Application of Machine Vision in Fabric Production Online Inspection System

First, the cloth production on-line detection overview Automatic detection technology as a rapid, real-time, accurate collection and processing of information, high-tech, has gradually become the national economy informatization, and enhance the competitiveness of indispensable technical tools and tools. In the modern production line, cloth needs to determine whether the color of the cloth is acceptable and whether there are impurities and the amount of impurities on the cloth. Because the production line runs faster and requires less impurity-resolving diameters, it is difficult to do real-time inspections manually, the post-sampling inspection efficiency is low, and the products after spot inspection still have the possibility of defects. Computer automation is just right for rapid and real-time detection. The cloth production on-line inspection system is based on machine vision technology to quickly and efficiently detect the color and presence of impurities in fabrics. Machine vision is the use of machines instead of the human eye to make measurements and judgments. The machine vision system refers to converting a captured object into an image signal through a machine vision product (ie, an image capturing device) and transmitting it to a dedicated image processing system. The signal is converted into a digital signal according to information such as pixel distribution, brightness, and color; the image system is These signals perform various operations to extract the characteristics of the target, and in accordance with the result of the discrimination, the content of the image is identified or the equipment operation in the field is controlled.

2. Project Requirements During the production process of cloth, highly repeatable and intelligent work such as cloth quality inspection can only be completed by manual inspection. Many modern inspection workers can often see this work after the modern assembly line. The process, while adding huge labor costs and management costs to the company, still cannot guarantee 100% inspection pass rate (ie "zero defect"). The detection of cloth quality is repetitive, error prone and inefficient.

The automated transformation of the assembly line makes the cloth production line a fast, real-time, accurate, and efficient assembly line. In the assembly line, the color and quantity of all fabrics are automatically confirmed (hereinafter referred to as “cloth detection”). Machine vision automatic identification technology is now used to complete the work previously done manually. In high-volume fabric testing, manual inspection of product quality is inefficient and inaccurate, and machine vision inspection methods can greatly increase production efficiency and production automation.

III. Project Plan Machine vision uses a computer to process and analyze image information and make conclusions without human intervention. Machine vision is characterized by automation, objectivity, non-contact, and high precision. Compared with image processing systems in the general sense, machine vision emphasizes accuracy and speed, and reliability in industrial field environments.

In machine vision applications, include the following processes:

Image acquisition through the optical system, the camera captures the image, the image is converted into digital format and into the computer memory.

The image processing processor uses different algorithms to process image elements that have an important influence on the decision, such as color recognition of the image, area, length measurement, image enhancement, edge sharpening, noise reduction, and the like.

The feature extraction processor recognizes and quantifies key features of the image, such as the color of the cloth and the shape of the impurities. Then this data is transmitted to the control program.

The decision and control processor's control program makes conclusions based on the data received. For example, these data include whether the diameter of the impurity is within the required specification or whether the color of the cloth is acceptable.

The vision system generally includes: a light source, an optical system, a camera, an image processing unit, an image analysis processing software, a monitor, a communication/input/output unit, and the like. The composition of the cloth inspection machine vision system is shown in the figure below. The output of the vision system is the result of the calculation after processing - the amount of various impurities. After the computer system obtains the test result in real time, the command motion system or the input/output system performs a corresponding control action (such as sorting).

The following is a detailed plan design specification for the cloth detection system:

Image Processing Software In the machine vision system, the processing technology of visual information mainly relies on image processing methods. It includes image enhancement, data encoding and transmission, smoothing, edge sharpening, segmentation, feature extraction, image recognition and understanding. After these processes, the quality of the output image is improved to a considerable degree, which not only improves the visual effect of the image, but also facilitates computer analysis, processing, and recognition of the image.

Feature Extraction Recognition General Fabric Detection (Automatic Recognition) First, use a high-definition, high-speed camera lens to take a standard image, set a certain standard based on this, and then shoot the detected image, and then compare the two. However, it is more complicated in the fabric quality inspection project:

1. The content of the image is not a single image, and the number, size, color, and position of impurities present in each measured area are not necessarily the same.

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

3. Due to the rapid movement of the fabric, reflections of light may cause a large amount of noise in the image.

4. On the assembly line, the cloth is tested and there are real-time requirements.

Due to the above reasons, image recognition processing should adopt corresponding algorithms to extract the characteristics of impurities, perform pattern recognition, and achieve intelligent analysis.

We use the color and blob tools in the Stemware Machine Vision Software Development Kit - CVB, which is suitable for the development of color pattern recognition and spot detection.

Color Detection In general, the images acquired from the color CCD camera are RGB images. That is, each pixel consists of three components: a red (R) green (G) basket (B) to represent a point in the RGB color space. The problem is that these color differences are different from the feeling of the human eye. Even small noise can change the position in the color space. So no matter how similar our human eyes feel, it is not the same in the color space. For 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 differences in the color space.

Blob Detection According to the processed image obtained above, impurity stains are detected on a solid background according to requirements, and the area of ​​the excellent patches is calculated to determine whether it is within the detection range. Therefore, the image processing software has the function of separating the target, detecting the target, and calculating its area.

Blob Analysis (Blob Analysis) analyzes the connected pixels of the same pixel in the image. This connected space is called a Blob. The stain in the processed image after binarization can be considered as a blob. The Blob analysis tool can separate the target from the background and calculate the number, position, shape, orientation, and size of the target. It can also provide the topology between related spots. Instead of using individual pixels to analyze one by one during the process, the rows of the graph are manipulated. Each line of the image uses run length coding (RLE) to represent the adjacent target range. This algorithm greatly increases the processing speed compared to pixel-based algorithms.

The result processing and control application stores the returned result in a database or user-specified location and controls the mechanical part to perform the corresponding movement based on the result.

According to the result of the recognition, it is stored in the database for information management. From now on, information can be searched and inquired at any time. Managers can be informed of the busyness of a certain period of time and make arrangements for the next step; they can learn about the quality of fabrics in the near future.

Fourth, the user interface and the operation project require the use of machine vision technology to intelligently identify all the impurities in the cloth on the assembly line and their quantity and size. According to the project requirements, we design as follows:

(1) Image display area: The color image collected by the camera is displayed in real time, and the system recognizes cloth information in real time according to the current image content.

(2) Information display area: The content of the image - the number of various impurities is displayed in real time in the form. The current status of the system (such as: real-time detection, stop detection, trigger signal status) is displayed in the status display column in real time so that the operator can understand the system status.

(3) Information Management Area: Managers can view the statistics of the pipeline at any time. The operator can flexibly configure the system configuration information (such as: database configuration, control module communication configuration, identification parameter correction). Rights Management Users of control system operations, for example: Only senior operators can configure system information; only those with the appropriate permissions can view statistics.

Fifth, fabric color learning tools We have developed cloth color learning tools, this tool friendly interface, easy to operate.

Cloth color learning tool A color should provide multiple template images for training, which can improve the recognition ability. After the learning is completed, it is saved as a CLF file, and the pattern recognition is then recognized according to the saved features.

VI. Conclusion The vision system involves optical and image processing algorithms. It is a highly specialized product in itself. Especially in the entire identification control system, it must cooperate with the motion control system to complete the follow-up operations. In the visual system of the project, the color feature values ​​of the identification object are extracted, and then the pattern recognition method is used to identify the unqualified area and then use spot analysis to determine whether it is an impurity. At the same time, the selection of various components in the entire system and the design of the user interface were mentioned.

In short, the application of machine vision systems can significantly reduce inspection costs, improve product quality, and accelerate production speed and efficiency. For modern companies, it is no doubt that they will be at the forefront of competition if they are aware of the trends in technology development and put it first.

Sheet Metal Forming processes are those in which force is applied to a piece of sheet metal to modify its geometry rather than remove any material. The applied force stresses the metal beyond its yield strength, causing the material to plastically deform, but not to fail. By doing so, the sheet can be bent or stretched into a variety of complex shapes. 

Sheet Metal Forming

Sheet Metal Forming,Metal Forming,Metal Forming Machine,Sheet Metal Forming Machine

Chongqing Jin Te Rui Machine Co.,Ltd ,

Posted on