The charger usually has a lot of information on its surface, such as charging voltage, output voltage, current, production place, batch number, manufacturer and so on. These information is usually silk screen. However, silk screen will often appear  printing defect, such as offset, shading, ink, missing words, lack of zoned, dust, bubbles, and so on. For the applications mentioned in this article, it is still a good situation. Because the print surface of the charger is flat, the product is smooth and white. In addition, some products will be printed on the arc surface, the surface of the product may be frosted surface, the color is not necessarily white, may be black, gray and so on, so it is trouble to deal with.

Machine Vision Light :

You can try to use the coaxial light, light source, light source shadowless dome ring. For the white products can try the UV light source, that may have good results.

FA Camera :

Because of detecting silk screen poor appearance, the resolution must be high. Only in this way can we detect some details. So try to use a high resolution industrial camera at any cost.

FA Lens :

Try to use high quality industryl lens, such as bi- telecentric lens and HD resolution.

Image Processing Algorithm :

Mainly consider the gold template matching method to achieve. Because the main test of such products is the gold template matching. Create one or a set of standard templates to compare. For NI VISION, to do gold template matching need to be able to locate accurately, if the location is inaccurate, it is easy to misjudge.

Machine Vision Project Implementation Difficulty:

Although the project is relatively difficult, but for this project, it is possible to implement.  In the software above, it is recommended to consider the use of HALCON or VISIONPRO like some of the more powerful library functions to deal with.