A PCB SMD Solder Quality Inspection Method Based on Dual-Path Region Segmentation and Color Clustering

During the PCB production process, interference from environmental light, reflections, and equipment vibrations often results in captured images suffering from noise, uneven illumination, and low contrast, which directly impairs subsequent defect detection. To address these challenges, we propose a multi-stage cascaded preprocessing strategy that integrates grayscale conversion, histogram equalization, Gaussian and median filtering, adaptive Otsu thresholding, and morphological operations to effectively enhance image quality. For solder quality inspection of surface-mounted components, a dual-path region segmentation algorithm is designed, which combines contour analysis with projection analysis to achieve accurate segmentation of both the component bodies and the solder pad regions. Additionally, we leverage the H channel in the HSV color space for color clustering and mapping, enabling automatic solder quality assessment. Experimental results on 461 real-world production images demonstrate the effectiveness of our approach, achieving an inspection accuracy of 0.99 and a recall rate of 1.00, thereby confirming the strong robustness and promising applicability of our system in complex industrial environments.