Spatio-temporal segmentation of moving objects using edge features in infrared videos

Saliency detection using mutual consistency-guided spatial cues combination

“…While luminance is the most widely used attribute given to its high information content for infrared images, another attribute, namely contour, is another important visual characteristic for saliency map computation in infrared images, for it can make a region visually unique and attractive to an observer's attention when compared to the rest of the image [24][25][26][27]. Due to its unique attributes, we wish to take advantage of contour as a complement to luminance contrast to construct the contour saliency map, so as to further distinguish between the salient and non-salient regions in infrared images.…”