Color image stitching based on color invariant and SURF
Gao Shengfen · Computer Engineering and Applications Journal · 2013
A method of color image stitching based on color invariant and SURF(Speeded Up Robust Feature)is proposed. The SURF feature points are extracted by the color invariable information calculated from color image in stead of gray information,and then to improve the accuracy of the transformation matrix, the mismatch points are removed in accordance with similarity transformation principle during the match process. To get a realistic seamless image, pixels of the image within the overlap region are smoothly transited by the brightness adjustment principle. Experimental results show that many more match points can be extracted in this method while retaining fast computational speed and high accuracy rather than traditional SURF algorithm.Stitching trace can be eliminated effectively, and also illumination robust has improved with the method mentioned above. It's proved that the algorithm proposed in this paper has certain practical value in the fields of images stitching and geological analysis.