PCB inspection in the context of smart manufacturing

Abstract

Manual assembly of Printed Circuit Board (PCB) in small- and medium-sized enterprises is error-prone and tedious. A small change in orientation of the components might affect the functionality of PCB to an unpredictable extent. We have focused on Automated Optical Inspection (AOI) system for PCB inspection problem in the context of smart manufacturing. The user interface is empowered with bi-directional task—(I) Industrial worker can inspect whether ICs exist or not by comparing with corresponding Gerber file (II) Automatic Inspection of PCB shall be performed in the absence of Gerber file. This is a preliminary step in the direction of automated real-time analysis of PCBs. The work presented in this paper deals with the image acquisition phase of inspection too, which is relatively less investigated when compared to image processing phase. Present automatic inspection is achieved with the histogram of hue values defined in the Hue Saturation Value (HSV) color space. Color distribution-based segmentation is carried out by taking the background color as the dominant value in the histogram. Then in the next step, features extracted from the resultant parts are compared with template ICs to obtain robust results and confirmed ICs are annotated in the live image. We have evaluated the efficiency of the algorithm in four different lighting conditions—outdoor light, LED, CFL, and incandescent lights and observed that 12 W LED works best with 2500 lx illumination. Then with the 12 W LED, we have analyzed accuracy with different camera resolution. We have observed that algorithm works best in the resolution of 1920 × 1080 which is both time consuming and cost-effective.