Model - Based Segmentation of

RICHARD D. HOLBEN Ford Aerospace & Communications Corp. exchange of heat between the target boundary surface and the immediate surroundings (background). The loss of contrast between the target and background with range to the observer is mainly caused by atmospheric scatter and absorption [5]. As a result, the boundary of the target in the image is not very sharp and many times there is an intensity gradient across the boundary. In the automatic recognition of tactical targets in FLIR images, an accurate and precise representation of the boundary of the targets is desired [3, 71. It is very important since the features used in the classification of the target are normally We present the use of gray scale intensities together with the based On the and gray Of the segmented edge information present in a fomard-looking infrared (FLIR) target and therefore, the performance Of a or a structural classifier critically depends on the results of segmentation [2]. Generally, only the gray scale of the image is used to extract the target from the background. Thus, the resulting segmentation image to obtain a precise and accurate segmentation of a target A