Application of machine vision in cloth quality inspection

Machine vision is to use machines instead of human eyes for measurement and judgment. Machine vision systems use machine vision products (that is, image pickup devices, divided into two types: CMOS and CCD) to convert the captured target into image signals and send them to a dedicated image processing system To obtain the morphological information of the subject, and convert it into digital signals according to pixel distribution, brightness, color and other information. The image system performs various operations on these signals to extract the characteristics of the target, and then controls the field equipment operations based on the discrimination results. .

Today, iJUNCO talks with everyone about the application of machine vision in fabric quality inspection.

Application of machine vision technology in fabric defect detection.png

Application of machine vision in fabric defect detection
 

The inspection of cloth quality is a repetitive labor. Many modern inspection workers can often be seen behind the modern assembly line to carry out this process, which adds huge labor and management costs to the enterprise, but still cannot guarantee the inspection pass rate, which is easy. Errors and inefficiencies.

In large-scale cloth inspection, manual inspection of products has low quality efficiency and low precision. Therefore, more and more lines have been transformed automatically, so that the cloth production line has become a fast, real-time, accurate, and efficient pipeline. On the assembly line, the automatic recognition technology of machine vision is used to complete the work previously done manually, and the color and quantity of all fabrics are automatically confirmed (hereinafter referred to as "cloth detection"). This detection method can greatly improve production efficiency and The degree of automation of production.

Application of machine vision technology in the appearance inspection of cloth.png

Application of machine vision in the appearance inspection of cloth

Feature extraction identification

Generally, cloth detection (automatic recognition) first uses a high-definition, high-speed camera lens to take standard images, sets a certain standard based on this, then takes the detected image, and then compares the two.

But it is more complicated in the cloth quality inspection project:

1. The content of the image is not a single image, and the quantity, size, color, and position of impurities in each measured area may not be consistent.

2. The shape of the impurities is difficult to determine in advance.

3． Due to the rapid movement of the cloth reflecting light, there may be a lot of noise in the image.

4． On the assembly line, there is a requirement for real-time detection of cloth.

Due to the above reasons, corresponding algorithms should be adopted in image recognition processing to extract the characteristics of impurities, perform pattern recognition, and realize intelligent analysis.

Application of machine vision cloth color difference detection.png

The overall structure of machine vision in color difference detection system

Color detection

Generally speaking, images obtained from color CCD cameras are RGB images. In other words, each pixel is composed of three components, red (R), green (G), and blue (B), to represent a point in the RGB color space. The problem is that these color differences are different from the human eye's perception, even if they are small. Noise also changes the position in the color space, so no matter how close our eyes feel, it is not the same in the color space. Based on the above reasons, we need to convert RGB pixels into another color space CIELAB, the purpose is to make our human eyes feel as close as possible to the color difference in the color space.

Blob detection

According to the processed image obtained above, according to the requirements, the impurity spot is detected under a solid color background, and the area of the outstanding spot is calculated to determine whether it is within the detection range. Therefore, image processing software should have the functions of separating targets, detecting targets, and calculating their area.

Results processing and control

The application stores the returned results into a database or a location specified by the user, and controls the mechanical part to perform corresponding movements according to the results. According to the recognition result, it is stored in the database for information management, and the information can be retrieved and queried at any time in the future. The manager can know the busyness of the pipeline within a certain period of time, make arrangements for the next work, and can know the quality of the cloth. .