Infrared Image Enhancement Based on Multiscale Bilateral Detail Decomposition

Abstract

Abstract. In order to solve the problem that infrared images usually have a poor visual effect with low contrast and weak details, an infrared image enhancement method based on multiscale bilateral detail decomposition (MBDD) is proposed in this paper. The core of MBDD strategy lies in utilizing bilateral filter with different scale parameters to achieve multiscale detail decomposition for infrared image. On the one hand, abundant detail information in different scales is effectively extracted as detail sub-images to be enhanced; on the other hand, a set of corresponding weights is constructed adaptively. The output image is a fused version from enhanced detail scale sub-images. Experimental results show that the proposed method outperforms other conventional methods for infrared image enhancement.