Infrared Image Adaptive Denoising Algorithm in Contourlet Transform with Sharp Frequency Localization

Infrared image is usually disturbed by various noises during capturing and transporting, so it should be firstly processed by denoising. A new adaptive denoising algorithm based on a new Contour let Transform with Sharp Frequency Localization is proposed in this paper. The hierarchical adaptive denoising threshold of new Contour let coefficient is firstly estimated respectively by each location from different direction, then the noisy image is denoising with soft threshold related to the transform scale and direction. In order to reduce significant amount of aliasing components which are located far away from the desired support because of the new Contour let Transform, cycle spinning is employed to overcome the lack of translation invariance property and suppress pseudo-Gibbs phenomena around singularities of denoising image. Numerical experiments on infrared noisy image show that the algorithm given by this paper is significantly superior to some other usual arithmetic based on contour let, which can get better PSNR and visual quality.