Infrared Image Segmentation via Fast Fuzzy C-Means with Spatial Information

Forward looking infrared (FLIR) image segmentation is crucial for automatic target recognition (ATR). This paper presents a new clustering algorithm for FLIR image segmentation. We combine the standard fuzzy c-means algorithm (FCM) with two-dimensional histogram of the image, and modify the membership function of the FCM for taking into account the spatial information of image data. Since the FCM is computationally cumbersome, we also present a fast and efficient implementation of the proposed method. Experiments show that the proposed method can segment the infrared image properly and fast, also has the better robust performance to noise