Infrared image segmentation using Enhanced Fuzzy C-means clustering for automatic detection systems

This paper proposes Enhanced Fuzzy C-means technique (EFCM) based infrared image segmentation and its broad application in Automatic detection systems. The EFCM based image segmentation is able to approximate the exact number of clusters present in the image. EFCM based segmentation is applied on various infrared images that can be used for automatic detection systems and compared with widely used clustering techniques such as K-means and EM. Clustering performance has been compared in terms of well-proven and widely accepted validation indices, Global Silhouette Index and Separation Index. The segments or clusters obtained from above mentioned clustering methods have been assessed visually. Automatic Detection Systems based on EFCM can help in reducing complexities present in conventional systems.