After years of development, machine vision technology has become more and more mature. Machine vision systems are indispensable for realizing factory automation. The high-precision requirements of assembly positioning, quality inspection, product identification, size measurement, and high-speed production lines are no longer available to the human eye. Satisfy. Once costs drop, performance increases, and applications increase, just like the changes brought about by the popularization of wireless networks, machine vision software and hardware products will become standard in the manufacturing industry.
How to reduce the cost? There are many imaginable spaces. Embedded machine vision system will be a good choice, and the core of the system is of course inseparable from an industrial computer. Since the 1990s, microprocessor, semiconductor technology and machine vision technology have been spiraling upward together: microprocessor and semiconductor technology are the birthplaces of machine vision. Europe and the United States have applied image technology to them in large quantities, and then slowly evolved into today’s machine vision. technology. The popularity of machine vision in Europe and America is also mainly reflected in the semiconductor and electronic industries. Now, microprocessor performance has become more powerful, power consumption has been greatly reduced, and the size has become more compact, but the price has not increased.
The working principle of machine vision is very simple. It is very similar to a single-function industrial robot. For example, it uses a camera to obtain the image of the detected object, and the special light source enables the features of the detected object to be enhanced in the image, such as size features and surface defects. Features etc. The images with clear characteristics are transmitted to the industrial computer at high speed, and the characteristics are calculated by professional image analysis and processing software. According to the calculation results, the industrial computer can control the actuator to perform classification, removal, alarm and other actions. If compared with humans, then cameras are like human eyes, industrial computers are like human brains, and software is like human thoughts. Then the actuators are like human hands and feet. In fact, the machine vision system is like a person with fast thinking, quick actions, and tirelessness.
The vision system generally includes light source, lens, CMOS camera, image processing unit, image processing software, monitor, communication/input output unit, etc. The output of the vision system is not an image or video signal, but a detection result (such as size data) after arithmetic processing.
Machine vision testing is to use a machine instead of the naked eye for measurement and judgment, so it requires high accuracy. First, a CMOS camera is used to convert the captured target into an image signal, which is sent to a dedicated image processing system, and converted into a digital signal according to the pixel distribution, brightness, and color information. The image system performs various operations on these signals to extract the characteristics of the target, such as: area, length, number, location, etc. Finally, output results according to preset tolerances and other conditions, such as: size, angle, offset, number, pass/fail, yes/no, etc. After the upper computer obtains the detection result in real time, it directs the motion system or I/O system to perform corresponding control actions.
Anrecson industrial computer product configuration:
Intel® Core series processors
12V, non-ECC DDR4 SODIMM x 2, supports up to 32G DDR4
USB 3.0 x 4, USB 2.0 x 2 with high-performance CPU, strong GPU performance, convenient for visual applications
SATA/mSATA, GPIO x 8, COM x 4
The IPC-805 embedded industrial computer has 4 USB 3.0 ports, 2 USB 2.0 ports, 2 intel Gigabit Ethernet ports, and VGA, HDMI support. By balancing flexibility, efficiency and size, system integration is realized, satisfying and surpassing Ideal for customer needs.
In recent years, industrial computer technology has developed rapidly. Relying on the strong support of industrial computer, machine vision technology has been increasingly applied to the online inspection of industrial production, realizing rapid and complete inspection of product quality, and then achieving product quality Of zero defects. Industry 4.0 is about connecting production technology and information technology, and machine vision is one of the most important basic technologies that provide information for Industry 4.0. Embedded systems will play a more important role in future machine vision systems. They can realize the design of compact products, meet the requirements of most image processing, and compared with independent PC-based systems, their integration with factory processes is higher.
