Drone Surveillance Research Articles

Prize Collecting Multiagent Orienteering: Price of Anarchy Bounds and Solution Methods

We propose and address a new variation of the team orienteering problem (TOP) in which all members of the team are independent self-interested agents. The prize-collecting nature emanates from the fact that the prize available at a node of the traversal graph can be collected only by a single visiting agent. The problem is motivated by situations in which team members (agents) must accomplish tasks toward a common goal but are unable to communicate, such as a fleet of surveillance drones operating in a communication denied area or with remote pilots operating independently. We explore three policies for minimizing the amount of inefficiency these self-interested agents can bring into the situation. We analyze these policies in a game-theoretic framework and show upper bounds on the Price of Anarchy (PoA) ranging from ≈1.582 to unbounded depending on the policy, network type, and the number of players k. This is done by extending well-known PoA bounds for valid utility systems to a leader-follower setting. The PoA also depends on the behavior of the agents, which could not have goodwill toward other agents. In many cases, we are able to provide examples that establish the tightness of the bounds. Finally, solution methods are provided for each of these policies. Numerical results computed by these solution methods are then presented and compared with the optimal centrally coordinated solutions.

Misconduct Committed by (Civilian) Private Contractors in Peacekeeping Operations

Abstract In peacekeeping operations, private companies are frequently and increasingly engaged by the United Nations to carry-out a wide-range of activities that can potentially impinge on human rights. This article deals with two recent cases of misconduct committed by contractors whose activities, albeit not on the face of it involving the threat or (lethal or not-lethal) use of force, nonetheless caused harm to individuals. The first case-study relates to the mismanagement of sanitary waste, while the second case addresses the (mis)use of unarmed surveillance drones. Against this backdrop, the article purports to assess whether, and under what conditions, wrongdoing committed by private contractors gives rise to the international responsibility of the United Nations. The study explores, firstly, the question of the ‘direct’ attribution of such conduct to the United Nations, based on the qualification of the contractors as agent of the Organization, as understood in the Draft Articles on the Responsibility of International Organizations. It then analyzes the issue of the ‘indirect’ responsibility of the United Nations for failing to have sufficient oversight of the outsourced activities. In this respect, it highlights the crucial role played by internal accountability mechanisms, in particular the Office of Internal Oversight Services, in appraising the monitoring measures taken by the Organization with regard to the practice of contractors and in recommending remedial actions.

Impact of lift propeller drag on the performance of eVTOL lift+cruise aircraft

Electric vertical takeoff and landing (eVTOL) aircraft of the lift+cruise configuration are significantly impacted by the drag of the lift propellers even though they are only briefly used for takeoff and landing. This work aims to quantify the drag penalty of these high lift propellers, its impact on the performance of a lift+cruise eVTOL aircraft, and potential benefits from retracting the propellers during cruise. An eVTOL model is tested in the wind tunnel with propellers extended and retracted. Results show that retracting the propellers reduces parasitic drag by 38%. This measured drag reduction is used to estimate the performance of a passenger eVTOL and a surveillance drone employing a propeller retraction system. A passenger carrying eVTOL lift+cruise aircraft with a propeller retraction system could cruise 21% faster, at the same mission range, or increase range by 13% at the same flight speed. When endurance is the main mission objective, the benefits are smaller, in the order of 7%, and the potential gains are more sensitive to the weight of the retraction system.

A provably secure and lightweight authentication scheme for Internet of Drones for smart city surveillance

Smart cities are made of different interconnected components that exchange data and facilitate the urban living for citizens. Unmanned Aerial Vehicles, known as drones, are components used in various civilian applications, such as agriculture, package delivery, and surveillance. To enhance drone’s flying safety and quality of service, a promising idea is to construct the Internet-of-Drones, where the drones are employed to collect the information, and citizens communicate with the drones of a particular flying zone via their mobile devices to obtain the data in real-time. However, due to the high sensitivity of the information, and the openness of communication media, privacy and security issues should be addressed. Specifically, an efficient and secure authentication scheme is needed to enable users and drones to authenticate each other and share a session key. Due to the drone’s limited resources and energy, the authentication scheme should be efficient regarding the computation overhead while providing high security. In this article, we devise a secure authentication scheme based on elliptic curve for drones to secure smart city surveillance. We demonstrate that our scheme is provably secure in the random oracle model, supporting the security requirements and resisting known attacks while incurring low computation and communication costs. • We devise an authentication scheme for IoD environments to provide a safe and smart-life for citizens in a smart city. • The proposed protocol is provably secure in the Random Oracle Model. • The security of the proposed protocol is validated formally through the Scyther tool. • The proposed protocol provides resistance against well-known potential security attacks. • The performance analysis showed that proposed protocol is more efficient than other competing protocols.

Wind turbine blade structural state evaluation by hybrid object detector relying on deep learning models

Surveillance drones are remarkable devices for monitoring, as they have strong spatial and remote sensing capabilities. The prompt detection of peripheral damage to the blades of wind turbines is necessary to reduce downtime and prevent the potential failure of wind farms. Computer vision breakthroughs with deep learning have developed and been refined over time, mainly using convolution neural networks. From this perspective, we suggest a deep learning model for monitoring and diagnosing the blade health of wind turbines based on images captured by surveillance drones. The main limitations of standard monitoring devices are their poor detection accuracy and lack of real-time performance, making it complex to obtain the attributes of blades from aerial images. Based on the foregoing, this study introduces a method for increasing detection accuracy when carrying out operations in real time using You Only Look at Once version 3 (YOLOv3). We train and evaluate three deep learning models on the wind turbine image dataset. We find that many aerial images are unclear because of blurred motion. As avoiding such low-resolution images for training can affect accuracy, we use a super-resolution convolution neural network to reconstruct a blurred picture as a high-resolution one. The computational results demonstrate that YOLOv3 outperforms traditional models in terms of both accuracy and handling time.

Beach-user perceptions and attitudes towards drone surveillance as a shark-bite mitigation tool

Management of human-wildlife conflict is often challenging and complex, particularly when the conflict involves sharks. New technologies are being trialled in New South Wales, on Australia's east coast, to accommodate the community demand for increased beach-user protection that does not harm marine wildlife. Drones (or unmanned aerial vehicles), are one of a suite of potential tools that can address both these demands. We released an online survey to assess beach-user perceptions and attitudes toward drones on NSW beaches as a shark surveillance tool. From 439 respondents, we found the use of drones on coastal beaches was accepted by the majority of people surveyed (88%) due to perceptions of reduced impact on sharks, and the relatively low cost. Drone surveillance was also the preferred approach for bather protection overall. Arguably the most vulnerable beach-user group for a shark bite incident, surfers, claimed the highest level of awareness of the use of drones for shark surveillance, but also indicated lower confidence in their utility compared to other groups. The study demonstrates an overall social licence regarding the use of drones for shark surveillance purposes, with the levels of support likely to increase with further public education efforts and improvements to the efficacy of drone-based surveillance.

Self-Controllable Super-Resolution Deep Learning Framework for Surveillance Drones in Security Applications

This paper proposes a self-controllable super-resolution adaptation algorithm in drone platforms. The drone platforms are generally used for surveillance in target network areas. Thus, super-resolution algorithms which are for enhancing surveillance video quality are essential. In surveillance drone platforms, generating video streams obtained by CCTV cameras is not static, because the cameras record the video when abnormal objects are detected. The generation of streams is not predictable, therefore, this unpredictable situation can be harmful to reliable surveillance monitoring. To handle this problem, the proposed algorithm designs superresolution adaptation. With the proposed algorithm, the shallow model which is fast and low-performance will be used if the stream queue is near overflow. On the other hand, the deep model which is highperformance and slow will be used if the queue is idle to improve the performance of super-resolution.

Facial recognition system using LBPH face recognizer for anti-theft and surveillance application based on drone technology

Providing security to the citizens is one of the most important and complex task for the governments around the world which they have to deal with. Street crimes and theft are the biggest threats for the citizens and their belonging. In order to provide security, there is an urgent need of a system that is capable of identifying the criminal in the crowded area. This paper proposes a facial recognition system using Local Binary Patterns Histogram Face recognizer mounted on drone technology. The facial recognition capability is a key feature for a drone to have in order to find or identify the person within the crowd. With the inception of drone technology in the proposed system, we can use it as a surveillance drone as well through which it can cover more area as compared to the stationary system. As soon as the system identifies the desired person, it tags him and transmits the image along with the co-ordinates of the location to the concerned authorities using mounted global positioning system. Proposed system is capable of identifying the person with the accuracy of approximately 89.1%.

Structural and modal analysis of hybrid low altitude self-sustainable surveillance drone technology frame

Structural and modal analysis of different versions of a special type of drone frame using carbon fibre and ABS materials are carried out in this paper using commercially available simulation software. The drone discussed in this paper is capable of self-sustaining in terms of power by generating its energy from solar cells and using a chamber filled with lighter than air materials. Thus, to support the drone and its components for such applications four versions of the frame namely directly connected triangular frame, short armed semi-circular frame, long armed semi-circular frame and short armed rectangular frame with two different materials such as carbon fibre and Acrylonitrile Butadiene Styrene (ABS) plastic were analysed by static structural and frequency analysis. The maximum permissible stress, strain displacement and frequency values were obtained from the simulation. The analysis indicates that the directly connected triangular frame and long armed semi-circular frame have less strength compared to other two versions but the frames can be manufactured easily. Whereas short armed semi-circular frame and short armed rectangular frames are complex to manufacture but they have high strength to withstand the induced stress.

A hybrid approach for search and rescue using 3DCNN and PSO

Search and rescue are essential applications of disaster management in which people are evacuated from the disaster-prone area to a safer place. This overall process of search and rescue can be more efficient if an automated system can quickly locate the human or area where rescue is required. To provide a faster and accurate search of those places, this paper proposes a novel approach to search and rescue using automated drone surveillance. In this paper, a complex scene classification problem is solved using the proposed 3DCNN model. The proposed model uses spatial as well as temporal features of the video for the classification of the scene as help or non-help in the natural disaster. Due to the unavailability of such kind of dataset, it is impossible to train the model. Therefore, it is essential to develop a dataset for search and rescue. The proposed dataset is a first and unique dataset for scene classification using drone surveillance. The major contribution of this paper is (1) a novel 3DCNN powered model for scene classification in drone surveillance, (2) to develop the required dataset for the training of scene classification model, and (3) particular swarm optimization (PSO)-based hyper-parameter tuning for getting the best value of multiple parameters used for training the model. Our hybridization of parameter tuning with PSO helps for the convergence of parameter values of proposed 3DCNN model, and the proposed scene classification model (3DCNN+PSO) is applied to the dataset. The proposed model gives an impressive performance to help situation identification with 98% training and 99% validation accuracy.

“Design fiction” skirts reality to provoke discussion and debate

In October 2015, researchers presented an unusual paper at a computer science conference in London. The paper described the promising results of a pilot project in which a local community used surveillance drones to enforce car parking restrictions and to identify dog owners who failed to clean up after their pets. Controlled by four elderly retirees, the drones buzzed around the city and directed council officials on the ground (1). To prompt discussion about the ethics of using drones for surveillance, researchers described a fanciful pilot project in which a local community used surveillance drones to enforce car parking restrictions. It’s one example of design fiction. Image credit: Shutterstock/Alex Yuzhakov. The paper and its accompanying video generated lively discussion about the ethics and regulation of drone use among delegates at the CHI PLAY conference. But there was a catch: The paper, the video, and the pilot scheme were fictional, as the researchers admitted at the end of both the paper and the presentation. The researchers had invented the scenario as a way to focus attention on how drone technology—a topic of study for some of the people in the room—could shape and change society. The team thought that presenting the idea as if it were real—for example, showing familiar street signs in the video warning drivers about a drone-controlled zone—would provoke discussion about a future in which such use of technology was considered mundane. The practice is called design fiction. Originally used in product design, the approach is finding increasing use in scientific and medical fields as a way to explore the possible consequences of technological development. These projects are not so much experiments designed to test a hypothesis as they are orchestrated scenarios designed to provoke forward-thinking discussion and debate. From climate science and artificial intelligence to wearable …

Residual Energy Analysis in Cognitive Radios with Energy Harvesting UAV under Reliability and Secrecy Constraints.

The integration of unmanned aerial vehicles (UAVs) with a cognitive radio (CR) technology can improve the spectrum utilization. However, UAV network services demand reliable and secure communications, along with energy efficiency to prolong battery life. We consider an energy harvesting UAV (e.g., surveillance drone) flying periodically in a circular track around a ground-mounted primary transmitter. The UAV, with limited-energy budget, harvests radio frequency energy and uses the primary spectrum band opportunistically. To obtain intuitive insight into the performance of energy-harvesting, and reliable and secure communications, the closed-form expressions of the residual energy, connection outage probability, and secrecy outage probability, respectively, are analytically derived. We construct the optimization problems of residual energy with reliable and secure communications, under scenarios without and with an eavesdropper, respectively, and the analytical solutions are obtained with the approximation of perfect sensing. The numerical simulations verify the analytical results and identify the requirements of length of sensing phase and transmit power for the maximum residual energy in both reliable and secure communication scenarios. Additionally, it is shown that the residual energy in secure communication is lower than that in reliable communication.

A learnheuristic approach for the team orienteering problem with aerial drone motion constraints

This work proposes a learnheuristic approach (combination of heuristics with machine learning) to solve an aerial-drone team orienteering problem. The goal is to maximise the total reward collected from information gathering or surveillance observations of a set of known targets within a fixed amount of time. The aerial drone team orienteering problem has the complicating feature that the travel times between targets depend on a drone’s flight path between previous targets. This path-dependence is caused by the aerial surveillance drones flying under the influence of air-resistance, gravity, and the laws of motion. Sharp turns slow drones down and the angle of ascent and air-resistance influence the acceleration a drone is capable of. The route dependence of inter-target travel times motivates the consideration of a learnheuristic approach, in which the prediction of travel times is outsourced to a machine learning algorithm. This work proposes an instance-based learning algorithm with interpolated predictions as the learning module. We show that a learnheuristic approach can lead to higher quality solutions in a shorter amount of time than those generated from an equivalent metaheuristic algorithm, an effect attributed to the search-diversity enhancing consequence of the online learning process.

Drone-surveillance for search and rescue in natural disaster

Due to the increasing capability of drones and requirements to monitor remote areas, drone surveillance is becoming popular. In case of natural disaster, it can scan the wide affected-area quickly and make the search and rescue (SAR) faster to save more human lives. However, using autonomous drone for search and rescue is least explored and require attention of researchers to develop efficient algorithms in autonomous drone surveillance. To develop an automated application using recent advancement of deep-learning, dataset is the key. For this, a substantial amount of human detection and action detection dataset is required to train the deep-learning models. As dataset of drone surveillance in SAR is not available in literature, this paper proposes an image dataset for human action detection for SAR. Proposed dataset contains 2000 unique images filtered from 75,000 images. It contains 30000 human instances of different actions. Also, in this paper various experiments are conducted with proposed dataset, publicly available dataset, and stat-of-the art detection method. Our experiments shows that existing models are not adequate for critical applications such as SAR, and that motivates us to propose a model which is inspired by the pyramidal feature extraction of SSD for human detection and action recognition Proposed model achieves 0.98mAP when applied on proposed dataset which is a significant contribution. In addition, proposed model achieve 7% higher mAP value when applied to standard Okutama dataset in comparison with the state-of-the-art detection models in literature.

Smart Irrigation and Intrusions Detection in Agricultural Fields Using I.o.T.

Agriculture plays an important role in the life span of human being not only for their survival but for the better economic growth of the country too. Precision agriculture is the new trending term in the field of technology whose main motive is to reduce the workload of the farmers and increase the productivity of the farms by using technologies like I.O.T, WSNs, Remote Sensing, Drone surveillance and many more. In this paper, we show the work done by our cost effective and reliable device whose aim is to irrigate fields only when there is a need of water and to provide information about detection of any intrusion in agricultural fields. The information is sent to the farmers by using cloud application. The performance of our system is measured in terms of intrusion detection and moisture of soil for irrigation.

The Limits of Recognition: Rethinking Conventional Critiques of Drone Warfare

The Limits of Recognition:Rethinking Conventional Critiques of Drone Warfare Rebecca A. Adelman (bio) A Google search for "anti-drone clothing" mostly and predictably yields an array of t-shirts printed with slogans protesting drone warfare. But the results also include something far more interesting: Adam Harvey's Stealth Wear line, garments "fabricated with silver-plated fabric that reflects thermal radiation, enabling the wearer to avert overhead thermal surveillance."1 These shirts, burqas, hijabs, and hoodies muffle the heat signature of the human body, presumably making it undetectable to the drones loitering above. This apparel is designed to foil drone operators by rendering the wearer unidentifiable as a human, and consequently invisible as a target, thereby breaking the circuit of recognizability on which drone surveillance and warfare fundamentally depend. Within this circuit, to be recognizable as a living human is to be recognizable as a target, and hence to be endangered. While Stealth Wear operates on the assumption that a human rendered unrecognizable as such will be protected by this type of invisibility, much antidrone activism presumes that humans who are recognizable as such will be protected by their visibility. Anti-drone activists often seek to emphasize the humanness of the people who are killed by drone strikes and call for drone operators, as well as the states and militaries who prosecute drone warfare, to recognize that humanness. The anti-drone faith in recognition is grounded in two conceits. The first is that drone technologies—with their remoteness, mediation, and putative similarity to video games—dehumanize their targets and thus encourage and facilitate wanton violence on the part of their operators. The second, and corollary, notion follows: if drone operators recognized the [End Page 93] beings on their displays as human, they would be unable or unwilling to kill them. Both of these premises, I suggest, are faulty. Indeed, there is increasing evidence against them, including the high rate of PTSD among drone operators, which suggests they are keenly aware of the consequences of their actions. But the figure of the oblivious drone operator who needs only to be enlightened persists in popular culture, as well as in activist and scholarly critiques of drone warfare. Consequently, anti-drone work often proceeds as an attempt to disabuse drone operators of their supposed ignorance through an emphasis the humanness of their targets. According to a March 2019 BBC count, the Trump administration approved 2243 strikes during its first two years, while working steadily to relax reporting requirements about drone casualties.2 Critics were already dismayed by the Obama administration's embrace of drone warfare, and this form of violence persists, sprawls, and sometimes intensifies, with no sign that it will abate (or that the American public will lose the appetite for it). In short, the durability and entrenched nature of drone warfare suggest a need to rethink prevailing critical frameworks for opposing it. In this article, I query the limits of recognition as a strategy of anti-drone resistance. I begin with a consideration of the politics of recognition and humanity in drone warfare. This overview serves as the foundation for my subsequent exploration of approaches to recognition operative in three artifacts of drone warfare: a 2014 report from the Stimson Center Task Force on US Drone Policy; a public art installation in Pakistan called Not a Bug Splat; and the playful photographic meditation on drone subjectivity by the IOCOSE collective, "Drone Selfies." I conclude with a consideration of the emerging awareness of PTSD in drone operators and the economies and politics of recognition operative within it. Ultimately, I demonstrate that the emphasis on recognition amounts to a misidentification of the conditions by which drone warfare proceeds and, moreover, of the means by which it might be resisted. Recognition in Anti-Drone Activism The discourse of recognition in anti-drone activism is rooted in liberalism and reliant on a range of beliefs about individuality, freedom, and humanity. It hinges on a pair of assumptions: that the systematic lack of recognition of targets' humanity enables the persistence of drone warfare and, consequently, that adequate recognition of that humanity would short-circuit the willingness of both nation-states and drone operators to engage in...

Node and Edge Drone Surveillance Problem With Consideration of Required Observation Quality and Battery Replacement

Our study introduces a drone routing problem in which drones fly to capture photos for surveillance purposes after a disaster. The drones perform observations on nodes and edges representing populated areas and road segments of a network from multiple altitudes. Each target node and edge requires observation at least once with a certain required quality. When the drones fly at a relatively high altitude, they can simultaneously capture low-quality photos and a large number of observed target nodes and edges. However, high-quality photos and narrow observation areas can be captured from a relatively low altitude. Each drone has a limited battery capacity and thus must return to the depot for battery replacement. This study routes the drones to satisfy the required photo quality of all target nodes and edges while minimizing the makespan of the surveillance by all drones. Our study is the first to examine a multiple-drone routing problem while considering flight altitude-dependent observation quality, battery replacement, node and edge combination, and minimizing the makespan. Our problem is formulated as a mixed integer linear programming (MILP) model. Firefly and adaptive-reactive tabu search algorithms are proposed. The latter outperforms the former and obtains better solutions than those in the MILP model for small-sized instances within a given short computation time.

Maximising reward from a team of surveillance drones: a simheuristic approach to the stochastic team orienteering problem

We consider the problem of routing a team of unmanned aerial vehicles (drones) being used to take surveillance observations of target locations, where the value of information at each location is different and not all locations need be visited. As a result, this problem can be described as a stochastic team orienteering problem (STOP), in which travel times are modelled as random variables following generic probability distributions. The orienteering problem is a vehicle-routing problem in which each of a set of customers can be visited either just once or not at all within a limited time period. In order to solve this STOP, a simheuristic algorithm based on an original and fast heuristic is developed. This heuristic is then extended into a variable neighbourhood search (VNS) metaheuristic. Finally, simulation is incorporated into the VNS framework to transform it into a simheuristic algorithm, which is then employed to solve the STOP. [Received 5 January 2019; Revised 15 June 2019; Accepted 13 October 2019]

Path Optimization of a Single Surveillance Drone Based on Reinforcement Learning

Path Optimization of a Single Surveillance Drone Based on Reinforcement Learning

A Review of Content-Based Video Retrieval Techniques for Person Identification

The rise of technology spurs the advancement in the surveillance field. Many commercial spaces reduced the patrol guard in favor of Closed-Circuit Television (CCTV) installation and even some countries already used surveillance drone which has greater mobility. In recent years, the CCTV Footage have also been used for crime investigation by law enforcement such as in Boston Bombing 2013 incident. However, this led us into producing huge unmanageable footage collection, the common issue of Big Data era. While there is more information to identify a potential suspect, the massive size of data needed to go over manually is a very laborious task. Therefore, some researchers proposed using Content-Based Video Retrieval (CBVR) method to enable to query a specific feature of an object or a human. Due to the limitations like visibility and quality of video footage, only certain features are selected for recognition based on Chicago Police Department guidelines. This paper presents the comprehensive reviews on CBVR techniques used for clothing, gender and ethnic recognition of the person of interest and how can it be applied in crime investigation. From the findings, the three recognition types can be combined to create a Content-Based Video Retrieval system for person identification.