Search

Moving object detection is a fundamental task on smart CCTV systems, as it provides a focal point for further investigation. In this study, an algorithm for moving object detection in video, which is thresholded using a stationary wavelet transform (SWT), is developed. In the detection steps, the authors perform a background subtraction algorithm; the obtained results are decomposed using discrete stationary wavelet transform 2D, and the coefficients are thresholded using Birge-Massart strategy. This leads to an efficient calculation method and system compared to existing traffic estimation methods.

In this paper an improved ultraspherical window is developed for designing quadrature mirror filters banks (QMF) with the help of a real coded genetic algorithm (RCGA). In fact, the ultraspherical window is modified by adding a parameter (α) which to improve the spectral parameters. Then, RCGA is used to find optimal values of the adjustment parameters, the side-lobes ratio of ultraspherical window and the cut-off frequency of the low-pass prototype filter. This latter, is used to derive all the filters of QMF banks. When the developed QMF banks are exploited for speech enhancement algorithm based on wavelets techniques it gives good performances in terms of Perceptual Evaluation of Speech Quality (PESQ) and Signal to Noise Ratio (SNR). In addition to this, when compared to Dolph-Chebyshev, Kaiser and the original Ultraspherical windows the modified Ultraspherical window gives better performance referred to the obtained simulation results.