任意形状曲线刃边的点扩散函数估计方法

遥感图像的点扩散函数估计是降低光学模糊、提高质量的必要前提。倾斜直线刃边法受到靶标极大的限制。现有的曲线刃边估计法虽然克服了倾斜刃边法只能适用于直线刃边的缺点,但是会因坐标拉伸导致估计值误差偏大。利用移动窗口的思想,提出了一种基于投影法的任意形状曲线刃边法,并在理论上说明了可行性。试验验证过程中先进行线性拟合刀刃边缘点,再对选定窗口内一定行数或列数的灰度值运用投影法采样,并对不同采样窗口的采样中心进行对齐处理,剔除不适合的样本点后插值和重采样,最终得到估计的点扩散函数。对于曲率在0.001～0.01的刃边,在较强模糊下,峰值信噪比依旧能保持在35 dB以上,测得点扩散函数峰值的误差可以控制在20%以内。相比传统刃边法和曲线拟合法,峰值信噪比平均能提高10 dB以上,且具有一定的抗噪性。

An algorithm for the estimation of point spread function based on curve edge of arbitrary shape

The estimation of point spread function of remote sensing images is the necessary premise to improve the quality and to reduce the optical blur. The oblique straight edge method is greatly limited by the target. Though existing curve edge methods overcome the drawback that the oblique edge methods can only be applied to the straight edges, but will result in large estimation value error because of the stretching coordinate. An curve edge method of arbitrary shape based on the edge method of projection is proposed using the idea of moving windows. And the feasibility is explained theoretically. In the process of experimental verification, the proposed method first fits the edge points linearly, then uses the projection method to sample the gray values of the rows or columns in a selected window and aligned the sampling centers of different windows, finally estimates the point spread function using interpolation and resampling after eliminating the improper sampling points. For the edges with curvatures from 0.001 to 0.01, the peak signal-to-noise ratio can still stay above 35 dB even though under strong blur. The error of peak of the point spread function can be controlled within 20%. Compared with the traditional edge method and curve fitting method, our method has better results for all kinds of curves that the average peak signal-to-noise ratio can be improved by more than 10 dB, and has somewhat anti-noise performance.