基于点扩散函数的理想亚像素边沿生成方法

On Method of Producing Ideal Sub-pixel Edge Based on PSF

针对数字成像系统无法提供精确亚像素精度图像的现状,提出了一种生成理想图像的方法,解决了亚像素定位算法评价过程中需要的高精度图像源问题.首先根据光学成像系统模型CCD采样理论以及圆孔衍射理论建立以二维高斯函数为核的点扩散函数(PSF);然后生成待测试的模拟量的理想边沿图像,将各个像素分割成N*N个子像素,以几何方法计算出各数字化的子像素;再将PSF作用于该图像得到确定边沿的数字理想图像.实验研究表明,与邻域平均方法相比,所提方法生成图像边沿过度平滑更接近实际物理成像.

Objective evaluation of subpixel algorithm accuracy and noise immunity requires sub-pixel accurate images,and digital optics imaging system is unable to provide accurate sub-pixel precision images.To solve this problem,the paper proposes a method of producing ideal image.Firstly,PSF has been established according to the principle of optics imaging system,CCD＇s sampling principle and Diffraction Theory.And then,a digitalized sub-pixel image of continues ideal image to be tested has been generated by dividing each pixel into N＊N sub-pixels and each sub-pixel＇s value calculated by geometry method.Finally,the target ideal image with precision edge has been obtained by convolution PSF and the sub-pixel image.Experimental studies show that,compared with the average algorithm,field over the edge of the proposed method to generate an image closer to the actual physical image.