Block Wiener-based image registration for moving target indication

作者:

Highlights:

摘要

This paper develops a block Wiener-based image registration method to align consecutive frames in a video sequence. The main purpose of this registration technique is to serve as a processing step in a moving target indication (MTI) algorithm, where subsequent steps include frame-to-frame differencing and thresholding. Experiments show that the block Wiener-based approach is computationally efficient and effective in terms of facilitating a MTI. These experiments compare the block Wiener-based against traditional global parametric registration, pixel-based Wiener registration, block-based motion compensation, and state-of-the-art phase-based registration.

论文关键词:Image registration,Wiener filtering,Least mean-squared estimation,Moving target indication

论文评审过程:Received 19 March 2007, Revised 8 July 2008, Accepted 18 July 2008, Available online 8 August 2008.

论文官网地址:https://doi.org/10.1016/j.imavis.2008.07.008