Adaptive PCB plate defect visual positioning detection and classification system

• Main Authors: ZHANG MEIJIE, LI XIANWEI, WEI DENGMING, ZHANG LU, WANG HUALONG, HU XIAOQIANG, ZHANG PENGZHONG
• Format: Patent
• Language: Chinese; English
• Published: 04.06.2019
• Subjects: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

The invention provides an adaptive PCB plate defect visual positioning detection and classification system. A whole set of processes of PCB detection is covered. The system comprises an offline pre-processing contour point attribute classification stage, online PCB whole plate alignment, defect detection and defect classification. A template matching method is used to align a whole plate, and a speed is fast and precision is high. A contour out-of-tolerance algorithm is used to detect a defect, the precision reaches a sub-pixel level, the speed is fast, and universality is high. Attribute classification is performed on contour points so as to satisfy different tolerances generated because different types of defects match with different detection standards. In the invention, a detection speed and precision of a PCB are increased, detection efficiency is effectively improved, false detection or leak detection conditions of defects are reduced, the automatic detection of PCB surface defects is realized, manpower