An Automatic Recognition Method for PCB Visual Defects

Zhongqiu Zhang, Xiaodong Wang, Shan Liu, Li Sun, Liye Chen, Yangming Guo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

19 Scopus citations

Abstract

Aiming at the problems of low recognition rate and poor efficiency in traditional manual visual inspection of PCB apparent defects, an effective automatic recognition approach is proposed in this paper. The approach firstly obtains the apparent defect region image based on template matching defect detection algorithm. And then, the features of the apparent defect region image are extracted by combining the gray histogram features and the geometric features. Finally, this paper uses the optimized Support Vector Machine (SVM) classifier for automatic recognition and classification. The simulation results show that the method proposed for PCB's apparent defect in this paper can reach more than 96% recognition rate, which is an effective and feasible automatic recognition method.

Original languageEnglish
Title of host publicationProceedings - 2018 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2018
EditorsChuan Li, Dian Wang, Diego Cabrera, Yong Zhou, Chunlin Zhang
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages138-142
Number of pages5
ISBN (Electronic)9781538660577
DOIs
StatePublished - 2 Jul 2018
Event2018 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2018 - Xi'an, China
Duration: 15 Aug 201817 Aug 2018

Publication series

NameProceedings - 2018 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2018

Conference

Conference2018 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2018
Country/TerritoryChina
CityXi'an
Period15/08/1817/08/18

Keywords

  • SVM
  • apparent defect
  • automatic recognition
  • feature combination

Fingerprint

Dive into the research topics of 'An Automatic Recognition Method for PCB Visual Defects'. Together they form a unique fingerprint.

Cite this