Research on the application of binary-like coding and Hough circle detection technology in PCB traceability system

Abstract

With the rapid development of smart devices, printed circuit board (PCB), as the core components of smart devices, not only proposed more and more demand for the quality and precision of their production, but also the controllability of their management and the traceability of manufacturing. The traditional processes for PCB including: acid–base corrosion, water washing, grinding and so on. Due to the traditional spray bar code, thunder engraving two-dimensional code and other methods cannot be used for PCB board identification, but the drilling method has been utilized. At present, the major methods for drilling identification and matrix drilling identification are time consuming, and low identification speed. To tackle this problem, in this paper, we combine the advantages of binary coding and the industry characteristics of PCB production, proposed a kind of binary-like code to identify the position of PCB production, in which, each digit of the binary code can be identified by four holes of a square with equal distance, and can represent the numbers from 0 to 9. Besides, fewer letters and the starting position of the code on the basis of few punching points. This method will greatly improve the efficiency of drilling identification. The system uses the photoelectric analysis module to automatically scan the hole array code on the penal (PNL) board to obtain the original image. Then, the CCD image is preprocessed, mainly including Gaussian filter for image smoothing and unsharp mask operator for image enhancement. Then, Hough detection algorithm is utilized to locate the hole, which can effectively identify the holes. Secondly, the horizontal and vertical directions of these points can be obtained according to the distance between the center points of each hole. Furthermore, the image is righted, and the starting position can be obtained. At the same time, Hough detection algorithm is used to segment the image threshold, and each character is decoded into a corresponding number. The simulation results show that the method is simple, accurate and effective. Finally, the identified digital coding sequences are compared with the database of online code reading application system, and the reliable identification and automatic identification trace-ability of PCB production are realized, and the automatic identification time is less than 1 s.