CN114789927A - Artificial intelligent control method and system for textile fabric gray cloth winding machine - Google Patents

Artificial intelligent control method and system for textile fabric gray cloth winding machine Download PDF

Info

Publication number
CN114789927A
CN114789927A CN202210697774.9A CN202210697774A CN114789927A CN 114789927 A CN114789927 A CN 114789927A CN 202210697774 A CN202210697774 A CN 202210697774A CN 114789927 A CN114789927 A CN 114789927A
Authority
CN
China
Prior art keywords
area
concave
convex
fabric
gray
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
Application number
CN202210697774.9A
Other languages
Chinese (zh)
Other versions
CN114789927B (en
Inventor
秦伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nantong Hengzhen Textile Technology Co ltd
Original Assignee
Nantong Hengzhen Textile Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nantong Hengzhen Textile Technology Co ltd filed Critical Nantong Hengzhen Textile Technology Co ltd
Priority to CN202210697774.9A priority Critical patent/CN114789927B/en
Publication of CN114789927A publication Critical patent/CN114789927A/en
Application granted granted Critical
Publication of CN114789927B publication Critical patent/CN114789927B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65H—HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES
    • B65H18/00—Winding webs
    • B65H18/08—Web-winding mechanisms
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65H—HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES
    • B65H26/00—Warning or safety devices, e.g. automatic fault detectors, stop-motions, for web-advancing mechanisms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/0014—Image feed-back for automatic industrial control, e.g. robot with camera
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/64—Analysis of geometric attributes of convexity or concavity
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65H—HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES
    • B65H2701/00—Handled material; Storage means
    • B65H2701/10—Handled articles or webs
    • B65H2701/17—Nature of material
    • B65H2701/174—Textile, fibre
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30124—Fabrics; Textile; Paper

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Geometry (AREA)
  • Quality & Reliability (AREA)
  • Robotics (AREA)
  • Treatment Of Fiber Materials (AREA)

Abstract

The invention relates to the technical field of material testing and analysis, in particular to an artificial intelligent control method and system for a textile fabric gray fabric winding machine. By utilizing the material analysis and test means, the invention can accurately determine the flatness of the fabric, and finally realizes the reliable adjustment of the extension tension of the fabric.

Description

Artificial intelligent control method and system for textile fabric gray cloth winding machine
Technical Field
The invention relates to the technical field of material testing and analysis, in particular to an artificial intelligent control method and system for a textile fabric gray cloth winding machine.
Background
Textile production is an important industry in development and construction of China, textile fabric gray cloth is used for manufacturing molded clothes, and in a production process program of the textile fabric gray cloth, the textile fabric gray cloth needs to be rolled by a fabric rolling machine to be bundled, so that the situation that the fabric gray cloth is stacked disorderly is avoided, and the situation that the fabric is attached with dirt, folded and the like is avoided.
Because the weaving surface fabric embryo cloth is many by the fibrous material composition, in the rolling operation in-process, but crease-resistant rolling machine need be earlier to the surface fabric kunshan, there is certain power of dragging to the surface fabric, when because roll core rolling speed is unchangeable, when extension tension is not enough, the position of surface fabric than the fold need consume the longer time just can be carried to rolling up the core under the conveying of compaction roller, can be because of the compaction roller is too late to carry and piles up to some extent, the extension tension has made the condition that the fibre degree of dragging is too big and the batting appears in some surface fabric positions easily. The existing method for adjusting the extension tension of the fabric requires manual adjustment according to experience, and has the disadvantages of strong subjectivity and poor adjustment reliability.
Disclosure of Invention
The invention aims to provide an artificial intelligent control method and system for a textile fabric gray cloth rolling machine, which are used for solving the problem that the existing fabric extension tension is unreliable to adjust.
In order to solve the technical problem, the invention provides an artificial intelligent control method of a textile fabric gray cloth winding machine, which comprises the following steps of:
acquiring a fabric operation image of a winding machine, acquiring a fabric gray fabric image according to the fabric operation image, and further acquiring a fabric gray fabric image;
performing wrinkle identification according to the gray level image of the fabric gray fabric to obtain each convex area and each concave area, and further determining the area of each convex area and each concave area;
determining the overall concave-convex degree of the fabric according to the gray level image of the fabric gray fabric, the gray level values of all pixel points in all the convex areas and the concave areas and the areas of all the convex areas and all the concave areas;
performing skeletonization treatment on each convex area and each concave area to obtain ridge lines of each convex area and each concave area;
determining the tortuosity of the ridge line of each convex area and each concave area according to the pixel value of each pixel point on the ridge line of each convex area and each concave area;
determining the slope line of the ridge line of each convex area and each concave area according to the ridge line of each convex area and each concave area and the pixel value of each pixel point on the ridge line;
determining the kurtosis of the ridge line of each convex area and each concave area according to the gray value of each pixel point on the slope line of the ridge line of each convex area and each concave area;
determining the sharpness of each convex area and each concave area according to the tortuosity and the kurtosis of ridge lines of each convex area and each concave area, and further determining the comprehensive sharpness of the fabric;
the flatness degree of the fabric is determined according to the overall concave-convex degree and the comprehensive sharp degree of the fabric, and the extension tension of the winding machine to the fabric is controlled according to the flatness degree of the fabric.
Further, the performing wrinkle identification to obtain each of the convex area and the concave area includes:
performing edge detection on the gray level image of the fabric gray fabric to obtain each convex area and each initial concave area;
determining a gray variance value corresponding to each initial concave area according to the gray value of each pixel point in each initial concave area;
and screening each initial concave area according to the gray variance value corresponding to each initial concave area, thereby obtaining each concave area.
Further, the determining of the overall concave-convex degree of the fabric comprises:
determining a flat area on the gray level image of the fabric gray level cloth according to the gray level image of the fabric gray level cloth, each convex area and each initial concave area;
determining a gray mean value of the flat area according to the gray values of all pixel points in the flat area on the gray image of the fabric gray grey cloth;
calculating a sum of gray values of all pixel points in all convex regions minus a gray average of the flat region, thereby obtaining the height of each convex region;
calculating a sum of gray values of all pixel points in all concave areas minus a gray average of the flat areas so as to obtain depths of all concave areas;
and calculating the overall concave-convex degree of the fabric according to the height of each convex region, the depth of each concave region and the area of each convex region and each concave region.
Further, a calculation formula corresponding to the overall concave-convex degree of the fabric is as follows:
Figure 436642DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 822624DEST_PATH_IMAGE002
the degree of the overall concave-convex of the fabric,
Figure 71203DEST_PATH_IMAGE003
is a firstiThe area of each of the convex regions is,
Figure 199169DEST_PATH_IMAGE004
is a firstiThe height of each of the convex regions is,
Figure 482383DEST_PATH_IMAGE005
is the area of the jth concave region,
Figure 773687DEST_PATH_IMAGE006
is the depth of the jth concave region,
Figure 509562DEST_PATH_IMAGE007
is the total number of the convex regions,
Figure 426571DEST_PATH_IMAGE008
the total number of concave regions.
Further, a calculation formula for determining the degree of curvature of the ridge line of each convex region and each concave region is as follows:
Figure 829871DEST_PATH_IMAGE009
wherein, the first and the second end of the pipe are connected with each other,
Figure 26497DEST_PATH_IMAGE010
the degree of meandering of the ridge line of each convex region or concave region,
Figure 249668DEST_PATH_IMAGE011
on the ridge line of each convex or concave regioniThe gray value of +1 pixel point,
Figure 721101DEST_PATH_IMAGE012
on the ridge line of each convex or concave regioniThe gray value of each pixel point, m, is the total number of pixel points on the ridge line of each convex region or concave region.
Further, the step of determining a slope line of the ridge line of each of the convex region and the concave region includes:
determining a maximum pixel value pixel point on the ridge line of each convex area and a minimum pixel value pixel point on the ridge line of each concave area according to the pixel values of the pixel points on the ridge lines of each convex area and each concave area;
and determining the perpendicular line of the tangent line of the ridge line of each convex region at the maximum pixel value pixel point corresponding to the ridge line of each convex region and the perpendicular line of the tangent line of the ridge line of each concave region at the minimum pixel value pixel point corresponding to the ridge line of each concave region according to the ridge line of each convex region and each concave region, the maximum pixel value pixel point on the ridge line of each convex region and the minimum pixel value pixel point on the ridge line of each concave region, so as to obtain the slope line of the ridge line of each convex region and each concave region.
Further, the determining the kurtosis of the ridge line of each of the convex region and the concave region includes:
determining the mean value of pixel values on the slope lines of the ridge lines of the convex areas and the concave areas according to the pixel values of the pixel points on the slope lines of the ridge lines of the convex areas and the concave areas;
calculating the difference value between the pixel value of each pixel point on the slope line of the ridge line of each convex area and each concave area and the mean value of the pixel values on the corresponding slope line, thereby obtaining the fourth-order central moment and the second-order central moment of the ridge line of each convex area and each concave area;
and calculating the ratio of the fourth-order central moment of the ridge line of each convex area and each concave area to the square of the corresponding second-order central moment, thereby obtaining the kurtosis of the ridge line of each convex area and each concave area.
Further, the corresponding calculation formula of the kurtosis of the ridge line of each convex area and each concave area is as follows:
Figure 962595DEST_PATH_IMAGE013
wherein, the first and the second end of the pipe are connected with each other,
Figure 861281DEST_PATH_IMAGE014
the kurtosis of the ridge line of each convex region or concave region,
Figure 571748DEST_PATH_IMAGE015
the pixel value of the ith pixel point on the slope line of the ridge line of each convex region or each concave region,
Figure 315713DEST_PATH_IMAGE016
the average value of the pixel values on the slope line of the ridge line of each convex region or each concave region is represented by t, and the t is the total number of the pixel points on the slope line of the ridge line of each convex region or each concave region.
Further, a calculation formula corresponding to the determined flatness degree of the fabric is as follows:
Figure 942872DEST_PATH_IMAGE017
wherein, the first and the second end of the pipe are connected with each other,
Figure 746880DEST_PATH_IMAGE018
in order to ensure the smoothness of the fabric,
Figure 944643DEST_PATH_IMAGE002
is the overall concave-convex degree of the fabric,
Figure 492299DEST_PATH_IMAGE019
is made of flourThe comprehensive sharpness of the materials.
The invention also provides an artificial intelligence control system of the textile fabric gray cloth winding machine, which comprises a processor and a memory, wherein the processor is used for processing the instructions stored in the memory so as to realize the artificial intelligence control method of the textile fabric gray cloth winding machine.
The invention has the following beneficial effects: the method comprises the steps of obtaining a fabric gray image through a fabric operation image of a winding machine, carrying out material analysis and test on the fabric gray image, determining each convex area and each concave area, further determining the overall concave-convex degree of the fabric, determining the ridge line of each convex area and each concave area and the slope line of the ridge line, further determining the tortuosity and the kurtosis of the ridge line of each convex area and each concave area, combining the tortuosity and the kurtosis of the ridge line of each convex area and each concave area, determining the comprehensive sharp degree of the fabric, combining the overall concave-convex degree and the comprehensive sharp degree of the fabric, further determining the flatness degree of the fabric, and finally realizing the control of the extension tension of the winding machine to the fabric. According to the invention, by acquiring the fabric operation image of the winding machine and utilizing material analysis and test means, the flatness of the fabric can be accurately determined, and finally the reliable adjustment of the extension tension of the fabric is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of an artificial intelligence control method of a textile fabric gray cloth winding machine of the invention;
fig. 2 is a schematic view of a fabric fold in an embodiment of the invention.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of the embodiments, structures, features and effects of the technical solutions according to the present invention will be given with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
Tensile regulation of extension of textile fabric crease-resistance rolling machine influences surface fabric gray cloth quality and operating efficiency, and this embodiment utilizes computer vision technique, through handling the surface fabric image of shooting, according to the characteristic analysis of image, calculates the roughness of surface fabric, and then obtains the required best extension tension of kun exhibition surface fabric, adjusts the extension power of rolling machine, makes the surface fabric kun exhibition to optimum, avoids the surface fabric to pile up or tear the surface fabric.
Specifically, the embodiment provides an artificial intelligence control method for a textile fabric gray cloth winding machine, and a corresponding flowchart is shown in fig. 1, and the method includes the following steps:
acquiring a fabric operation image of a winding machine, acquiring a fabric gray fabric image according to the fabric operation image, and further acquiring a fabric gray fabric image;
performing wrinkle identification according to the gray level image of the fabric gray fabric to obtain each convex area and each concave area, and further determining the area of each convex area and each concave area;
determining the overall concave-convex degree of the fabric according to the gray level image of the fabric gray fabric, the gray level values of all pixel points in all convex areas and concave areas and the areas of all convex areas and concave areas;
performing skeletonization treatment on each convex area and each concave area to obtain ridge lines of each convex area and each concave area;
determining the tortuosity of the ridge line of each convex area and each concave area according to the pixel value of each pixel point on the ridge line of each convex area and each concave area;
determining slope lines of the ridge lines of the convex areas and the concave areas according to the ridge lines of the convex areas and the concave areas and pixel values of pixel points on the ridge lines;
determining the kurtosis of the ridge line of each convex area and each concave area according to the gray value of each pixel point on the slope line of the ridge line of each convex area and each concave area;
determining the sharpness of each convex area and each concave area according to the tortuosity and the kurtosis of ridge lines of each convex area and each concave area, and further determining the comprehensive sharpness of the fabric;
and determining the flatness of the fabric according to the overall concave-convex degree and the comprehensive sharp degree of the fabric, and controlling the extension tension of the fabric of the winding machine according to the flatness of the fabric.
The above-mentioned artificial intelligence control method for the textile fabric gray cloth winding machine is described in detail below with reference to specific embodiments.
The method comprises the following steps: and acquiring an image shot by a camera above the winding machine, and performing semantic segmentation to identify the fabric image.
This embodiment needs clear surface fabric surface image, according to surface fabric surface image characteristic, calculates the roughness of surface fabric, adjusts the extension power of rolling machine. All the surface images of the fabric on the winding machine need to be collected, and the characteristic information of the fabric surface in the images is identified.
In order to clearly display wrinkles on the surface of the fabric, a single-side LED lamp is used as a light source for illumination, light and shade changes can be formed at the wrinkles due to the existence of the wrinkles, the wrinkles are sharper, and the light and shade changes are more severe, so that the wrinkle characteristics of images can be described based on the phenomenon, and the wrinkles can be used for distinguishing the fabrics with different flatness.
The present embodiment adopts a DNN semantic segmentation manner to identify the target in the segmented image, and the relevant content of the DNN network is as follows:
a. the used data set is a fabric image data set on the winding machine acquired in a overlooking mode.
b. The pixels to be segmented are divided into 2 types, namely the labeling process of the training set corresponding to the labels is as follows: in the semantic label of the single channel, the label of the pixel at the corresponding position belonging to the background class is 0, and the label of the pixel belonging to the fabric is 1.
c. The task of the network is classification, so the loss function used is a cross entropy loss function.
Therefore, the processing of the fabric image on the winding machine is realized through the DNN, and the connected domain information of the fabric surface in the image is obtained.
Step two: and judging the flatness of the fabric according to the characteristic analysis of the fabric image.
And calculating a convex area and a concave area of each wrinkle according to the difference between the flat area and the wrinkle area of the surface of the fabric, analyzing the change characteristics of the gray value in each wrinkle area, and calculating the severity of each single wrinkle. And comprehensively obtaining the flatness of the current fabric.
The process for obtaining the fabric flatness in the embodiment comprises the following steps: a) The masked area in the folds of the fabric was calculated according to the Canny edge test.
b) And distinguishing a real wrinkle concave area according to the characteristics of the wrinkle concave area of the fabric.
c) And calculating the flatness of the fabric according to the gray level change of each wrinkle area.
The following are specific developments:
a) the masked area in the folds of the fabric was calculated according to the Canny edge test.
When the fabric is wrinkled, the gray level of the presented image is unstable. The gray values of the areas where the wrinkles are recessed are small, and the gray values of the areas where the wrinkles are protruding are large, as shown in fig. 2.
Therefore, the surface image of the fabric is subjected to graying treatment, then noise and miscellaneous points in the image are removed by using median filtering, and then convex areas and concave areas of folds are identified and segmented by using Canny edge detection, wherein the number of the convex areas and the number of the concave areas are respectively
Figure 711316DEST_PATH_IMAGE007
And
Figure 951804DEST_PATH_IMAGE020
。
calculating the area of each convex region of the fold, i.e.Obtaining a convex surface product set by the number of pixel points
Figure 371284DEST_PATH_IMAGE021
Then calculating the area of each fold concave area to obtain a concave area set
Figure 988210DEST_PATH_IMAGE022
. And finally, calculating gray values of all pixel points in the flat area on the fabric, summing the gray values, and taking the average value as R.
Due to the use of a single-sided light source for illumination, the convex area of the fold can generate shadows, which can be in the flat area of the fabric. When the segmentation is identified, the segmentation may be judged as a wrinkle concave area, which affects the calculation of the flatness of the fabric, so that the wrinkle concave area needs to be further analyzed.
b) And distinguishing a real wrinkle concave area according to the characteristics of the wrinkle concave area of the fabric.
When the flat area of the fabric is influenced by the shadow, the numerical difference of the gray value of each pixel point in the area is not obvious, and the gray value of each pixel point in the fold concave area gradually becomes smaller along with the downward depth of the concave pit. Therefore, the shaded flat area and the corrugated concave area can be analyzed according to the change of the gray value in the corrugated concave area.
Firstly, gray values of all pixel points in a fold concave area are calculated to obtain a set
Figure 809536DEST_PATH_IMAGE023
And n is the number of pixel points in the region. Recalculating sets
Figure 470193DEST_PATH_IMAGE024
Thereby obtaining a set of
Figure 376969DEST_PATH_IMAGE024
Standard deviation of (2)
Figure 266428DEST_PATH_IMAGE025
The calculation formula is as follows:
Figure 207839DEST_PATH_IMAGE026
wherein, aggregate
Figure 304977DEST_PATH_IMAGE024
Standard deviation of (2)
Figure 699049DEST_PATH_IMAGE025
The smaller the value of (b), the smaller the change in the gray level value in the description area, and the flatter the fabric.
Taking a plurality of flat areas on the current fabric, and calculating the standard deviation of each position
Figure 392199DEST_PATH_IMAGE025
Obtaining a set, and calculating the mean value of the set
Figure 922537DEST_PATH_IMAGE027
This is used as a threshold. When here the concave region is wrinkled
Figure 924997DEST_PATH_IMAGE025
Value less than
Figure 806365DEST_PATH_IMAGE027
Then, this region is a flat region. When the concave area is wrinkled
Figure 303206DEST_PATH_IMAGE025
Is greater than
Figure 219209DEST_PATH_IMAGE027
This region is a true recessed region. From this, the true fold concave area has
Figure 877724DEST_PATH_IMAGE008
A is prepared from
Figure 498585DEST_PATH_IMAGE028
. I.e. the area set of the fold valley regions is
Figure 533537DEST_PATH_IMAGE029
。
c) And calculating the flatness of the fabric according to the gray level change of each wrinkle area.
In conclusion, all the wrinkle convex areas and concave areas on the fabric are obtained, and the influence on the flatness is different due to different shape characteristics of the concave-convex areas. The effect of each fold on the overall flatness needs to be analyzed.
Firstly, a convex region is taken, the sum of the gray value of each pixel point in the region minus R is calculated, and a set is obtained
Figure 304047DEST_PATH_IMAGE030
Wherein
Figure 133463DEST_PATH_IMAGE007
The data in the set may represent the height of each point protrusion for the number of raised areas of the current fabric. Then a concave area is taken, the sum value of the gray value of each pixel point in the area after being subtracted by R is calculated, and a set is obtained
Figure 238691DEST_PATH_IMAGE031
In which
Figure 342913DEST_PATH_IMAGE008
The data in the set may represent the depth of each point depression, which is the number of concave regions on the current fabric.
So far, the overall concave-convex degree F of the current fabric can be obtained:
Figure 436771DEST_PATH_IMAGE001
the former term is the sum of products of the areas of the corrugated convex regions and the convex heights of the corrugated convex regions, the latter term is the sum of products of the areas of the corrugated concave regions and the concave depths of the corrugated concave regions, and the sum of the areas and the concave depths represents the overall concave-convex degree F of the current fabric.
And then calculating the sharpness of the whole wrinkle according to the height change of the convex area and the depth change of the concave area. The process of refining the wrinkle region, i.e. reducing the lines of the image from a multi-pixel width to a unit pixel width, is also called skeletonization. Thereby obtaining the peak ridge lines of the corrugated convex regions and the valley ridge lines of the concave regions.
Then analyzing the gray value change of the thinned ridge line, and firstly calculating a gray value set from one end to the other end of the ridge line of a fold area
Figure 92881DEST_PATH_IMAGE032
And m represents the length of the ridge line, namely the number of pixel points of the ridge line.
Then in the collection
Figure 419826DEST_PATH_IMAGE012
The former value is used to subtract the latter value to obtain the ridge fluctuation, which is summed up to take the mean value G. The degree of tortuosity of the ridge line is represented by G, and the calculation formula is as follows:
Figure 62160DEST_PATH_IMAGE009
wherein the content of the first and second substances,
Figure 541682DEST_PATH_IMAGE033
and the gray scale difference from one point to the next point on the ridge line is represented, m is the number of pixel points on the ridge line, and G is the average value of the gray scale difference on the ridge line and represents the tortuosity of the ridge line. The larger the value of G, the more tortuous the peak or valley-ridge lines of the wrinkles, and the greater the degree of wrinkles.
And if the ridge line is the corrugated convex area, taking the maximum gray value point on the ridge line. If the ridge line is the fold concave area, the minimum gray value point on the ridge line is taken. And a straight line perpendicular to the tangent line of the ridge line at the point is made in the fold area, and the straight line can represent slopes on the left and right sides of the highest part of the ridge line of the peak of the fold or the lowest part of the ridge line of the valley. Counting the gray value set from one end to the other end of the straight line
Figure 978480DEST_PATH_IMAGE034
WhereintIndicating the length of the two ramps.
It is known that kurtosis is the degree of steepness of the distribution of the measurement data, and the distribution map is more steep when the kurtosis value is larger, and the distribution map is more squash when the kurtosis value is smaller. Thus can be assembled by calculation
Figure 809033DEST_PATH_IMAGE015
The steepness of the slope is judged according to the kurtosis H of the slope. First computing a set of distances
Figure 976096DEST_PATH_IMAGE015
Average value of (2)
Figure 310126DEST_PATH_IMAGE016
Obtaining a set
Figure 917825DEST_PATH_IMAGE015
Central moment of
Figure 219362DEST_PATH_IMAGE035
The calculation formula is as follows:
Figure 203498DEST_PATH_IMAGE036
the kurtosis is the ratio of the fourth-order central moment to the square of the second-order central moment, and the calculation formula is as follows:
Figure 392034DEST_PATH_IMAGE037
wherein the content of the first and second substances,
Figure 436214DEST_PATH_IMAGE038
representation collection
Figure 975779DEST_PATH_IMAGE015
The fourth-order central moment of (a) is,
Figure 747295DEST_PATH_IMAGE039
representation collection
Figure 55917DEST_PATH_IMAGE015
The second-order central moment of (a) of (b),
Figure 5418DEST_PATH_IMAGE015
representation collection
Figure 766701DEST_PATH_IMAGE015
The value of the (i) th value,
Figure 341908DEST_PATH_IMAGE016
is a set
Figure 505036DEST_PATH_IMAGE015
Average value of (a). The larger the value of the kurtosis H is, the steeper the slopes of the left and right sides of the highest position of the crest line or the lowest position of the valley line of the corrugation are.
Thus, the sharpness degree R of the whole wrinkle is obtained, and the calculation formula is as follows:
Figure 625438DEST_PATH_IMAGE040
g is the tortuosity of a crest line or a trough line of the fold, H is the steepness of slopes on the left side and the right side of the highest position of the crest line or the lowest position of the trough line of the fold, and the sum R of the two represents the integral sharpness of the single fold.
Calculating the overall sharpness of each fold to obtain a set
Figure 405176DEST_PATH_IMAGE041
Wherein
Figure 244126DEST_PATH_IMAGE042
All wrinkles are indicated. Summing the two, and calculating the comprehensive sharpness degree Q of the current fabric as follows:
Figure 527339DEST_PATH_IMAGE043
to sum up, the obtained flatness W of the current fabric is as follows:
Figure 84223DEST_PATH_IMAGE017
wherein, F represents the overall concave-convex degree of the current fabric, Q represents the comprehensive sharp degree of the current fabric, the larger the sum of the F and the Q is, the more and more serious the wrinkle in the fabric is, and therefore the inverse ratio W is used for representing the flatness degree of the current fabric.
Step three: and adjusting the extension tension of the winding machine to the fabric according to the flatness of the fabric to enable the fabric to be unfolded to the optimal state.
Obtaining the flatness W of the current fabric according to the step two, and automatically setting a threshold value according to different requirements of an implementer
Figure 820098DEST_PATH_IMAGE044
When the value of the flatness degree W is not more than the threshold value
Figure 471528DEST_PATH_IMAGE044
In the process, the extension tension of the winding machine on the fabric is increased until the value of the flatness degree W is greater than the threshold value
Figure 343669DEST_PATH_IMAGE044
And the winding machine keeps the extension tension at the moment.
The embodiment also provides an artificial intelligence control system of the textile fabric gray cloth winding machine, which comprises a processor and a memory, wherein the processor is used for processing the instructions stored in the memory so as to realize the artificial intelligence control method of the textile fabric gray cloth winding machine. Because the artificial intelligence control method of the textile fabric gray cloth winding machine is described in detail in the above content, the description is omitted here.
It should be noted that: the above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.

Claims (10)

1. An artificial intelligence control method for a textile fabric gray cloth winding machine is characterized by comprising the following steps:
acquiring a fabric operation image of a winding machine, acquiring a fabric gray fabric image according to the fabric operation image, and further acquiring a fabric gray image;
performing wrinkle identification according to the gray level image of the fabric gray fabric to obtain each convex area and each concave area, and further determining the area of each convex area and each concave area;
determining the overall concave-convex degree of the fabric according to the gray level image of the fabric gray fabric, the gray level values of all pixel points in all convex areas and concave areas and the areas of all convex areas and concave areas;
performing skeletonization treatment on each convex area and each concave area to obtain ridge lines of each convex area and each concave area;
determining the tortuosity of the ridge line of each convex area and each concave area according to the pixel value of each pixel point on the ridge line of each convex area and each concave area;
determining slope lines of the ridge lines of the convex areas and the concave areas according to the ridge lines of the convex areas and the concave areas and pixel values of pixel points on the ridge lines;
determining the kurtosis of the ridge line of each convex area and each concave area according to the gray value of each pixel point on the slope line of the ridge line of each convex area and each concave area;
determining the sharpness of each convex area and each concave area according to the tortuosity and the kurtosis of the ridge line of each convex area and each concave area, and further determining the comprehensive sharpness of the fabric;
and determining the flatness of the fabric according to the overall concave-convex degree and the comprehensive sharp degree of the fabric, and controlling the extension tension of the fabric of the winding machine according to the flatness of the fabric.
2. The artificial intelligence control method of the textile fabric greige cloth rolling machine according to claim 1, wherein the step of performing wrinkle identification to obtain each convex area and each concave area comprises the following steps:
performing edge detection on the gray level image of the fabric gray cloth to obtain each convex area and each initial concave area;
determining a gray variance value corresponding to each initial concave area according to the gray value of each pixel point in each initial concave area;
and screening each initial concave area according to the gray variance value corresponding to each initial concave area, thereby obtaining each concave area.
3. The artificial intelligence control method for the textile fabric gray cloth winding machine according to claim 2, wherein the determining of the overall concave-convex degree of the fabric comprises:
determining a flat area on the gray level image of the fabric gray level cloth according to the gray level image of the fabric gray level cloth, each convex area and each initial concave area;
determining a gray mean value of the flat area according to the gray values of all pixel points in the flat area on the gray image of the fabric gray grey cloth;
calculating a sum of gray values of all pixel points in each convex area minus a gray average value of the flat area, thereby obtaining the height of each convex area;
calculating a sum of gray values of all pixel points in all concave areas minus a gray average of the flat areas so as to obtain depths of all concave areas;
and calculating the overall concave-convex degree of the fabric according to the height of each convex region, the depth of each concave region and the area of each convex region and each concave region.
4. The artificial intelligence control method for the textile fabric gray cloth winding machine according to claim 3, characterized in that the calculation formula corresponding to the overall concave-convex degree of the fabric is:
Figure 70493DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 11905DEST_PATH_IMAGE002
the degree of the overall concave-convex of the fabric,
Figure 577884DEST_PATH_IMAGE003
is as followsiThe area of each of the convex regions is,
Figure 706377DEST_PATH_IMAGE004
is a firstiThe height of each of the convex regions is,
Figure 399527DEST_PATH_IMAGE005
is the area of the jth concave region,
Figure 444712DEST_PATH_IMAGE006
is the depth of the jth concave region,
Figure 666746DEST_PATH_IMAGE007
is the total number of the convex regions,
Figure 282535DEST_PATH_IMAGE008
the total number of concave regions.
5. The artificial intelligence control method of the textile fabric gray cloth winding machine according to claim 1, characterized in that the calculation formula for determining the corresponding degree of the tortuosity of the ridge line of each convex area and each concave area is as follows:
Figure 231905DEST_PATH_IMAGE009
wherein the content of the first and second substances,
Figure 616750DEST_PATH_IMAGE010
the degree of meandering of the ridge line of each convex region or concave region,
Figure 515743DEST_PATH_IMAGE011
on the ridgeline of each convex or concave regioniThe gray value of +1 pixel point,
Figure 884408DEST_PATH_IMAGE012
on the ridgeline of each convex or concave regioniThe gray value of each pixel point, m, is the total number of pixel points on the ridge line of each convex region or concave region.
6. The artificial intelligence control method for the textile fabric blank rolling machine according to claim 1, wherein the step of determining the slope line of the ridge line of each of the convex area and the concave area comprises:
determining a maximum pixel value pixel point on the ridge line of each convex region and a minimum pixel value pixel point on the ridge line of each concave region according to the pixel values of the pixel points on the ridge lines of each convex region and each concave region;
and determining the perpendicular line of the tangent line of the ridge line of each convex area at the maximum pixel value pixel point corresponding to the ridge line of each convex area and the perpendicular line of the tangent line of the ridge line of each concave area at the minimum pixel value pixel point corresponding to the ridge line of each concave area according to the ridge line of each convex area and the concave area, the maximum pixel value pixel point on the ridge line of each convex area and the minimum pixel value pixel point on the ridge line of each concave area, and thus determining the slope line of the ridge line of each convex area and each concave area.
7. The artificial intelligence control method for the textile fabric blank rolling machine according to claim 6, wherein the determining the kurtosis of the ridge line of each of the convex area and the concave area comprises:
determining the mean value of pixel values on the slope lines of the ridge lines of the convex areas and the concave areas according to the pixel values of the pixel points on the slope lines of the ridge lines of the convex areas and the concave areas;
calculating the difference value between the pixel value of each pixel point on the slope line of the ridge line of each convex area and each concave area and the mean value of the pixel values on the corresponding slope line, thereby obtaining the fourth-order central moment and the second-order central moment of the ridge line of each convex area and each concave area;
and calculating the ratio of the fourth-order central moment of the ridge line of each convex area and each concave area to the square of the corresponding second-order central moment, thereby obtaining the kurtosis of the ridge line of each convex area and each concave area.
8. The artificial intelligence control method of the textile fabric gray cloth winding machine according to claim 7, characterized in that the corresponding calculation formula of the kurtosis of the ridge line of each convex area and each concave area is as follows:
Figure 919360DEST_PATH_IMAGE013
wherein, the first and the second end of the pipe are connected with each other,
Figure 689870DEST_PATH_IMAGE014
the kurtosis of the ridge line of each convex or concave region,
Figure 34132DEST_PATH_IMAGE015
the pixel value of the ith pixel point on the slope line of the ridge line of each convex region or each concave region,
Figure 890093DEST_PATH_IMAGE016
the value is the mean value of pixel values on the slope line of the ridge line of each convex region or each concave region, and t is the total number of pixel points on the slope line of the ridge line of each convex region or each concave region.
9. The artificial intelligence control method of the textile fabric gray cloth winding machine according to claim 1, characterized in that,
the calculation formula for determining the flatness of the fabric is as follows:
Figure 463156DEST_PATH_IMAGE017
wherein, the first and the second end of the pipe are connected with each other,
Figure 88173DEST_PATH_IMAGE018
the smoothness of the fabric is the degree of smoothness of the fabric,
Figure 603337DEST_PATH_IMAGE002
the degree of the overall concave-convex of the fabric,
Figure 681014DEST_PATH_IMAGE019
is the comprehensive sharpness of the fabric.
10. An artificial intelligence control system for a textile fabric greige cloth winding machine, which is characterized by comprising a processor and a memory, wherein the processor is used for processing instructions stored in the memory to realize the artificial intelligence control method for the textile fabric greige cloth winding machine according to any one of claims 1-9.
CN202210697774.9A 2022-06-20 2022-06-20 Artificial intelligent control method and system for textile fabric gray cloth winding machine Active CN114789927B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210697774.9A CN114789927B (en) 2022-06-20 2022-06-20 Artificial intelligent control method and system for textile fabric gray cloth winding machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210697774.9A CN114789927B (en) 2022-06-20 2022-06-20 Artificial intelligent control method and system for textile fabric gray cloth winding machine

Publications (2)

Publication Number Publication Date
CN114789927A true CN114789927A (en) 2022-07-26
CN114789927B CN114789927B (en) 2022-09-23

Family

ID=82463569

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210697774.9A Active CN114789927B (en) 2022-06-20 2022-06-20 Artificial intelligent control method and system for textile fabric gray cloth winding machine

Country Status (1)

Country Link
CN (1) CN114789927B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117163730A (en) * 2023-11-02 2023-12-05 南通明富纺织品有限公司 Production cloth rolling system based on internet of things

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1434289A (en) * 1995-10-27 2003-08-06 日本钢管株式会社 Method for detecting surface defect and device thereof
CN103093475A (en) * 2013-01-28 2013-05-08 海信集团有限公司 Image processing method and electronic device
CN105523417A (en) * 2014-10-20 2016-04-27 柯尼卡美能达株式会社 Image forming apparatus and image formation system
CN205652888U (en) * 2016-05-31 2016-10-19 马春阳 A receive unwinding device for rolling up type product processing
CN112224868A (en) * 2020-10-17 2021-01-15 东莞科卓机器人有限公司 FPC (Flexible printed Circuit) light bar feeding method based on CCD (Charge coupled device)

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1434289A (en) * 1995-10-27 2003-08-06 日本钢管株式会社 Method for detecting surface defect and device thereof
CN103093475A (en) * 2013-01-28 2013-05-08 海信集团有限公司 Image processing method and electronic device
CN105523417A (en) * 2014-10-20 2016-04-27 柯尼卡美能达株式会社 Image forming apparatus and image formation system
CN205652888U (en) * 2016-05-31 2016-10-19 马春阳 A receive unwinding device for rolling up type product processing
CN112224868A (en) * 2020-10-17 2021-01-15 东莞科卓机器人有限公司 FPC (Flexible printed Circuit) light bar feeding method based on CCD (Charge coupled device)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117163730A (en) * 2023-11-02 2023-12-05 南通明富纺织品有限公司 Production cloth rolling system based on internet of things
CN117163730B (en) * 2023-11-02 2024-03-12 南通明富纺织品有限公司 Production cloth rolling system based on internet of things

Also Published As

Publication number Publication date
CN114789927B (en) 2022-09-23

Similar Documents

Publication Publication Date Title
CN115311292B (en) Strip steel surface defect detection method and system based on image processing
CN116977358B (en) Visual auxiliary detection method for corrugated paper production quality
CA3053219C (en) Real-time, full web image processing method and system for web manufacturing supervision
CN109035195B (en) Fabric defect detection method
CN114842007B (en) Textile wear defect detection method based on image processing
CN114789927B (en) Artificial intelligent control method and system for textile fabric gray cloth winding machine
CN109816645B (en) Automatic detection method for steel coil loosening
CN111861996A (en) Printed fabric defect detection method
CN114862855B (en) Textile defect detection method and system based on template matching
CN115311507B (en) Building board classification method based on data processing
CN115311267B (en) Method for detecting abnormity of check fabric
CN115359053A (en) Intelligent detection method and system for defects of metal plate
CN116523899A (en) Textile flaw detection method and system based on machine vision
CN115063424B (en) Textile bobbin yarn detection method based on computer vision
CN115131348A (en) Method and system for detecting textile surface defects
CN105572143B (en) The detection method of rolled material surface periodic defect in calender line
CN114998321A (en) Textile material surface hairiness degree identification method based on optical means
CN115311279A (en) Machine vision identification method for warp and weft defects of fabric
Jing et al. Automatic recognition of weave pattern and repeat for yarn-dyed fabric based on KFCM and IDMF
US10388011B2 (en) Real-time, full web image processing method and system for web manufacturing supervision
CN115239961A (en) Method for monitoring working state of laser cutting machine
CN115018792A (en) Deep-drawing part wrinkling detection method based on computer vision technology
EP3459045B1 (en) Real-time, full web image processing method and system for web manufacturing supervision
Shuang et al. Surface defect segmentation and detection of printing roller based on improved FT algorithm
CN114210741B (en) Strip steel surface roughness online detection device and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant