A model of saliency-based selective attention for machine vision inspection application

Authors:
Xiao-Feng Ding;Li-Zhong Xu;Xue-Wu Zhang;Fang Gong;Ai-Ye Shi;Hui-Bin Wang
Affiliations:
College of Computer and Information Engneering, Hohai University, Nanjing, China;College of Computer and Information Engneering, Hohai University, Nanjing, China and Institute of Communication and Information System Engineering, Hohai University, Nanjing, China;College of Computer and Information Engneering, Hohai University, Nanjing, China;College of Computer and Information Engneering, Hohai University, Nanjing, China;College of Computer and Information Engneering, Hohai University, Nanjing, China and Institute of Communication and Information System Engineering, Hohai University, Nanjing, China;College of Computer and Information Engneering, Hohai University, Nanjing, China and Institute of Communication and Information System Engineering, Hohai University, Nanjing, China
Venue:
ICANNGA'11 Proceedings of the 10th international conference on Adaptive and natural computing algorithms - Volume Part II
Year:
2011

Citing 4
Cited 0

A Model of Saliency-Based Visual Attention for Rapid Scene Analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Selective Attention-Based Method for Visual Pattern Recognition with Application to Handwritten Digit Recognition and Face Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Copper Strip Surface Defects Inspection Based on SVM-RBF

ICNC '08 Proceedings of the 2008 Fourth International Conference on Natural Computation - Volume 02
Applying wavelets transform, rough set theory and support vector machine for copper clad laminate defects classification

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

A machine vision inspection model of surface defects, inspired by the methodologies of neuroanatomy and psychology, is investigated. Firstly, the features extracted from defect images are combined into a saliency map. The bottom-up attention mechanism then obtains "what" and "where" information. Finally, the Markov model is used to classify the types of the defects. Experimental results demonstrate the feasibility and effectiveness of the proposed model with 94.40% probability of accurately detecting of the existence of cropper strips defects.