基于细分小波与局部投影熵相结合的图像匹配算法

利用第二代小波-细分小波计算量小和其具有多分辨分析的特点，结合局部投影熵的优势，提出了一种基于细分小波与局部投影熵相结合的图像匹配算法．该算法首先利用立方B-样条细分小波对模板图像和实时图像进行多尺度分解；然后在粗尺度上利用局部投影熵进行匹配，确定细一层的候选匹配区域；重复这一过程，直到尺度为零时止，便得到最终匹配结果．通过仿真实验表明该算法可以有效地提高匹配速度，并具有一定抗噪能力和较高的匹配精度．

Image Matching Algorithm Based on Combining Subdivision Wavelet with Local Projection Entropy

In this paper an image matching algorithm based on combining subdivision wavelet with local projection entropy is presented, with using the feature of computational cost scanty of second generation wavelet-subdivision wavelet and the dominance of the local projection entropy. The programs of the algorithm are as follows: firstly, execute multiresolution analysis and decompose template image and real time image; secondly, according to local projection entropy match image of coarse layer and decide candidate matching area of finer layer. Repeating the process till the zero scale; finally gain the patching result. Experimental results show that the algorithm can effectively promote the speed of matching. It also possesses higher matching precision and the ability of resisting noise.