Image Fusion And Airport Detection Based On Ridgelet Bi-frame

It is of great importance to find methods that can sparsely represent natural images for researchers in the fields of mathematics, computer vision and data compression. Motivated by the failure of wavelet, much work has been done to look for new systems superior to separable wavelet systems. Ridgelet theory was developed under such a background. A new image representation system referred to as ridgelet bi-frame is developed, which not only inherits the key idea of the orthonormal ridgelet, but also provides a more generalized notion. Apply the ridgelet bi-frame to linear feature detection, Image fusion and airport detection is the main work in this paper. The research results included in this article are as follows:The papers apply the ridgelet bi-frame and curvelet bi-frame to images fusion. Based on the directional properties of ridgelet bi-frame subband, a new method for anisotropic images fusion method is presented, which combines with local variance of subband. The method is respectively applied to the multifocus, the medical and the remote image fusions in the experiments. Experiments show that they have achieved satisfactory results, especially in edge. Considering the problems for linear feature detection, a novel one based on ridgelet bi-frame for localized linear feature detection is given. All kinds of experiments show that the linear feature detection method based on ridgelet bi-frame is able to accurately detect the direction and position of a straight line, and can effectively suppress noise. we propose a new airport detection method for SAR images based on ridgelet bi-frame. At First, treat a simple preprocessing to images, and then extract straight lines form edge images, finally line tracking and binarization are employed. The experimental results show that the new algorithm is speedy and simple, meanwhile, it does not require complicated post-processing.