An Improved Active Contour Model for Medical Images Segmentation
Mohammed Deriche, Sami El Ferik, شريف محمد صالح الشريف
كلية علوم وهندسة الحاسب الآلي-جامعة الملك فهد للبترول والمعادن · السعودية
Image Segmentation is an image processing technique which is used to subdivide an image into regions in which each one contains components having similar properties or characteristics. Its goal is to “simplify and/or change the representation of an image into something that is more meaningful and easier to analyze”. This approach has many applications. One of its most important applications is the extraction of tumor areas from medical images as a first step in the therapy planning of cancer patients. This objective can be achieved by many image segmentation approaches. One of these successful techniques is the active contour (Snake); its idea is based on a flexible curve (or surface) which is dynamically adapted to required edges or objects in an image. In this work an improvement over a current active contour model is proposed. The new model was tested using real CT images and has given promising results. MASTER OF SCIENCE DEGREE King Fahd University for Petroleum & Minerals, Dhahran - Saudi Arabia JANUARY 2010