Machine vision-based electrical connector pin identification and quality inspection method research

The quality of each component in the aerospace industry is integral to the safety of the entire system, with the quality inspection of electrical connectors being a critical aspect of ensuring safety. This paper researches a machine vision-based electrical connector pin automatic identification and quality inspection method for actual engineering needs. Due to the issues of sample scarcity and high computational power consumption associated with deep learning methods, our method uses the captured pin images and employs a template-matching-based algorithm for the automatic recognition of pins and calculation of their skew degree. The correct rate of pin identification is 100%, and skewed pin detection accuracy is better than 0.05 mm. The proposed method can promptly address the current issues of low efficiency, high rate of missed inspections, and chaotic management in the quality detection of electrical connector pins.