Low complexity contrast enhancement algorithm for nighttime visual surveillance

This paper presents a new algorithm for nighttime contrast enhancement. The proposed algorithm modifies the traditional histogram equalization algorithm to maintain the color information of the original nighttime images. The algorithm has a low computational cost that makes it suitable for real-time hardware implementation. In addition, its efficient hardware implementation is detailed on a Xilinx Spartan3A DSP XC3SD3400A device. The targeted resolution is 1920×1080@30fps. The hardware prototype utilizes 6.45% of the slices, 1.59% of the Block RAMs, and 7.14% of the DSP48As available in Spartan3A DSP XC3SD3400A device.

[1]  Tomi Räty,et al.  Survey on Contemporary Remote Surveillance Systems for Public Safety , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[2]  Takeo Kanade,et al.  Denighting: Enhancement of nighttime images for a surveillance camera , 2008, 2008 19th International Conference on Pattern Recognition.

[3]  Leonard McMillan,et al.  Video enhancement using per-pixel virtual exposures , 2005, ACM Trans. Graph..

[4]  Dmitry B. Goldgof,et al.  Understanding Transit Scenes: A Survey on Human Behavior-Recognition Algorithms , 2010, IEEE Transactions on Intelligent Transportation Systems.

[5]  Lucia Maddalena,et al.  A Self-Organizing Approach to Background Subtraction for Visual Surveillance Applications , 2008, IEEE Transactions on Image Processing.

[6]  Tieniu Tan,et al.  A survey on visual surveillance of object motion and behaviors , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[7]  Tieniu Tan,et al.  Context Enhancement of Nighttime Surveillance by Image Fusion , 2006, 18th International Conference on Pattern Recognition (ICPR'06).