From Particle Mechanics to Pixel Dynamics: Utilizing Stochastic Resonance Principle for Biomedical Image Enhancement

摘要: There is a noteworthy analogy between the statistical mechanical systems and digital image processing systems. We can make pixel gray levels of an correspondence to discrete particles under thermodynamic noise (Brownian motion) that transits binary state transition from weaksignal strong-signal whereas noisy signal enhanced in imaging One such phenomenon physical stochastic resonance (SR) where gets by adding small amount mean-zero Gaussian noise. A local change made based upon current values pixels boundary elements immediate neighborhood. However, this random, generated sampling conditional probability distribution. These distributions are dependent on global control parameter called “temperature” (Geman & Geman, 1984). At low temperature coupling tighter means images appear more regular at higher induce loose neighboring appears or blurred image. particular optimum these comes much closer fashion similarly got arranged leads degradation further enhances signal. In chapter, we discuss application principle biomedical Some applications detection transmission, restoration, enhancement segmentation. Stochastic certain nonlinear which synchronization input occurs when optimal additional inserted into system (Gammaitoni et al., 1998). ubiquitous conspicuous phenomenon. The climatic model addressing apparently periodic occurrences ice ages weak, external was thought be first theoretical phenomenon, concept put forward (Benzi 1981). Since after discovery Benzi, there has been increasingly attracting various fields like physics 1998), (Anishchenko 1999), chemistry (Horsthemke Lefever, 2006), biology neurophysiology (Moss 2004), (Morse Evans,