sends the fabric analysis result drawn in step S24 to intelligent hand
In machine, and analysis result is shown to user.
Using the different knitted fabrics of above method continuous monitoring, contrasted with manual measurement, it is as a result as shown below：
Knitted fabric density measure error in length and breadth
As can be seen from the above table, using the knitted fabric technique rapid analysis method of the present invention, it must effectively can apply to
The analysis of a variety of different fabric constructions, and analysis result and the error of manual analysis are smaller (being respectively less than 6%), are ensureing accurate
On the premise of degree, analyze speed can must be effectively improved.
The preferred embodiment of the present invention is the foregoing is only, but the present invention is not limited to this, is familiar with the skill in the field
Scope of any change that art personnel are done to the embodiment of the present invention all without departing from claims of the present invention.Cause
This, as long as the improvement and conversion made on the basis of its general principles, are regarded as falling into the protection model of the present invention
In enclosing.
Claims (5)
1. a kind of knitted fabric industrial analysis method based on smart mobile phone, comprises the following steps：
S1 smart mobile phones gather the front-back two-sided image of knitted fabric and are sent to server, including：
S12 textile images are gathered：User makes fabric be presented in preview pane and ensure in preview pane extremely by adjusting camera site
Few textile image containing a complete cycle size；
S13 is pre-processed for the first time：Determined after the image collected in step S12 is amplified into 3-6 times by smart mobile phone
Position cuts and carries out the AWB processing of dynamic threshold, to obtain the textile image for meeting fabric analysis requirement；
S14 images are sent：The image sending module of smart mobile phone sends out the textile image in step S13 by pretreatment for the first time
Deliver to server；
S2 servers receive the textile image that described smart mobile phone sends and are handled, analyzed, and then send out analysis result
It is sent to smart mobile phone；
The step S2 includes：
S21 textile images are received：Server receive by smart mobile phone image sending module send through for the first time it is pretreated
Textile image；
Second of pretreatment of S22：The textile image that server is received to step S21 carries out gray processing processing；
The processing of S23 textile images：Threshold transformation, adaptive-filtering, quick Fu are carried out respectively to the image after step S22 gray processings
In leaf transformation, edge detection process, to obtain the periodic feature point of textile processes positive and negative image；
S24 textile images are analyzed：The periodic feature point of positive and negative image in step S23 is analyzed, fabric point is drawn
Result is analysed, the fabric analysis result is including art flower is wide, art flower is high, the horizontal close, finished product of finished product is indulged close, institutional framework and passed
Circulation；
S25 analysis results are fed back：The fabric analysis result is fed back in smart mobile phone；
The step S24 textile images analysis includes following sub-step：
Textile processes front frequency domain figure pictures of the S241 after S23 is handled is distributed the bright of rule in the horizontal and vertical directions
Band, horizontal direction bright band represents fabric row information, and vertical direction bright band represents fabric stringer information, frequency domain figure picture is entered respectively
Row is both horizontally and vertically projected；
The central point of S242 selected level direction projection images, vertically point by point scanning gradation of image, ash is arrived until scanning
Using the pixel as starting point when angle value is 255, continue vertically point by point scanning, be up to scanning gray value next time
Stop when 255, the pixel count m, the m of writing scan are the frequency between adjacent two row of fabric；Selected vertical direction
The central point of projected image, point by point scanning gradation of image, makees the pixel when scanning to gray value is 255 in the horizontal direction
For starting point, continue point by point scanning in the horizontal direction, until scanning next time stops when being 255 to gray value, writing scan
Pixel count n, the n are the frequency between adjacent two stringer of fabric；
The wide W of art flower of fabric described in S243idth, the high H of art flowereight, the horizontal close W of finished productPCClose C is indulged with finished productPCTable can be distinguished
It is shown as：
Width=n
Height=m
<mrow>
<msub>
<mi>W</mi>
<mrow>
<mi>P</mi>
<mi>C</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mi>r</mi>
<mo>&times;</mo>
<mi>n</mi>
</mrow>
<mrow>
<mn>2.54</mn>
<mi>M</mi>
</mrow>
</mfrac>
</mrow>
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>P</mi>
<mi>C</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mi>r</mi>
<mo>&times;</mo>
<mi>m</mi>
</mrow>
<mrow>
<mn>2.54</mn>
<mi>N</mi>
</mrow>
</mfrac>
</mrow>
Wherein r=pixel/inch, is the resolution ratio for the image that the smart mobile phone is shot, and M, N are the textile image of location cutting
Pixel size；
S244 knows for the verso images of fabric to multiple target area parameters for shape characteristic in the fabric edge figure of extraction
Not, judged pixel-by-pixel with element marking method, the coil underlap edge parameters in acquisition Weaving Cycle, and then analyze,
Calculate the institutional framework of fabric and pass circulation.
2. a kind of knitted fabric industrial analysis method based on smart mobile phone according to claim 1, it is characterised in that institute
Stating white balancing treatment method in step first time pretreatment S13 includes：
Textile image is YC by RGB patten transformations by S131bCr, N number of part is divided the image into after converting, calculates every respectively
The C of individual partbAnd CrAverage value Mb、MrAnd both mean square deviation Db、Dr, computational methods are as follows：
<mrow>
<msub>
<mi>D</mi>
<mi>b</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>b</mi>
</msub>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>M</mi>
<mi>b</mi>
</msub>
<mo>|</mo>
<mo>)</mo>
</mrow>
<mi>N</mi>
</mfrac>
</mrow>
<mrow>
<msub>
<mi>D</mi>
<mi>r</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<mo>|</mo>
<msub>
<mi>C</mi>
<mi>r</mi>
</msub>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>M</mi>
<mi>r</mi>
</msub>
<mo>|</mo>
<mo>)</mo>
</mrow>
<mi>N</mi>
</mfrac>
</mrow>
The average value M in each region of S132 statistics textile imagesb、MrAnd mean square deviation Db、DrIt is used as the value of whole image；
S133 determines the white reference point in textile image as follows,
|Cb(i,j)-(Mb+Db×sign(Mb)) | 1.5 × D of ＜b
|Cr(i,j)-(Mr+Dr×sign(Mr)) | 1.5 × D of ＜r
10% white reference point is used as final white reference point before brightness value；
S134 calculates each white reference point average brightness R of textile imageav, Gav, Bav, the gain R of each passagegain, Ggain, Bgain
With image in YCbCrY maximum Y under patternmax；
<mrow>
<msub>
<mi>R</mi>
<mrow>
<mi>g</mi>
<mi>a</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>Y</mi>
<mi>max</mi>
</msub>
<msub>
<mi>R</mi>
<mrow>
<mi>a</mi>
<mi>v</mi>
</mrow>
</msub>
</mfrac>
</mrow>
<mrow>
<msub>
<mi>G</mi>
<mrow>
<mi>g</mi>
<mi>a</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>Y</mi>
<mi>max</mi>
</msub>
<msub>
<mi>G</mi>
<mrow>
<mi>a</mi>
<mi>v</mi>
</mrow>
</msub>
</mfrac>
</mrow>
<mrow>
<msub>
<mi>B</mi>
<mrow>
<mi>g</mi>
<mi>a</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>Y</mi>
<mi>max</mi>
</msub>
<msub>
<mi>B</mi>
<mrow>
<mi>a</mi>
<mi>v</mi>
</mrow>
</msub>
</mfrac>
</mrow>
S135 calculates the final R ' G ' B ' of each passage of textile image：
R '=R × Rgain
G '=G × Ggain
B '=B × Bgain
3. a kind of knitted fabric industrial analysis method based on smart mobile phone according to claim 2, it is characterised in that step
When using the statistics in step S132 in rapid S133, cast out Db<0.05MbAnd/or Dr<0.05MrWhen value.
4. a kind of knitted fabric industrial analysis method based on smart mobile phone according to claim 1, it is characterised in that institute
The gray processing stated in step S22 converts completion by below equation：
Gray (i, j)=0.11R (i, j)+0.59G (i, j)+0.3B (i, j)
5. a kind of knitted fabric industrial analysis method based on smart mobile phone according to claim 4, it is characterised in that institute
Stating step S23 includes：
S231 threshold transformations：Using the method for threshold transformation by the further two-value of image of the fabric tow sides gray processing of extraction
Change, the threshold transformation method of gradation of image is as follows：
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>x</mi>
<mo><</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>255</mn>
</mtd>
<mtd>
<mrow>
<mi>x</mi>
<mo>&GreaterEqual;</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
S232 adaptive-filterings：By the noise in Adaptive noise cancellation fabric two-face image, according to the local variance of image
Adjustment filtering output, it is ensured that image border and detail of the high frequency.
S233 carries out Fast Fourier Transform (FFT) (FFT) by following formula and is converted to the space area image of textile processes direct picture frequently
Spectrogram：
<mrow>
<mi>F</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>&Integral;</mo>
<msubsup>
<mo>&Integral;</mo>
<mrow>
<mo>-</mo>
<mi>&infin;</mi>
</mrow>
<mi>&infin;</mi>
</msubsup>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mi>exp</mi>
<mrow>
<mo>(</mo>
<mo>-</mo>
<mi>j</mi>
<mn>2</mn>
<mi>&pi;</mi>
<mo>(</mo>
<mrow>
<mi>u</mi>
<mi>x</mi>
<mo>+</mo>
<mi>v</mi>
<mi>y</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>x</mi>
<mi>d</mi>
<mi>y</mi>
</mrow>
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>&Integral;</mo>
<msubsup>
<mo>&Integral;</mo>
<mrow>
<mo>-</mo>
<mi>&infin;</mi>
</mrow>
<mi>&infin;</mi>
</msubsup>
<mi>F</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mi>exp</mi>
<mrow>
<mo>(</mo>
<mo>-</mo>
<mi>j</mi>
<mn>2</mn>
<mi>&pi;</mi>
<mo>(</mo>
<mrow>
<mi>u</mi>
<mi>x</mi>
<mo>+</mo>
<mi>v</mi>
<mi>y</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>u</mi>
<mi>d</mi>
<mi>v</mi>
</mrow>
S234 carries out rim detection to the textile processes verso images after the processing of step S231, S232, obtains textile processes
The edge graph of reverse side.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610522846.0A CN106198542B (en) | 2016-07-05 | 2016-07-05 | A kind of knitted fabric industrial analysis method based on smart mobile phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610522846.0A CN106198542B (en) | 2016-07-05 | 2016-07-05 | A kind of knitted fabric industrial analysis method based on smart mobile phone |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106198542A CN106198542A (en) | 2016-12-07 |
| CN106198542B true CN106198542B (en) | 2017-10-20 |
Family
ID=57465306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610522846.0A Active CN106198542B (en) | 2016-07-05 | 2016-07-05 | A kind of knitted fabric industrial analysis method based on smart mobile phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106198542B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108729011A (en) * | 2018-05-16 | 2018-11-02 | 东华大学 | For automatic assembly line to the system for rapidly identifying and method of garment piece fabric positive and negative |
| CN108921819B (en) * | 2018-05-29 | 2021-02-02 | 黎明职业大学 | Cloth inspecting device and method based on machine vision |
| CN111325707B (en) * | 2018-12-13 | 2021-11-30 | 深圳中科飞测科技股份有限公司 | Image processing method and system, and detection method and system |
| CN109635758B (en) * | 2018-12-18 | 2021-07-09 | 武汉市蓝领英才科技有限公司 | Intelligent building site video-based safety belt wearing detection method for aerial work personnel |
| CN113205476A (en) * | 2020-08-04 | 2021-08-03 | 黄永芹 | Platform and method for intelligently identifying germ target |
| CN112734383B (en) * | 2021-01-12 | 2024-04-12 | 浙江理工大学 | Method for automatically identifying clothing technology |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149325A (en) * | 2007-10-09 | 2008-03-26 | 浙江理工大学 | Textile water-proof performance automatic detection method and detection system |
| US8259208B2 (en) * | 2008-04-15 | 2012-09-04 | Sony Corporation | Method and apparatus for performing touch-based adjustments within imaging devices |
| CN102523366B (en) * | 2011-12-13 | 2014-07-09 | 上海工程技术大学 | Automatic fabric weaving style analyzing system and method |
| CN102788793A (en) * | 2012-03-31 | 2012-11-21 | 江南大学 | Method for measuring density of weft knitted fabric based on spectrum analysis |
| CN103604937B (en) * | 2013-11-22 | 2015-06-10 | 青岛大学 | Fabric analysis system and method based on biaxial mechanical stretch processing |
| CN104730073A (en) * | 2015-02-16 | 2015-06-24 | 中国土产畜产进出口总公司 | Quantitative analysis method and system for dishes contained in plates |
| CN205352932U (en) * | 2015-12-14 | 2016-06-29 | 高安来 | Portable pesticide residue detecting system |
| CN105675601B (en) * | 2016-03-04 | 2019-05-31 | 江苏大学 | The system and detection method of a kind of field quick detection catechol and its derivative |
-
2016
- 2016-07-05 CN CN201610522846.0A patent/CN106198542B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106198542A (en) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106198542B (en) | A kind of knitted fabric industrial analysis method based on smart mobile phone | |
| CN102297867A (en) | Detection system for assembly quality of wiring harness | |
| CN201512707U (en) | Printing error detection device based on intelligent camera | |
| CN102523366B (en) | Automatic fabric weaving style analyzing system and method | |
| CN104021533B (en) | A kind of real time imaging noise-reduction method and device | |
| CN102790862A (en) | Method for dynamically improving effect of cell phone front camera | |
| CN102368821A (en) | Adaptive noise intensity video denoising method and system thereof | |
| CN106339657B (en) | Crop straw burning monitoring method based on monitor video, device | |
| CN107426894A (en) | Classroom lighting automatic detection and control system and its method based on intelligent video identification | |
| CN107403180A (en) | A kind of numeric type equipment detection recognition method and system | |
| DE102009029391A1 (en) | Image processing apparatus and image processing method | |
| CN113655002B (en) | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology | |
| CN101668226B (en) | Method for acquiring color image with best quality | |
| CN108154490A (en) | Based on the high-voltage transmission line insulator image enchancing method for improving fuzzy set theory | |
| CN212442146U (en) | Household garbage recognition system | |
| CN110288623A (en) | The data compression method of unmanned plane marine cage culture inspection image | |
| CN109594319A (en) | A kind of pck count intelligent detection device and method | |
| CN103458158B (en) | The control method of time of exposure of line-scan digital camera and system thereof | |
| CN109727198A (en) | A kind of brightness of image optimization method based on Face Detection | |
| CN106093051A (en) | Paper roll tangent plane burr detection method based on machine vision and device | |
| CN104048966A (en) | Big-law-based cloth cover defect detection and classification method | |
| CN107504905A (en) | A kind of reticulated mesh size deviation ratio detecting system based on machine vision | |
| CN111192332B (en) | Smoke detection-based smoke machine control method and smoke machine | |
| CN107564041A (en) | A kind of detection method of visible images moving air target | |
| CN104458759B (en) | A kind of irregular automatic cloth inspecting machine from correction system of gray scale |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |