Self-learning machine vision

被引：0

作者：

Acuity Imaging Inc, Nashua, United States ^{[1
]}

机构：

来源：

Ind Comput | / 6卷 / 12-15期

关键词：

Image analysis - Inspection - Learning systems - Manufacturing data processing - Neural networks - Quality control - Signal filtering and prediction;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

A technique is presented for simplifying the training process of machine vision systems in judging the quality of parts they were designed to inspect. The show-and-go technology trains a vision system by simply showing it several examples of good parts. The system can then `remember' what characteristics constitute a good part and can use this information to inspect the rest of the parts. Several characteristics of the show-and-go machine vision are discussed.

引用