Recently, there is a growing interest in building a surveillance system which can run robustly under different environments, such as nighttime, rain, fog, sunshine, etc. In this project, we mainly concern the first situation. The night vision images are generally low quality images obtained by CCD cameras in surveillance applications. Carrying out detection, tracking and behavior understanding on nighttime video is a challenging problem because of the following reasons:
(1) Low contrast. We can not clearly extract moving objects from the dark background. Most color-based methods will fail on this matter if the color of the moving objects and that of the background are similar.
(2) Low signal to noise ratio. Signal to noise ratio is usually very low due to high ISO (ISO is the number indicating camera sensors sensitivity to light). Using a high ISO number can produce visible noise in digital photos. But low ISO number means less sensitivity to light.
(3) Low environment information. Enviroment information, in other words, the context information of a scene, affects the way people perceive and understand what has happened. Due to the limited information in nighttime video, understanding behavior of people in video becomes more difficult.
Solutions:
Make use of the high quality background of the day to help enhance the context of nighttime images.
Experimental Results:
Related Publications:
1. Yinghao Cai, Kaiqi Huang, Yunhong Wang and Tieniu Tan, “Context Enhancement of Nighttime Surveillance by Image Fusion”, Pattern Recognition, 2006. ICPR 2006. 18th International Conference on, vol.1, no., pp.980-983, 2006. (presentation).
2. A. Ilie, R. Raskar, and J. Yu. Gradient domain context enhancement for fixed cameras. In Proc. of ACCV. Jeju Island, Korea, January 2004.
3. R. Raskar, A. Ilie, and J. Yu. Image fusion for context enhancement and video surrealism. In Proc. of the 3rd international symposium on Non-photorealistic animation and rendering(NPAR), pages 85–152. Annecy, France, June 2004.
4. J. Li, S. Z.Li, Q. Pan, and T. Yang. Illumination and motionbased video enhancement for night surveillance. In Proc.of the 2nd Joint IEEE International Workshop on VS-PETS, pages 169–175. Beijing, China, October 2005.
