Multiresolution EM Algorithm

II. Data Modeling Image segmentation is an essential step in many advanced imaging applications. Accurate segmentation is required for volume measurements, medical diagnosis, image guided procedures, and 3D rendering. In this paper, we have developed a new multiresolution algorithm that extends the well-known expectation maximization (EM) algorithm. The proposed algorithm is based on generating scaled versions of the original image using wavelet analysis. We used synthetic images for tests and produced the results in confusion matrix so that it is easy to be understood, which proved more accurate segmentation than EM algorithm. Moreover we have applied our algorithm on real magnetic resonance image (MRI) of the human brain. A set of observed data X={xi | I=1,... N} can be modeled as to be generated from a mixture of random processes X1, X2, ... XK, with joint probability distribution . It is usually assumed that the processes represent independent identically distributed random variables, then we can write ) ,... , ( 2 1 K X X X f ) θ , f( ) ,... , ( 1 2 1 ∏ = = K k k K x X X X f