Uterus Image Segmentation Using Multi-feature EM Algorithm Based on Gabor Filter

Abstract

In this work the image is segmented effectively based on texture feature by reducing the noise. For effective image segmentation Expectation-Maximization (EM) algorithm based on Gabor filter is used. The EM algorithm is applied on 2D Ultrasonic image of uterus and tested. The Gabor function has been recognized by its multiresolution properties and the precision of locating the texture features in the spatial domain. The approach consists of two steps: finding an effective and discriminative set of texture features using Gabor filters and EM based image segmentation. The work is simulated using MAT LAB (Matrix laboratory).