Image Denoising and Enhancement with Evolutionary Programming in Wavelet Domain

Abstract

Unmanned Aerial Vehicles (UAV) digital images are often badly degraded by noise during dynamic acquisition and transmission process. Denoising is very important and difficult for UAV-vision Guided, because natural scene image is complicated and having lots of the edges and texture details. The image denoising and enhancement algorithm based on adaptive dual-tree discrete wavelet packets (ADDWP) which combine the dual-tree discrete wavelet packets and evolutionary programming is proposed in this paper. In order to reduce the noise and enhance detail for a UAV image by the EP (evolutionary programming) in wavelet domain. Firstly, The de-noising threshold is estimated by EP in the wavelet domain. Secondly, Noise is reduced in the fine high-frequency sub-bands of each decomposition level, respectively, so that the maximum signal-noise ratio can be obtained in the high-frequency sub bands, respectively. Thirdly, The enhancement parameters is estimated by EP in the wavelet domain, and the detail is adaptive enhanced by an improved non-linear gain operator in the coarse high-frequency sub bands of each decomposition level, respectively. The experiment result improves our algorithm is efficient.