Multi-sensor Surveillance System Research Articles

Engineering a multi-sensor surveillance system with secure alerting for next-generation threat detection and response

This paper presents an advanced Surveillance Warning System (SWS) designed for next-generation video surveillance applications. While contemporary alert systems have improved analytics engines, their binary threat detection using only optical imaging sensors has limitations in low-light conditions, system security, and multilevel threat detection (i.e., low, moderate, and critical). Our approach integrates a novel alert module with an enhanced video analytics engine, contributing to a comprehensive framework integrating surveillance, access control, multilevel threat detection, and multi-tiered alert routing. We develop a Video Surveillance and Motion Detection (VSMD) application utilizing optical imaging, Passive Infrared (PIR), and fingerprint sensors for multilevel threat assessment, access control, and security. Unlike existing systems, we develop an intelligent alert module generating threat-specific alerts transmitted securely through cellular phones, the Internet, and encrypted radio frequency (RF) signals, significantly enhancing real-time threat awareness and response capabilities. Validation against existing solutions highlights the system's adaptability, rapid response, and overall security advancements in video surveillance.

Absent Multiview Semisupervised Classification.

With the advent of vast data collection ways, data are often with multiple modalities or coming from multiple sources. Traditional multiview learning often assumes that each example of data appears in all views. However, this assumption is too strict in some real applications such as multisensor surveillance system, where every view suffers from some data absent. In this article, we focus on how to classify such incomplete multiview data in semisupervised scenario and a method called absent multiview semisupervised classification (AMSC) has been proposed. Specifically, partial graph matrices are constructed independently by anchor strategy to measure the relationships among between each pair of present samples on each view. And to obtain unambiguous classification results for all unlabeled data points, AMSC learns view-specific label matrices and a common label matrix simultaneously. AMSC measures the similarity between pair of view-specific label vectors on each view by partial graph matrices, and consider the similarity between view-specific label vectors and class indicator vectors based on the common label matrix. To characterize the contributions of different views, the p th root integration strategy is adopted to incorporate the losses of different views. By further analyzing the relation between the p th root integration strategy and exponential decay integration strategy, we develop an efficient algorithm with proved convergence to solve the proposed nonconvex problem. To validate the effectiveness of AMSC, comparisons are made with some benchmark methods on real-world datasets and in the document classification scenario as well. The experimental results demonstrate the advantages of our proposed approach.

МЕТОДИКА ВЫЯВЛЕНИЯ ЛОЖНЫХ ПРЕДНАМЕРЕННО ФОРМИРУЕМЫХ СИГНАЛОВ АЗН-В

The International Civil Aviation Organization recommends to use advanced surveillance systems: multi-sensor surveillance system and automatic dependent surveillance broadcast. Ground stations for automatic dependent surveillance have a relatively low cost and can partially replace secondary surveillance radars in the future. Currently, the development of automatic dependent surveillance has slowed down, because the system has vulnerabilities: the complete openness of the 1090ES data link, the lack of encryption and protection against false intentionally generated signals. There are a large number of equipment on the market for receiving and processing on-board transponder signals; for specialists it's not difficult to assemble transmitting equipment that generates such signals. This poses a serious threat to information security and flight safety. The article proposes a method that allows, through a comprehensive analysis of the aircraft trajectory and the nature of the level of received signals, identify false signals generated intentionally. The proposed technique is based on identification of abnormal characteristics of false signals. To search and identify anomalous characteristics, the Mahalanobis distance is estimated. To increase the distinguishability of data and to detect intentional interference, it is necessary to modernize the antenna system of the ground station by introducing an additional antenna with a different radiation pattern. The false signal generated by the ground source is detected by a constant value of the antenna directivity ratio. The developed technique was implemented in the MATLAB software and applied to evaluate real data received from the ground-based automatic dependent surveillance station located in the Arkhangelsk air traffic management center. The power levels of the received messages of automatic dependent observation received from L-410 aircraft operating in the area of local air lines were used as reference values.

Adaptive passive sensor selection for maneuvering target localization and tracking using a multisensor surveillance system

This paper investigates maneuvering-target tracking problem based on a multisensor system and interacting multiple model (IMM). The estimation is performed by a novel particle filter (PF) with a ca...

Modeling Design and Implementation of an Embeds System Real Time Over a Network of Wireless Sensors to Environmental Monitoring

Artificial Neurons Network (ANN) is used in the decision and control of dynamic systems which can be with a lack of superfluous information.it forces the use of fuzzy logic.
 
 For this reason, several methods and monitoring techniques have been implemented. This article presents a technique based on artificial neural networks implanted at the level of a multisensor surveillance system. It is a statistical learning method that displays optimal training and generalization performance in several domains, including the recognition domain of forms.
 
 In this case ANN based on raspberry PI card for decision node and arduino for the input and hidden nodes, in order to develop a complete platform environmental monitoring system. and hence enhance multi-Sensor wireless signals aggregation via multi-bit decision fusion. 
 
 The back-propagation algorithm generates a weight for all nodes in the networks, with aim of minimizing absolute error committed in fusion data and economics of electrical energy using artificial intelligence techniques. This algorithm is more efficient than the human being since it can reason and learn from its errors so as not to repeat them.
 
 Its main applications include a variety of data monitoring parameters (such as : temperature, humidity, gas sensor, … etc), that can be found in factory automation, for instance : home automation, remote monitoring and home device control, or it may be used in environment to make an exact decision in short time.

Optimal estimation of sensor biases for asynchronous multi-sensor data fusion

An important step in a multi-sensor surveillance system is to estimate sensor biases from their noisy asynchronous measurements. This estimation problem is computationally challenging due to the highly nonlinear transformation between the global and local coordinate systems as well as the measurement asynchrony from different sensors. In this paper, we propose a novel nonlinear least squares (LS) formulation for the problem by assuming the existence of a reference target moving with an (unknown) constant velocity. We also propose an efficient block coordinate decent (BCD) optimization algorithm, with a judicious initialization, to solve the problem. The proposed BCD algorithm alternately updates the range and azimuth bias estimates by solving linear least squares problems and semidefinite programs (SDPs). In the absence of measurement noise, the proposed algorithm is guaranteed to find the global solution of the problem and the true biases. Simulation results show that the proposed algorithm significantly outperforms the existing approaches in terms of the root mean square error (RMSE).

Adaptive maneuvering target tracking with 2-HFSWR multisensor surveillance system

The high frequency surface wave radar (HFSWR) with detection capability of hundreds of kilometers plays an important role in marine surveillance [1]–[8]. Nevertheless, HFSWR has low azimuth measurement accuracy, which causes great deviation of target location and tracking [9]. Meanwhile, the miniaturization of radar arrays leads to the limitation of antenna length, which further causes severe decline of azimuth accuracy. Moreover, the interference from external environment also decreases the azimuth accuracy, making the accurate tracking of vessels difficult. In recent years, the 2-HFSWR multisensor surveillance system has become a hot spot, because it only requires parameters of distance and velocity without using the azimuth measurement when tracking vessels. The 2-HFSWR multisensor surveillance system consists of two independent radars implementing the target location and tracking task by detecting the radial distance and velocity, respectively. The 2-HFSWR multisensor surveillance system can ensure the detection accuracy without azimuth measurement, which is advantageous to the miniaturization of HFSWR. However, there are still certain complicated cases to be solved for continuous tracking of vessels, for example sudden change in motion state, an inaccurate model, and unknown priori statistical properties of noise.

Development of Multi-functional Control Architecture for Multisensor Surveillance Systems

Development of Multi-functional Control Architecture for Multisensor Surveillance Systems

Evidential framework for data fusion in a multi-sensor surveillance system

The multi-sensor data fusion relies on a combination of information pieces to produce a more accurate or complete description of the environment. In this work, we considered the case of a surveillance system using several heterogeneous sensors in a network. In such a system, the data fusion objective is to merge the detections provided by the different sensors in order to count, locate and track all the targets in the monitored area. The problem was addressed in the context of Belief Function theory in order to cope with the high inaccuracy of information and the different forms of imprecision. In this framework, we developed a unified approach to model and merge the detections coming from various kinds of sensors with prior knowledge about target location derived from topographical elements. We showed that the developed belief model provided an efficient measurement for data association between tracks and detections. Considering scalable constraints for the system, the complexity and consistency of belief function representation should be controlled, which was achieved by implementing versatile discernment frames and by restricting the number of focal elements. The proof of concept of the proposed data fusion module was achieved by implementing it in an actual detection system. Real-world scenarios were used to draw some conclusions about localization performance and end-user perception. Further experiments were also performed on simulated data to focus on data association and belief function simplification subproblems.

A practical approach for active camera coordination based on a fusion-driven multi-agent system

In this paper, we propose a multi-agent system architecture to manage spatially distributed active (or pan-tilt-zoom) cameras. Traditional video surveillance algorithms are of no use for active cameras, and we have to look at different approaches. Such multi-sensor surveillance systems have to be designed to solve two related problems: data fusion and coordinated sensor-task management. Generally, architectures proposed for the coordinated operation of multiple cameras are based on the centralisation of management decisions at the fusion centre. However, the existence of intelligent sensors capable of decision making brings with it the possibility of conceiving alternative decentralised architectures. This problem is approached by means of a MAS, integrating data fusion as an integral part of the architecture for distributed coordination purposes. This paper presents the MAS architecture and system agents.

Auction-Based Consensus Mechanism for Cooperative Tracking in Multi-Sensor Surveillance Systems

This paper presents an auction-based consensusmechanism for cooperative targets tracking using minimum numbers of mobile sensors in order to reduce energy consumption due to sensor mobilization. After targets are detected, they are clustered using hybrid subtractive K-means clustering technique to reduce the number of trackers needed to track these detected targets. The proposed target tracking process is based on an Extended Kohonen neural network. In order to decrease the network sensitivity to initial conditions, a supervised learning technique is used to get the initial weights of unsupervised Extended Kohonen Map instead of random initialization. An auction-based consensus mechanism is used as a cooperation methodology between trackers during tracking. Monitoring sensors either remain stationary or begin following their targets is based on this mechanism. The simulation results confirms that the proposed approach outperforms other approaches in energy saving and achieves better coverage.

Use of Continuous Zoom on Electro-Optical Imaging Systems: Comparisons between Automatic and Manual Target Tracking

The use of continuous zoom in an electro-optical sensor system was investigated with respect to target tracking. Using a simulation of an operator-machine interface in an airborne multi-sensor surveillance system, targets were tracked by manually directing the sensor or by an automated tracker. It was hypothesized that frequency of using the continuous zoom would be higher in the manual tracking mode than in auto-tracking, and negatively correlated with tracking error. Sensor, and targets to be tracked, were either moving or stationary in three types of tracking scenarios. Results showed that the zoom function was used more often when tracking manually, although the way continuous zoom was used differed between the two tracking modes. Also, tracking error was lower when the zoom function was used in manual mode. Tracking error was additionally affected by whether or not the target and/or the sensor were moving or stationary. Results improve our understanding of the way complex sensor systems are used, and will assist in ascertaining whether providing a continuous zoom into optical imaging systems is of benefit to operators.

Development of intelligent multisensor surveillance systems with agents

Intelligent multisensor surveillance systems consist of several types of sensors, which are installed on fixed and mobile devices. These components provide a huge quantity of information that has to be contrasted, correlated and integrated in order to recognize and react on special situations. These systems work in highly dynamic environments, with severe security and robustness requirements. All these characteristics imply the need for distributed solutions. In these solutions, scattered components can decide and act with some degree of autonomy (for instance, if they become isolated), or cooperate and coordinate for a complete tracking of special situations. In order to cope with these requirements and to better structure the solution, we have decided to design surveillance system control as a multiagent system. This is done by applying an agent-orientated methodology, which is assessed with concrete scenarios.

A Fully Decentralized Multi-Sensor System For Tracking and Surveillance

In many tracking and surveillance systems, multisensor config urations are used to provide a greater breadth of measurement information and also to increase the capability of the system to survive individual sensor failure. Most architectures currently used in such systems rely on having either a central processor where global data fusion takes place or a central communi cations medium through which all messages between sensors must pass. Such centralized architectures give rise to problems with communication and computational bottlenecks and are susceptible to total system failure should the central facility fail. In this article we describe a multisensor surveillance system that achieves a high degree of survivability by employing a decentralized sensing architecture (Durrant-Whyte, Rao, and Hu 1990). Using CCD cameras and optical barriers the system is able to track (in real time) people and robots as they move around a factory room. The decentralized sensing architecture takes the form of a network of Transputer-based sensor nodes, each with its own processing facility, that together do not re quire any central processor, any central communication facility, or any common clock. In this architecture, computation is per formed locally, and communication occurs between any two nodes as and when required. The starting point for the surveillance system is an algo rithm that allows complete decentralization of the multisensor Kalman filter equations among a number of sensing nodes. We develop the algorithm to ensure that internodal commu nication is minimized and can take place without any prior synchronization between nodes. By designing a suitable track management scheme, we are able to extend the system to cope with multitarget tracking. Finally, the specific sensors used in the implementation are interfaced to the algorithm, and the results of the working system are given.