FEATURE EXTRACTION FROM MEDICAL IMAGES USING IMAGE PROCESSING TECHNIQUES

Abstract

Processing and analysis of medical images are an important area of Biomedical Instrumentation. Sophisticated computer vision systems are able to interpret the result of image analysis and describe various objects and their relationship in the image. Detection and measurement of bone, tissue, tumor etc from medical images are possible by an image analysis system. This dissertation focuses on extracting important features like edges or discontinuities from medical images. Feature extraction is an essentially preliminary step in most automatic pictorial pattern recognition and scene analysis problems. Various image processing techniques are used for image enhancement before feature extraction. Preprocessing of input image includes point operations and histogram equalization. This work can be considered as a first step towards a computer vision system for medical image analysis. The results of feature extraction can be used as input for next stages like image segmentation and classification