CN103018250B - Image detection system for seamlines of fabric in calendering process - Google Patents

Abstract

The image detection system of the fabric seam line in the calendering process belongs to the new textile detection field. In the calendering process, when the hot rubber roller is ironed to the seam line of the fabric, due to the sudden increase in the thickness of the seam, marks will be left on the surface of the rubber roller, which will be copied to the subsequent ironed fabric surface , affect the appearance and flatness of the cloth surface, and even produce defective products. Manually control the rise and fall of the hot rubber roller to avoid the seam line, there are visual errors, poor response speed and fatigue errors, etc., which will cause excessive avoidance or insufficient avoidance of the seam line. The invention proposes an automatic detection system for fabric seams based on embedded vision. The moving image of the fabric is collected by the CCD camera, input into the DSP, and the embedded algorithm is used to detect whether there is a seam in the image. When the seam is detected, the machine is controlled to lift the rubber roller, and the seam is automatically avoided to realize the fabric Intelligent image detection of seam lines reduces the labor intensity of workers.

Description

An Image Detection System for Fabric Seamline in Calendering Process

technical field

The invention relates to an image detection system for fabric seam lines in a calendering process, including image acquisition and image processing of moving fabrics, which belongs to the field of new textile detection, and specifically relates to technologies such as image acquisition and processing of embedded machine vision .

Background technique

In the calendering process often involved in textile enterprises, when the hot rubber roller is ironed to the seam line of the fabric, due to the sudden increase in the thickness of the seam, marks will be left on the surface of the rubber roller, which will be copied to the subsequent ironing On the surface of the ironed fabric, it will affect the appearance and flatness of the fabric surface, and even produce defective products. Therefore, in the traditional calendering process, it is necessary to manually control the rise and fall of the hot rubber roller to avoid the seam line and avoid the above-mentioned quality problems. However, this method has visual errors, reaction speed errors, and fatigue errors caused by continuous tension. These subjective factors may cause excessive seam line avoidance or insufficient avoidance.

In recent years, with the development of image processing technology, its application range has become more and more extensive. In the textile industry, many fabric-related inspections rely on the inspector's vision, such as fabric defect inspection, foreign fiber inspection and evaluation, etc. The development of image processing technology makes the above problems expected to be solved by automated methods.

Aiming at the problems existing in seam line detection in the calendering process, the invention proposes an automatic fabric seam line detection system based on image analysis technology. The moving image of the fabric during the calendering process is collected by a digital camera and input to the embedded chip to detect whether there is a seam line in the image. When the seam line is detected, the rubber roller in the machine is automatically controlled to lift the rubber roller to avoid Seamlines. The implementation process of the detection system proposed by the present invention is shown in FIG. 1 .

Contents of the invention

The object of the present invention is to provide an image detection scheme for fabric seamlines in the calendering process, which is used to realize automatic detection of fabric seamlines. The technical scheme that the present invention adopts is as follows:

(1) Use a digital camera to complete the collection of moving fabric images, input them into the embedded chip through the data communication interface, and complete the automatic detection of the seam line through image analysis. When the fabric seam line is detected, the rubber roller is controlled to automatically lift up , to avoid the fabric seam line, an example diagram of the system is shown in Figure 2;

(2) Input the collected image into the DSP development chip through the data communication line to complete the embedded visual inspection, and the data communication interface adopted is the AV interface;

(3) Through the characteristic parameters of the fabric image, it is judged whether there is a seam line in the image;

(4) When the fabric seam line is detected, the signal is sent to the single-chip microcomputer or PLC through the output interface to control the lifting of the rubber roller.

Description of drawings

Figure 1 Fabric seam line detection process

Figure 2 Schematic diagram of fabric seam detection system

Figure 3 fabric seamline image

Detailed ways

Install the digital camera above the moving fabric. It should be noted that the detection system is located in front of the rubber roller. When the seam line is detected, a trigger signal is output to control the lifting of the rubber roller. During the detection and control time, the fabric The moving distance should be less than the distance between the detection system and the rubber roller.

Since the width of the fabric seam line is consistent with the width of the fabric, it is not necessary to take an image of the entire seam line during detection, but only a part of it. Therefore, when the digital camera is installed, it is located on one side of the moving fabric, and the distance between the camera and the fabric is about 20cm. The system provides a double-sided transmitted white light source, so that the seam line in the captured fabric image is clearly identifiable. Figure 3 is the actual detection system Images of fabric seamlines captured in .

The image data is transmitted to the embedded chip through the data connection line. This system uses the data line of the AV interface, and the embedded chip uses DSP (DM642).

In the DSP chip, calculate the mean value and standard deviation parameters of the image transmitted by each frame, and further calculate the characteristic parameters of the coefficient of variation of the image. The calculation method is as follows:

$\mathrm{<span\; class="notranslate">\; cv</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&sigma;</span>}<span\; class="notranslate">\; /</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}$

in is the mean value of the image, σ is the standard deviation of the grayscale image, and CV is the coefficient of variation of the image.

The CV value reflects the relative change of image grayscale, and objectively reflects the grayscale dispersion degree of each group of feather cloth images, which can be used as a standard for detecting whether there is a seam line.

In this example system, the CV value of 0.07 is selected as the basis for judging whether there is a seam in the fabric image. When the CV value is greater than 0.07, it is considered that there is a seam in the image, and when the CV value is less than 0.07, it is considered that there is no seam in the image. Wire. The value of the judgment basis in the specific example system is closely related to the selection of the light source of the system, and the best judgment value can be obtained through example training.

Claims (2)

1. An image detection system for fabric seamlines in the calendering process, characterized in that: the automatic detection of fabric seamlines in the calendering process is realized by using embedded machine vision; The communication interface is input into the DSP chip, and the image detection algorithm is used to judge whether the image contains a seam line; the system includes an output interface, a rubber roller connected to a single-chip microcomputer or a PLC control machine, and when a seam line is detected, a trigger signal is output. Control the machine to lift the rubber roller to avoid the seam line; In the DSP chip, calculate the mean value and standard deviation parameters of the image transmitted by each frame, and further calculate the characteristic parameters of the coefficient of variation of the image. The calculation method is as follows: $\mathrm{<span\; class="notranslate">\; cv</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&sigma;</span>}<span\; class="notranslate">\; /</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}$ in, is the mean value of the image, σ is the standard deviation of the grayscale image, and CV is the coefficient of variation of the image. The CV value reflects the relative change of image grayscale, and objectively reflects the grayscale dispersion degree of each group of feather cloth images, which is used as the standard for detecting whether there is a seam line. 2. The system according to claim 1, characterized in that: a CCD digital camera is used to collect moving fabric images, the resolution of the camera is above 640 pixels×480 pixels, and the seam lines in the collected fabric images are required to be clearly identifiable.