关键词: 相似性度量, 特征点配对, 随机抽样一致性（RANSAC）, 图像拼接, 城市距离, 脊柱图像

Abstract: According to the limited vision of spine image, this paper presents a method of automatic spine image moasic algorithm, which based on optimized MSFE（Multi-Scale Feature Extraction）. This algorithm devises two improved ways. One way is presented that the size of template can automatically adjust along with the scale factor changing, the other is that a bidirectional feature points matching algorithm is proposed. City-distance is used to judge similarity. Random sample consensus is used to guarantee stability and decide the transformation parameters of the image stiching, and a weighted average is used in the step of image fusion. The mosaic results of multiple spine images show that the algorithm has good real-time performance and robustness.

Key words: similar measurement, matching of feature points, Random Sample Consensus（RANSAC）, image mosaic, city-distance, spine image