Plastic parts defect detection, which is a lot of applications in the field of plastics, belongs to one of the appearance defect detection. This kind of product usually comes out of the mold. There are many circumstances that cause the gap, such as insufficient feeding, foreign matter in the mold and so on. Therefore, testing the integrity of plastic parts is a necessary test.

Machine vision light source: Because the shape of each plastic product may be different, the light source is not universal. However, such products generally cannot use backlight as lighting. The positive gap can generally be considered low-angle ring light source lighting, strip light source lighting or coaxial light source lighting. The overall lack of shadowless light can be used, and if there is only a lack of surface material, shadowless light generally does not show up, and a low-angle light source or a coaxial light source may be required.

Industrial cameras: Generally speaking, the gap on the plastic parts will be relatively large, and the smaller gap is not easy to detect, so the resolution of the camera only needs to be general, 1.3 million pixels to 5 million pixels is almost. Of course, if the budget allows, the use of high-resolution cameras can also be, but the detection speed is slow, the price is relatively high.

Industrial lens: It is best to have a better image quality lens, as for the use of CCTV lens or telecentric lens, it depends on the specific cost budget. Generally there is not much need to use telecentric lenses, the price is too high, and the use of such a good performance of the product.

Image processing algorithm: The detection of this product is mainly based on the fact that the surface of OK product and NG product are not on the same plane, and the color or brightness of the surface will change when lighting, which brings the feasibility of measurement. Such as using particle analysis, gray scale measurement, etc. The popular deep learning used in defect detection is also relatively good, and the effect is better than the traditional algorithm.

Detection requirements: damage, crush

solution

SIPOTEK vision inspection software 1 set, SIPOTEK custom industrial computer 1 set, PHILIPS 22 "LCD monitor 1 set, SONY industrial camera 1 set, SIPOTEK custom camera adjustment module 1 set, telecentric HD optical industrial lens 1 set, custom optical adaptive ring light source 1 set, SIPOTEK custom control system 1 set, Panasonic servo motor 1 set, SIPOTEK custom reducer 1 set.

Detection effect

Dark spot

Dark spot

Surface defect

Surface damage

Amorphosis

Surface defect

Surface defect

Surface defect