An Image Matching and Fusion Algorithm Based on Improved SIFT
Image matching and fusion are the core processes of image stitching,affecting both the accuracy and naturalness of the stitched images.The Scale Invariant Feature Transform(SIFT)has been widely applied in the field of image processing due to its rotational invariance and robustness.However,it can lead to issues such as time-consuming feature extraction and mismatches.To address these issues,a new method combining SIFT and AGAST algorithms is proposed.Feature points are refined through KNN and RANSAC,and the fusion and stitching are completed using an improved weighted average method.The results demonstrate that this method significantly enhances the efficiency of feature extraction and matching,achieving better image fusion effects.
万方数据
维普
