A Color Distance Model Based on Visual Recognition

In computer vision, Euclidean Distance is generally used to measure the color distance between two colors. And how to deal with illumination change is still an important research topic. However, our evaluation results demonstrate that Euclidean Distance does not perform well under illumination change. Since human eyes can recognize similar or irrelevant colors under illumination change, a novel color distance model based on visual recognition is proposed. First, we find that various colors are distributed complexly in color spaces. We propose to divide the HSV space into three less complex subspaces, and study their specific distance models. Then a novel hue distance is modeled based on visual recognition, and the chromatic distance model is proposed in line with our visual color distance principles. Finally, the gray distance model and the dark distance model are studied according to the natures of their subspaces, respectively. Experimental results show that the proposed model outperforms Euclidean Distance and the related methods and achieves a good distance measure against illumination change. In addition, the proposed model obtains good performance for matching patches of pedestrian images. The proposed model can be applied to image segmentation, pedestrian reidentification, visual tracking, and patch or superpixel-based tasks.