Application of Deep Transfer Learning in Electronic Connector Recognition and Inspection

Abstract

As a key component for device interconnection, the quality inspection of electronic connectors has become a core link in the production process. This article is based on deep transfer learning to build a connector recognition and detection hardware system. Using a layered design approach, the five layer architecture of "data layer preprocessing layer transfer learning detection layer post-processing layer application layer" is analyzed, and a small sample connector defect detection method that integrates data augmentation is proposed, which has high application value.