In the field of computer vision, object detection consists of automatically finding objects in images by giving their positions. The most common fields of application are safety systems (pedestrian detection, identification of behavior) and control systems. Another important application is head/person detection, which is the primary material for road safety, rescue, surveillance, etc. In this study, we developed a new approach based on two parallel Deeplapv3+ to improve the performance of the person detection system. For the implementation of our semantic segmentation model, a working methodology with two types of ground truths extracted from the bounding boxes given by the original ground truths was established. The approach has been implemented in our two private datasets as well as in a public dataset. To show the performance of the proposed system, a comparative analysis was carried out on two deep learning semantic segmentation state-of-art models: SegNet and U-Net. By achieving 99.14% of global accuracy, the result demonstrated that the developed strategy could be an efficient way to build a deep neural network model for semantic segmentation. This strategy can be used, not only for the detection of the human head but also be applied in several semantic segmentation applications.

The propagation of ultra-wideband (UWB) signals is analyzed. The UWB radar output signals are formed both with transmitter and antenna. Moreover, they can be substantially affected due to electromagnetic wave propagation through walls. Experimental results and numerical simulations are analyzed and compared. The given examples show that the numerical simulations, which are very fast, could be reliable, when suitable wall parameters are used.

The paper deals with an accurate Time Difference Of Arrival (TDOA) measurement method for passive radar systems. The reached measurement accuracy of several tenths of ns provides a chance to design a short base radar system with a sufficient positioning accuracy even at long ranges. The feasibility is documented using a standard hyperbolic system model of the short base TDOA system based on reception of Automatic Dependent Surveillance-Broadcast (ADSB) messages. The anticipated radar system, deployed in a small radius used a centralized time of arrival measurement with optical cable transmission of the received signals to the measurement processor. The horizontal and vertical position errors of such a system are estimated for different radius of the radar receivers array.

In this paper an optimization scheme about the association of targets on multistatic radar system is discussed. The association of targets based on Probability Hypothetic Density function (PHD) is a challenging task for the parallel programming techniques. In this paper, a comparison of PHD computations between serial and optimized versions of the proposed algorithm is presented. The proposed optimized algorithm was implemented on Parallel Computing Toolbox under Matlab system.

This paper deals speaker localization using received acoustic signals on microphone array. The real time localization system is based on method for a speaker speech processing and extraction of sibilant sequence. It is shown, that this method leads to good position estimation accuracy in indoor systems.