In Machine Vision, defect detection function is one of the most widely used functions, which mainly detects various information of product surface. In modern industrial automation production, each process in the continuous mass production has a certain defective rate. Although the rate is small on its own, it becomes a bottleneck for enterprises to improve the yield rate after multiplication, and the cost of eliminating defects after the full process is much higher (for example, if the solder paste printing process is misaligned and the problem is not detected until the chip is tested online after assembly, the cost of rework will be more than 100 times the original cost) , therefore, it is very important for quality control and cost control to detect and reject the defective products in time, and it is also an important foundation for the further upgrade of manufacturing industry.

In the detection industry, compared with human vision, machine vision, obvious advantages

1) HIGH ACCURACY: Human Vision is 64 gray level, and the resolution of small targets is weak; machine vision can significantly improve the gray level, while the observation of Micron level targets;

2) fast speed: human can not see fast moving targets, machine shutter time can reach the microsecond level;

3) high stability: Machine Vision solve a very serious human problem, unstable, artificial visual inspection is very boring and hard work of the industry, no matter what you design the reward and punishment system, there will be a relatively high miss detection rate. But machine vision monitors have no fatigue problems, no mood swings, and anything you write in your algorithm is executed carefully every time. In the quality control greatly enhance the controllability of the effect.

4) Information Integration and retention: The amount of information obtained by machine vision is comprehensive and traceable, and related information can be easily integrated and retained.

To be free of all artificial visual inspection, there are still many difficulties to be solved in machine vision:

1) light source and imaging: in Machine Vision, high-quality imaging is the first step, because different materials surface reflection, refraction and other problems will affect the detection of the characteristics of the object under test, so light source and imaging can be said to be the first obstacle to be overcome in machine vision detection. For example, glass, reflective surface scratch detection, many times the problem is stuck in the integrated imaging of different defects.

2) feature extraction from low-contrast images in heavy noise environment: In heavy noise environment, it is difficult to distinguish true and false defects, which is also the reason why there is always a certain false detection rate in many scenes.

3) recognition of unexpected defects: In applications, it is often given some specific defect patterns, using machine vision to identify whether or not they occur. But it's often the case that many obvious flaws are missed because they haven't happened before, or because the patterns are too diverse. If it were a human, he would not have been asked to detect the defect in the operating procedure file, but he would have noticed it and had a better chance of catching it, and machine vision's "intelligence" at this point is currently harder to break through.

Returns List