Résumé: Moving object detection (MOD) has gained significant attention for its application in advanced video surveillance tasks. Region-of-Interest (ROI) detection algorithms are essential prerequisites for various applications, ranging from video surveillance to adaptive video coding. The simplicity and efficiency of MOD methods are critical when targeting energy-constrained systems, such as Wireless Multimedia Sensor Networks (WMSN). The challenge is always to reduce computational costs while preserving high detection accuracy. In this article, we present EVBS-CAT, an Enhanced Video Background Subtraction with a Controlled Adaptive Threshold selection method for low-cost surveillance systems. The proposed moving object detection method utilizes background subtraction (BS) with morphological operations and adaptive thresholding. We evaluate the algorithm using the Change Detection 2012 dataset. Through a computational complexity analysis of each step, we demonstrate the efficiency of the proposed MOD technique for embedded WMSN. The algorithm yields promising results compared to state-of-the-art MOD techniques in the context of embedded wireless surveillance.

Résumé: In recent years, the Discrete Tchebichef Transform (DTT) has gained popularity as a signal processing tool for image and video compression due to its efficient coding and decorrelation properties. However, in the context of real-time applications and embedded systems, it is critical to develop approximate algorithms with reduced complexity and energy consumption. While three DTT approximations have been proposed to date, there is still room for further improvements. To address this gap, we propose two new low-complexity DTT approximations that employ a modified deviation metric, resulting in better compression efficiency and reduced complexity. We validate our proposed methods by implementing them on the Xilinx Virtex-6 XC6VSX475T-1FF1759-2 Field Programmable Gate Array (FPGA) through rapid prototyping. Our proposed transformations exhibit superior performance in terms of hardware resources and energy consumption, particularly for 1D 8 inputs. Furthermore, compared to the state-of-the-art DTT approximations in image compression, our proposed transformations demonstrate a quality gain of up to 2 dB. Overall, our proposed approximations provide a promising trade-off between image quality, hardware resources, and energy consumption, making them ideal for real-time applications and embedded systems.

Résumé: The increase in demand for video delivery over the last few years has led to a need for more compression efficiency. High efficiency video coding (HEVC) offers a better compression rate compared to the preceding standard codecs. However, the robustness of the coded stream is reduced in the low delay mode used for real-time applications. When a bitstream is transmitted over a hostile network, there is a high probability of burst network packet loss which can result in the loss of the entire frame. To deal with these signal degradations that occur in the transmission channel, an HEVC encoder adaptation scheme based on spatial multiple description coding (MDC) is proposed. A comparative study with the single description coding (SDC) scheme has shown its efficiency in improving the quality of the reconstructed video in the five packet loss cases studied and yields an average gain of about 14 to 35.57%.

Résumé: This paper presents a MATLAB implementation of the exact real Discrete Tchebichef Transform (DTT) and Integer Discrete Tchebichef Transform (IDTT) with their inverse functions. It is a simple and scalable package that is the first of its kind for MATLAB. This software provides an efficient way to advance research related to DTT, IDTT, and their inverses. It is applicable to all domains where Chebyshev moments are employed, including feature extraction, image and video compression, and other related image processing. The package supports matrix generation of any size, from 2 × 2 to any desired size, which makes it scalable. The presented software is easy to use and generates the exact DTT transformation kernel, integer IDTT kernel, and diagonal matrix. The performance of DTT and IDTT-based JPEG compression compared to DCT-based JPEG compression was evaluated using standard metrics. Overall, this package is a valuable tool for researchers and practitioners working in the field of image and video processing.