Aiming at the shortcomings of the existing wedgelet compression arithmetics, a novel contour-representing algorithm based on wedgelets is presented in this paper. Firstly the input image is binarized and the most opti...Aiming at the shortcomings of the existing wedgelet compression arithmetics, a novel contour-representing algorithm based on wedgelets is presented in this paper. Firstly the input image is binarized and the most optimized wedgelets are found by means of quadtree framework. Then the contours are reconstructed by applying the wedgelets, the data volume is compressed, and the shortcomings of the contour representation based on normal wavelet are ameliorated, the better effect for the visualization is obtained, too.展开更多
Based on the analysis of the characteristics of synthetic aperture radar （SAR） images, a new edge detection method is proposed. The wedgelet transform is introduced into the area of SAR image speckle reduction for i...Based on the analysis of the characteristics of synthetic aperture radar （SAR） images, a new edge detection method is proposed. The wedgelet transform is introduced into the area of SAR image speckle reduction for it can provide a nearly optimal representation for images in the horizon class. The wedgelet filter has good ability in keeping edge and speckle reduction. Then, a ratio edge detector is applied after a process of speckle reduction. The experimental results show that the method outperforms substantially others visually.展开更多
基金The project was supported by the National Natural Science Foundation of China (60275012) and LiuHui Center for Ap-plied Mathematics , Nankai University & Tianjin University
文摘Aiming at the shortcomings of the existing wedgelet compression arithmetics, a novel contour-representing algorithm based on wedgelets is presented in this paper. Firstly the input image is binarized and the most optimized wedgelets are found by means of quadtree framework. Then the contours are reconstructed by applying the wedgelets, the data volume is compressed, and the shortcomings of the contour representation based on normal wavelet are ameliorated, the better effect for the visualization is obtained, too.
文摘Based on the analysis of the characteristics of synthetic aperture radar （SAR） images, a new edge detection method is proposed. The wedgelet transform is introduced into the area of SAR image speckle reduction for it can provide a nearly optimal representation for images in the horizon class. The wedgelet filter has good ability in keeping edge and speckle reduction. Then, a ratio edge detector is applied after a process of speckle reduction. The experimental results show that the method outperforms substantially others visually.
