Tracking object using a TOF camera using the intensity and depth chanel

Background subtraction and movement segmentation algorithms can be improved by merging the depth and color entries. In this paper, we propose a background subtraction method based on the Gaussian mixture models using color and depth information. To combine color and distance information, we used a probabilistic model based on the histogram of the area where movement was detected. In particular, we focused on solving color camouflage problem and depth denoising. For evaluating our method, we built a new dataset containing normal, color camouflage and depth camouflage situations. The dataset files consist of color, depth and ground truth image sequences. The proposed method shows greater resistance to changes in lighting, shadows, reflections and camouflage. Thus, this technique will help to robustly detect regions of interest as pre-processing in high-level image processing stages.