Feature-level image fusion for SAR and optical images

In this paper, a feature-level fusion is implemented on optical and SAR images to obtain a better result of target detection. Discrete Canny algorithm is employed for extracting edges in optical image and the modified ratio algorithm is implemented for extracting the edges in SAR image. Based on contours detected from both images, the feature-level image fusion technique is proposed. This method used affinity value between contours for implementing contour correlation. The experimental results demonstrated the effectiveness of proposed feature based fusion technique.