The video surveillance market is increasingly moving toward cheaper, efficient, portable, and high resolution systems. A typical video surveillance system consists of several cameras, which issue warnings or initiate smart reactions according to the analysis results of the captured video data. Many algorithms in video surveillance systems assume fixed lighting conditions for the monitored area. The performance of these algorithms is severely affected by the illumination changes of the monitored area. This paper presents an efficient intensity correction algorithm for high definition video surveillance applications. The new algorithm corrects both the global and the local intensity changes. It uses an apparent gain factor to correct the global intensity changes. In addition it corrects the local intensity changes using the local intensity mean and standard deviation. The proposed algorithm shows a promising performance when compared with other intensity correction algorithms. It has a low computational cost that makes it a suitable choice for real-time hardware implementation. This paper also presents a hardware implementation of the proposed algorithm using Xilinx Spartan3A digital signal processor (DSP) XC3SD3400A device. The targeted resolution is 1920 × 1080 at 30 f/s. The hardware prototype utilizes 17% of the slices, 21% of the block random access memories, and 24% of the DSP48As available in Spartan3A DSP XC3SD3400A device.

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.

An acquisition method of long pseudo-noise (PN) sequences in direct-sequence spread spectrum (DS/SS) systems is presented. The concurrent cross correlation of locally generated auxiliary signals with the incoming PN sequence produces binary search estimates leading to an initial approximation of the input signal's PN offset. The PN receiver then uses a closed-loop subsystem to determine the true offset of the incoming PN signal. New phase estimates are generated from successive corrections of likely errors in the binary search. The acquisition scheme's average acquisition time and its variance are analytically determined. Simulation results show that the proposed acquisition scheme is faster than the conventional serial receiver. A hybrid serial/parallel PN acquisition receiver is used for comparing the acquisition performance given similar hardware complexity. The hybrid receiver used the same number of parallel correlators as used in the proposed receiver's input phase estimator. Simulation results show that the proposed system provides an improved performance under normal operating conditions.

Direct sequence spread spectrum systems require a local replica of the pseudonoise (PN) sequence to be synchronized with the received signal's spreading code. A method is proposed to improve the PN sequence acquisition time and decrease the acquisition time variability. The proposed method is based on a binary search algorithm. At each stage of the algorithm, an auxiliary signal is used to formulate a new search region within the PN sequence, and at each stage the search region is reduced by half. The region containing the incoming PN phase is therefore obtained within O(log/sub 2/ (N+1)) steps. A closed loop coherent PN acquisition algorithm presented by Salih and Tantaratana (1996) motivated the use of the auxiliary signal, and the algorithm of Salih et al. is also used for comparison. Results indicate the algorithm presented in this paper can achieve a factor of two to three times improvement in the mean acquisition time with a significant decrease in the acquisition time variability.

Nonlinear filtering has found many practical applications in digital signal and image processing. The computation complexity of these filtering algorithms make them difficult for real-time hardware implementation. One of these nonlinear filters, which is based on fuzzy classification of each pixel to subgroups of its neighboring pixels, is considered hen for hardware implementation. The criteria of this filter are based on the local context which form the basis of the fuzzy rule. The filtering algorithm is slightly modified for implementation into a Xilinx Virtex series of FPGA for real-time processing of image sequences. Implementation details and recommendations for further improvement are discussed. Result of a simulation example from the proposed hardware implementation is also presented.

Filtering noise in real-time image sequences is required in some applications like medical imaging. The optimum approach in this case is in the form of adaptive 3-D spatial-temporal filter, which is generally very complex and prohibitive for real-time implementation. Independent processing of the image sequences, in spatial and temporal domains can resolve some of these implementation difficulties. Some of the existing spatial filters can easily be modified for real-time implementation. Adaptive temporal filters, however, are more involved. In this paper, an adaptive temporal filter is proposed that lend itself to hardware implementation for real-time temporal processing of image sequences. The proposed algorithm is based on adaptive Kalman filtering which is relatively simple and effective in its performance. Adaptation in this case is with respect to motion in the image sequence as well as variation of noise statistics. An efficient hardware implementation of this algorithm, based on FPGA technology, is proposed.