一种基于局部投影熵的图像匹配新算法

将图像熵和投影特征的概念结合引入到图像匹配中,定义了图像局部投影熵,并提出一种基于局部投影熵的图像匹配方法.基于局部投影熵的图像匹配方法具有较好的抗几何失真能力,并且提高了抗噪能力和在强光照条件下的匹配能力.结合采用分块、序贯检测及分层搜索等技术,进一步减少了计算量,实验结果表明这是一种简单而行之有效的图像匹配方法.

Image Matching Based on Local Projection Entropy

The information entropy that has been introduced to image matching is combined with the projection features and a image local projection entropy is defined. An image matching method based on the local projection entropy is proposed. This image matching method has very good anti-geometrical distortion properties, and the anti-noise capability is raised and good matching result can be got in high light conditions. In additional, some techniques are adopted: dividing image to some blocks, series testing and multi-resolution searching techniques. These techniques have led to the raising computing effect. Experimental results show that this algorithm can work effectively.