| Review: |
Image processing has traditionally been based on Fourier and spectral analysis, but in the last few decades new methods have emerged, such as stochastic approaches based on Gibbs/Markov random fields and Bayesian inference theory, variational methods that incorporate various geometric regularities, linear or nonlinear partial differential equations, and applied harmonic analysis centred on wavelets. These apparently distinct approaches, are, in fact, intrinsically connected. They share common grounds and roots. This book aims to integrate these diverse approaches. The authors explore three key aspects of image processing and analysis: modelling, model analysis and computation and simulation. There are chapters on: image analysis tools; image modelling and representation; image denoising; image deblurring; image inpainting; and image segregation. |