Abstract
In this paper, we propose a novel model of sparse representation for image denoising that we call an adaptive contourlet hidden Markov model (HMM)-pulse-coupled neural network (PCNN). In this study, we first adopted a contourlet transform to decompose a noisy image to be some subband coefficients of various directions at various scales. The contourlet emulated extremely well the sparse representation performance of human visual perception, such as its multiscale characteristics, geometric features, and bandpass properties. Second, we used an HMM method to create a statistical model that expressed the coefficient relationships in intrabands, interbands, intrascales, and interscales. Then we used an expectation-maximization training algorithm to obtain the state probability. The result included the state, scale, and direction, the position of the coefficient, the noisy image, and the parameter set of the HMM model. Third, we put the state probability into the PCNN model, which could adaptively optimize the parameters of the HMM model and get better coefficients of clean images. Finally, we transformed the image denoising problem into a Bayesian posterior probability estimation problem. We also reconstructed a denoised image based on the clean coefficients obtained from our proposed method. The experimental results show that the contourlet HMM-PCNN model proposed in this paper is superior to the contourlet with hidden Markov tree model and the wavelet threshold method.
Highlights
Research indicates that most information acquired by humans comes from human vision, which is visually selective; it can perfectly distinguish small amounts of important information from a large amount of visual information
To make better use of biological visual perception characteristics for image noise removal, we propose an adaptive contourlet-hidden Markov model (HMM)–pulse-coupled neural network (PCNN) model of sparse representation based on a contourlet-HMT model developed by Po and Do [21]
To find a better sparse image representation method, we propose an approach that takes advantage of the contourlet-HMM–PCNN model to simulate the characteristics of the biological visual perception mechanism in the receptive field of the simple cells
Summary
Research indicates that most information acquired by humans comes from human vision, which is visually selective; it can perfectly distinguish small amounts of important information from a large amount of visual information. This is called sparse coding [1], [2]. Horace Barlow [3] has proposed an effective coding hypothesis: the human visual perception system can adapt to its environment because the visual perception cells can effectively filter out statistically redundant external signals. The system effectively represents the infinite information of nature by using limited neurons. Computer vision and neural computing, parallel computing, associative memory, and other intelligent information-processing methods have become much more
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
