Intelligent Identification System Based On Machine Vision, Real-time Hub

In recent years, with the rapid growth of wheel production, wheelmanufacturer needs a lot of production lines. Only relying on artificial way toidentify all kinds of wheels on every industrial production line cannot satisfy theneed of reality and many problems have been exposed. For example, lowefficiency and high labor costs. At the same time, great labor intensity and thelimitations of the human eye lead to the error identification. So wheel enterprisesare in urgent need for systems of simple structure, high recognition rate and therequirements of the automatic production can be satisfied on the production linein order to improve productivity, reduce costs and increase the wheel recognitionaccuracy.Image processing and vision technology can identify complex targets andhas been broad used for detection of industrial products and intelligentclassification with the non-contact, high precision, strong anti-interference ability,etc. And the online intelligent identification of wheels is one of very typical.This paper designs a real-time online automotive wheel automaticidentification system by image processing and vision technique. This paperimplements the wheel intelligent identification through wheel imagepreprocessing, feature extraction, multi-parameter comprehensive processing andoffset identification. On this basis, PLC realizes automatic sorting operation ofdifferent kinds of wheels through connecting to the computer control system.The main innovation of this system is as follows:1) This paper puts forward wheel identification methods based on lookingfor center of wheel and rotating image template matching.2) This paper improves the recognition accuracy of wheels by usingmulti-parameter comprehensive technique.3) This paper guarantees the robustness of the system and enhances theaccuracy of system identification by combination wheel above images withwheel below laser distance measurement.