Multiresolution signal and image models aim to capture the statistical structure of smooth and singular regions.Unfortunately

models based on orthogonal wavelet transform suffer from shift-variance

which makes them be less accurate and real-time.In this paper

we extend the local contextual hidden markov model(LCHMM)modeling framework to the complex wavelet transform and proposed a new model

called local contextual markov model based on complex wavelet(C-LCHMM)

which features near shift-invariance and improved angular resolution and can exploit the local statistics of wavelet coefficients at a low computational complexity.