A novel enhancement algorithm for low-illumination images

For the image lack of illumination or uniformity, the image enhanced by histogram equalization is reducing gray scale for contrast, Leading to some image details disappeared. Wavelet analysis, with its local excellence both in space and frequency domain and its multi-resolution characteristic, can be focused to any details of the analysis object by adopting for high frequency fine gradually temporal or spatial step length. This paper adopts the method of wavelet transform , multi-scale decomposition and reconstruction of the original image, and constructs low-pass, high-pass filter for filtering processing, gets the wavelet transform coefficient matrix, respectively different frequency characteristics, with different frequency characteristics of low frequency and low frequency, high frequency and high frequency by four blocks of different linear transformation, and then wavelet inverse transformation to achieve the goal of image enhancement. The experimental results show that the algorithm not only enhances image contrast but also reserves more scene details. It is a better enhancement algorithm for low- illumination image.