Two improved methods of SIFT algorithm combined with harris
Shao Wei, Li Na, Sun Lijuan, Sui Shulin, Xiangpeng Liu · 2012
Due to the invariance of scale, rotation and illumination, SIFT algorithm is commonly used in image matching. According to Harris algorithm, two simplified methods of SIFT algorithm are developed to improve real-time efficiency and robustness of feature detection. In the stage of accurate keypoints localization, two improved methods are proposed to eliminate edge responses. One is the replacement of Hessian matrix by the local autocorrelation function of Harris; the other is using the measure of corner response function of the Hessian matrix. The experimental results show that the improved methods can increase the performance of SIFT algorithm.