Wavelet-based enhancement of color image using combined local variance and entropy analysis

This paper presents a novel wavelet-based image enhancement approach to reduce noise in a single-sensor color camera. Most traditional noise reduction methods tend to over-suppress high-frequency details. For overcoming this problem we first decompose the input image into flat and edge regions, and adaptively remove noise by multiplying alpha map and wavelet transform coefficients of LH, HL, and HH bands. After removing noise in the flat region, we further remove noise in edge regions by adaptive shrink of wavelet coefficients based on the entropy. Experimental result show the proposed algorithm is proved to be able to reduce noise without losing sharp details and is suitable for commercial low-cost such as digital cameras, CCTV, and surveillance systems.