Comparative Analysis and Evaluation of Image Segmentation Algorithms

Abstract - With the rapid developments of higher resolution imaging systems, larger image data are produced. To process the increasing image data with conventional methods, the processing time increases tremendously. Image segmentation is emerging as a solution for computer vision and image processing. With the help of several image processing algorithms efficiency of segmentation can be improved, and it is widely used in medical imaging (i.e. find tumor in MRI), robotic vision (i.e. vision-based navigation), and face recognition. New faster image processing techniques are needed with their complete database including algorithms detail, their shortcoming with expected solution and implementation, to keep up with the ever increasing image data size. The focus of our study is Watershed and Clustering algorithm with their modified version to get better result. Watershed and K-means algorithm are each considered for their speed, complexity, and utility. Implementation of each algorithm is then discussed. Finally, the experimental results of each algorithm are presented and discussed with quantitative and qualitative comparison.