Image Mosaic Based on Improved Logarithmic Polar Coordinate Transformation and Ransac Algorithm

Abstract

Complex factors with the electronic noise, X-ray scattering and uneven illumination often disturb the image registration. A new algorithm was proposed in this paper. The improved phase correlation algorithm based on log polar transformation was used to calculate parameters, such as rotation, scaling and translation. Then, the Harris corner matching points were extracted in overlapping positions and purified by the improved Ransac algorithm. Finally, images were processed by NSCT transform algorithm to make the image joint seemed smooth and natural. Experiments confirmed that, the new algorithm is accurate and efficient, and has high robustness to complex environment.