基于CCD尺寸测量中图像边缘退化及恢复技术研究  

Restoration of degraded edge in size measurement based on CCD image

作　　者：刘宾[1] 王黎明[1] 赵霞[1] 

机构地区：[1]中北大学电子测试技术国家重点实验室,山西太原030051

出　　处：《应用光学》2013年第6期995-999,共5页Journal of Applied Optics

基　　金：973计划项目(2011CB311804);国家自然基金项目(61071193);山西省高等学校优秀创新团队支持计划资助项目

摘　　要：针对直边衍射效应造成的图像边缘退化问题,建立边缘退化模型。从菲涅尔直边衍射理论出发,基于信号与系统理论的图像边缘退化数学描述方法,分析了造成图像边缘退化的原因。通过分析不同光源波长和成像物距下直边衍射强度分布曲线的特点,构造简单函数实现衍射光强分布的近似,进而得到退化系统传递函数;利用退化系统传递函数构造滤波器对获取图像的边缘进行恢复,从而提高尺寸测量精度。实验结果表明:经边缘恢复方法校正后的测量结果误差约为0.02mm,相对未校正数据的测量结果误差减小0.04mm,提高了尺寸测量精度。For image edge degradation problems caused by straight edge diffraction, an edge degradation model was established. Starting from Fresnel diffraction theory, reasons for image edge degradation were discussed based on mathematical description of image edge degradation of signal and system theory. By analyzing the characteristic of intensity distribution curve of straight edge dif- fraction in case of different wavelengths of light source and distances from image and object, a simple function was used to represent distribution of straight edge diffraction, and the transfer function of the degradation system was obtained. Then the transfer function of the degradation system was used to realize the restoration of the degraded edge. Experimental results indicate that the presented method has better effects on restoration of the degraded edge caused by straight edge diffraction. The measurement error based on corrected data is 0.02 mm, which reduces 0.04 mm comparing to the measurement error based on original data.

关 键 词：尺寸测量 直边衍射 边缘退化 边缘恢复 

分 类 号：TN913[电子电信—通信与信息系统] TP394.1[电子电信—信息与通信工程]