摘要
本文研究了Ce:KNSBN光折变联合变换相关,以Ce:KNSBN光折变晶体作为平方律转换器件,用波长为514.5nm的Ar+离子激光做写入光,在系统的傅里叶变换平面上放置光折变晶体记录输入靶物和参考物的联合变换功率谱,用632.8nm的He-Ne激光作为读出光,读出联合变换功率谱,输入到计算机进行处理,进而实现相关识别。实验研究了基于光折变晶体的传统联合变换相关器(PCJTC)和二元联合变换相关器(PBJTC)的识别能力。结果表明,PCJTC有着一定的相关识别能力,可实现靶物和参考物的相关识别,而采用了联合变换功率谱二值化后的PBJTC,其一级相关峰值较强而尖锐,中心直流零级峰相对减弱,并且有更大的相关信噪比,因此其相关识别能力也得到了一定的提高。
A photorefractive joint transform correlator for optical pattern recognition is presented. In the experimental implementation, a photorefractive Ce.. KNSBN crystal is used as the dynamic holographic medium to record the joint power spectra of the reference image and target image. It was experimentally shown that the correlator successfully performs the pattern recognition. And compared with the photorefractive classical correlator, the photorefractive binary joint transform correlator provides narrower half-width of the first-order level correlation peaks and higher discrimination ability.
出处
《人工晶体学报》
EI
CAS
CSCD
北大核心
2007年第6期1266-1269,共4页
Journal of Synthetic Crystals
关键词
光折变晶体
联合变换相关
联合变换功率谱
二值化
photorefractive crystal
joint transform correlator
joint transform power spectrum
binary
