一种抗尺度畸变的光学相关图像识别技术  

Optical correlation image recognition technology to resist scale distortion

作　　者：邵王君 华文深[1] 周中亮[1] 甘厚吉[1] 

机构地区：[1]军械工程学院光学与电子工程系,河北石家庄050003

出　　处：《应用光学》2008年第4期503-507,共5页Journal of Applied Optics

摘　　要：为提高光学相关图像识别的准确度和抗畸变能力,引入多重复合匹配滤波器概念,探讨一种新的识别思路。它融合了综合鉴别函数和K-L变换的经典思想,改善了单个匹配滤波器动态范围太小的状况,对图像尺度变换有较宽的适应范围,从而提高了光学模式识别系统的识别效率。仿真结果表明,对尺度缩小1/2的情况,采用此思路可使其滤波的刷新速度比常规方法提高4倍,从而验证了算法的有效性和可行性。A concept of multi-complex matched filter is introduced to improve the recognition of optical correlation image and the resistance of distortion, and a new way for the recognition is discussed. The synthetic discriminant function （SDF）was combined with K-L transform as a classical concept to effectively expand the dynamic range of single matched filter. Therefore, the scale distortion was restrained and the recognition efficiency of the optical pattern recognition （OPR） system was significantly improved. The simulation indicates that the concept can increase the update speed of filtering by a factor of 5, compared with that of the ordinary filtering, in the case of the distortion of half scale. The algorithm proved valid and feasible.

关 键 词：尺度畸变 综合鉴别函数 K—L变换 光学模式识别 多重复合匹配滤波器 

分 类 号：TN919.8[电子电信—通信与信息系统] TP391.4[电子电信—信息与通信工程]