Fuzzy Integral for Moving Object Detection

Detection of moving objects is the first step in many applications using video sequences like video-surveillance, optical motion capture and multimedia application. The process mainly used is the background subtraction which one key step is the foreground detection. The goal is to classify pixels of the current image as foreground or background. Some critical situations as shadows, illumination variations can occur in the scene and generate a false classification of image pixels. To deal with the uncertainty in the classification issue, we propose to use the Choquet integral as aggregation operator. Experiments on data sets in video surveillance have shown a robustness of the proposed method against some critical situations when fusing color and texture features. Different color spaces have been tested to improve the insensitivity of the detection to the illumination changes. Then, the algorithm has been compared with another fuzzy approach based on the Sugeno integral and has proved its robustness.

Experimental Results

We have applied our algorithm to different datasets: the first one is our Aqu@theque dataset used in a multimedia application, where the output images are 384×288 pixel. The second and the thirth ones are PETS 2000 and the PETS 2006 dataset applied in video surveillance. The output images of these two last datasets are respectively 768 × 576 and 720×576 pixels. For each datasets, we provide a comparison with another approach based on Sugeno integral [1]. The results are obtained without post processing and the threshold for each algorithm is optimized to give the best results.

Fig.1. shows results obtained on Aqu@theque dataset. The results obtained using the proposed method are better than using the method proposed by [1] with the same color space, i.e. Ohta. The results obtained with the Choquet integral using other color spaces, i.e. the HSV and YCrCb confirmed that optimum results are obtained using Choquet integral with the YCrCb color features.

Fig. 1. First row: The current image, the ground truth. Second row: Sugeno-Ohta, Choquet-Ohta. Third row: Choquet-HSV and Choquet-YCrCb.

[1] H. Zhang, D. Xu, “Fusing Color and Texture Features for Background Model", International Conference on Fuzzy Systems and Knowledge Discovery, Volume 4223, No. 7, pages 887-893, September 2006.

Publication

F. El Baf, T. Bouwmans, B. Vachon, “Fuzzy Foreground Detection for Infrared Videos”, Joint IEEE International Workshop on Object Tracking and Classification in and Beyond the Visible Spectrum, OTCBVS 2008, pages 1-6, Anchorage, Alaska, USA, 27 June 2008. [pdf]

F. El Baf, T. Bouwmans, B. Vachon, “Fuzzy Integral for Moving Object Detection”, IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2008, pages 1729-1736, Hong-Kong, China, 1-6 June 2008. [pdf]

F. El Baf, T. Bouwmans, B. Vachon, “Foreground Detection using the Choquet Integral”, International Workshop on Image Analysis for Multimedia Interactive Services, WIAMIS 2008, pages 187-190, Klagenfurt, Austria, May 2008. [pdf]