在低信噪比下实现快速图象匹配的准最佳组合相关方法

Suboptimal Combining Correlatin Methods for Fast Image Matching in Low SNR

本文首先导出了噪声中图象匹配的最佳算法,然后给出了两种在低SNR和短数据条件下实现两幅数字图象快速配准的组合相关方法。在实际地形图上的大量仿真匹配结果证明,这两种方法至少比最佳算法快两个数量级,精度也几乎达到最佳。满足了某些导航和精确制导问题中对实时配准的要求。

In this paper the optimal algorithm for image matching in noise is derived, and two combining correlation methods which can realize fast registration of two digital images in low SNR and short data are proposed. Monte-Carlo simulation to a real terrain image has shown that the two methods are two orders faster than the optimal algorithm at least and their precision are almost optimal. The two methods can satisfy the demands of real time matching insome navigation and guidance systems.