结合Fourier域约束的多帧盲解卷积图像恢复算法  

Multi-Frame Blind Deconvolution Algorithm Based on Fourier-Constrained

作　　者：王洛飞[1] 耿则勋[1] 魏小峰[1] 娄博[1] 刘衷瑞[1] 

机构地区：[1]信息工程大学测绘学院,河南郑州450052

出　　处：《测绘科学技术学报》2011年第4期283-287,共5页Journal of Geomatics Science and Technology

基　　金：国家自然科学基金资助项目(60778051)

摘　　要：大气湍流严重限制了地对空光学系统的成像质量,自适应光学系统对大气湍流的补偿仅仅是部分的、有限的。当前的图像恢复技术对重建过程中的PSF施加的先验约束主要有非负性、归一化、支持域等空间域约束。这里对图像的Fourier频谱特性进行分析,将一种基于Fourier域的先验约束引入到多帧迭代盲解卷积算法中。实验结果表明,该算法可有效对模糊图像进行恢复,尤其是对强湍流观测下的重度模糊图像具有较好的恢复效果。The imaging quality of ground-based observations of space objects is severely limited by atmosphere turbulence. The effects of the turbulence can only partially be compensated by adaptive optics. The prior constraints used in the current image restoration techniques include non-negative constraints, normalization and support region constraints. The Fourier spectrum of image was analyzed and then a multi-frame iterative blind deconvolution algorithm based on the Fourier constraints was proposed. The results of the experiment showed that the blurred images, especially the strong atmosphere turbulence degraded images can be restored effectively by this algorithm.

关 键 词：盲解卷积 大气湍流 先验约束 频谱缺陷 支持域 

分 类 号：P237[天文地球—摄影测量与遥感]