Image Enhancement Technique Based on Human Visual Perception and Multi-exposure Fusion for a Landslide-prone Area Monitoring System

摘要

In this paper, we propose a new technique for enhancing image quality and generating a representative image from a set of input images taken from a landslide-prone area monitoring camera at different times of a day. Thus, less-visible areas in the input images are different from one another. First, the proposed technique enhances each input image by deploying a scaling function based on human visual perception. Then, it fuses all input images and all enhanced images by using Gaussian and Laplacian pyramid-based blending. Experimental results show that the resulting image can improve the visibility of some shadowed details and that the objective evaluation results regarding image enhancement metric, universal image quality index, and perceptual similarity index are satisfying.