TEM Color Image Segmentation using Hill Climbing Algorithm

TEM image is rapidly gaining prominence in various fields. Since high resolution TEM image captures huge amount of information, it is important to understand image segmentation techniques on it. The goal of image segmentation is to cluster pixels into salient image regions, i.e., regions corresponding to individual surfaces, objects, or natural parts of objects. To have region based image retrieval features can be mapped to various feature spaces on the basis of grouping algorithms. Using low level features of luminance and color, some salient features can be determined. This paper makes use of Hill Climbing algorithm on nanoscopic TEM image to obtain some low level features. With the same size and resolution of input image, it gives high quality saliency maps.