PACS图像处理中SURF特征点提取算法实现

Implementation of SURF Feature Point Extraction Algorithm in PACS Image Processing

在PACS图像处理中经常会遇到分辨率和视野的矛盾问题。视野范围越大则分辨率越小,分辨率越高则视野范围越小。图像拼接技术可以有效地解决这个矛盾问题。图像拼接最终能否达到良好效果,最重要的一点就是选择一个鲁棒而快速的图像配准方法。首先介绍了图像拼接流程,然后详细分析比较了基于特征图像配准中不同特征点检测算法的优劣性。最后用C语言实现了SURF特征点检测算法,并用真实的PACS图像进行试验,验证了不同PACS图像特征点检测算法有效性。

In picture archiving ＆ communication system （PACS） image processing, the conflict problem between resolution and field of vision is often met. The greater is the field of vision, the smaller is the resolution. The problem can be solved by image stitching technology effectively. Choosing a robust and rapid image registration method is the most important thing for achieving good effect of image stitching. The image stitching process is introduced at first. Then the advantages and disadvantages of the detection algorithm for different feature points based on feature image registration are analyzed and compared in detail. The detection algorithm for speeded up robust features （SURF） point is realized through C language finally. And the real PACS images are used to test the effectiveness of detection algorithm for feature points in PACS images.