Abstract: Image segmentation using active contour models to improve image processing enhances object detection. Various segmentation methods have been proposed over in the past to improve the accuracy of segmentation results such as clustering, edge-based, region-based, template matching and hybrid methods. However, the image segmentation results of these methods are not ideal. Therefore, a small improvement in the results will have a huge impact on image processing, particularly for autonomous unmanned aircraft application. Recently, the Chan-Vese Model, a region-based method that uses active contour models, gained considerable research attention because of its improved image segmentation capability. This study presents a model that enhances the Chan-Vese algorithm model. The main idea of the proposed method is to reduce the computational time in image segmentation without affecting the segmentation result. Fitting term is defined as constant in the proposed model and the level set equation of the main domain continues to evolve the curve toward the boundary of the object. A total of 467 images from the Berkeley segmentation database are used to test the proposed method and analyze its performance. Results indicate that the proposed model achieves better segmentation result with low computational time compared with existing image segmentation methods
Ooi Qun Wong and Parvathy Rajendran, 2019. Image Segmentation using Modified Region-Based Active Contour Model. Journal of Engineering and Applied Sciences, 14: 5710-5718.