Undecimated double density wavelet transform based contrast enhancement technique using dynamic stochastic resonance

This paper describes an efficient and adaptive method for enhancing contrast of very dark grayscale and color images based on Undecimated Double Density Wavelet Transform (UDDWT) using Dynamic Stochastic Resonance (DSR). In the proposed UDDWT-DSR approach, the internal noise due to inadequate illumination of the image has been used for the enhancement purpose. Specifically the darkness is treated as a noise and is used to produce a noise-induced transition of the image from a low contrast to high contrast state. The proposed method first decomposes the given dark image into many sub bands using UDDWT. The stochastic resonance is induced in the approximation and detail coefficients of UDDWT transformed dark image in an iterative fashion. The DSR technique iteratively tunes the UDDWT coefficients using bi-stable system parameters. Computer simulations are carried out for both dark grayscale and color images to demonstrate the performance of the proposed method and the results are compared with existing contrast enhancement techniques. From the comparison it is observed that the proposed technique provides better performance in terms of Relative Contrast Enhancement Factor (F), Perceptual Quality Measures (PQM) and Color Enhancement Factor (CEF).