光学模式识别和人工智能处理器

Optical Pattern Recognition and Intelligence Processors

介绍采用光学信息处理方法实现人工视觉功能的研究工作．以提取光学特征和相关方法为主要特点的光学模式识别已经得到发展，但是还不能完全解决目标识别中比例尺、偏移、旋转、畸变、强干拢、部分阻挡、多目标等一系列问题．目前正努力发展光学人工智能处理器，包括符号处理、联想存储和处理、决策网络等．

The study on the artificial vision function using the optical information processings is de- scribed. Optical pattern recognition characterized by optical feature extraction and correlation has been developed. However, the problems on the scale-, shift-, rotation-, distortion-in variants, interference, partial shadowing and multi-target for object recognition have not been solved satisfactorily. Recently, the optical artificial intelligence processors, including symbol processor, asso- ciative memory and processor, and decision network, have been developed acceleratedly.