关键词: 颜色传递, 尺度不变特征（SIFT）, 特征匹配, 消除误匹配点

Abstract: This paper investigates the matching problem between two color images coming from the same scene under different illuminations, and proposes a new image matching algorithm with the Scale Invariant Feature Transform（SIFT）  based on global color transfer. Firstly, the new algorithm implements the global color transfer between two color images which have different illumination under the same scene, resulting in reducing the error caused by color difference. Then, the SIFT algorithm is used to extract the features information from the processed images and the primary matching between the source image and the processed one is finished. Lastly, the false matched features are discarded by using the Random Sample Consensus（RANSAC） algorithm. Simulation results indicate that the new method has better performance of color images’ matching than that in the literatures.

Key words: color transfer, Scale Invariant Feature Transform（SIFT）, feature matching, eliminate false matching points