All-optical Subtracted Joint Transform Correlator

Author:

Yu Wanji,Nakagawa Kiyoshi,Minemoto Takumi

Abstract

When the classical joint transform correlator (CJTC) was applied to pattern recognition, one of the most well known disadvantages was that the cross correlation signals between the object and the reference patterns as well as those between the reference patterns themselves would join closely in the same output plane. This fact causes inconvenience to identify the necessary output correlation signals between the object and the references, especially when a large number of references was used in the input image. Q.Zhan and T.Minemoto(1) proposed the subtracted joint transform correlator (SJTC) to remove the correlation signals between the references from the output plane, and the binnary subtracted joint transform correlator (BSJTC) to improve furthermore the output correlation results. According to the SJTC and BSJTC, the cross correlations between the references had been successfully removed from the output plane by subtracting the Fourier power spectrum of the references from the joint Fourier power spectrum of the input image. Thus far, the subtraction processing was performed by a computer. In this paper, a new all-optical SJTC system is proposed. In this system, a holographic interference was introduced to perform the subtraction of the Fourier power spectrum of reference patterns. There is not any digital processing by a computer in the system, it is possible to achieve the real time SJTC. The experimental arrangement of the system, the shift-invariant characteristics and the dscriminability of input patterns are described in the following.

Publisher

Optica Publishing Group

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3