Enhancing object quality based on saliency map and derivatives on color distances

In recent years, computers have become more and more important in human life and work. People used computers to control highway, traffic violation, etc. These jobs need process input images to detect interesting objects. This step is important in many computer vision applications such as image segmentation, object recognition, etc. There are a lot of methods to solve this problem. However, most of output images from them need enhance quality, and color change at object contour. In this paper, we propose a method for enhancing object quality. The proposed method uses saliency map based on global contrast and derivative on color distance. The proposed method is simple to know, easy to implement and efficient to apply. The results of the proposed method are better than those of the other methods at the saliency map quality when evaluated by using a large public dataset. We can control masks, and the extracted object quality by using a derivative operator on color distances and this idea brings the results as expected.