False Positive Signal Reduction using AI in Counter UAS

Improving radar-based surveillance systems by using background estimation and foreground activity analysis. The system determines persistent objects at radar background locations, filters radar tracks with background motion, and generates alerts based on radar tracks and learned foreground activity patterns. It also fuses radar and vision data to improve scene understanding. The background estimation involves detecting radar objects at fixed locations and removing them. The foreground analysis involves learning typical radar track paths and generating alerts for unusual activity. This helps filter out false positives from moving background objects.

Source

A threat coping system for unmanned aerial vehicles (UAVs) that can autonomously intercept and destroy other UAVs without false positives or misfirings. The system has a control device that sets a no-coping area around targets that shouldn't be engaged. It generates instructions to the threat UAV to only intercept if it won't hit the no-coping area. This prevents collateral damage to nearby aircraft or ground targets. The no-coping area is based on factors like target speed, size, direction, and weapon characteristics.

Source

Detecting and taking countermeasures against unmanned aerial vehicles (UAVs) using AI-assisted object classification and adaptive jamming/spoofing. The system receives signals reflected by objects, conditions them to improve resolution and SNR, then uses AI to classify if the object is a UAV. If so, it takes over control using jamming and spoofing. Jamming disorients the UAV for a time, then spoofing establishes communication to remotely maneuver it. The jamming and spoofing parameters adapt based on range and velocity.

Source

Unauthorized unmanned aerial vehicles (UAVs) pose threats to public safety and individual privacy. Traditional object-detection approaches often fall short during their application in anti-UAV technologies. To address this issue, we propose the YOLOv7-GS model, which is designed specifically for the identification of small UAVs in complex and low-altitude environments. This research primarily aims to improve the models detection capabilities for small UAVs in complex backgrounds. Enhancements were applied to the YOLOv7-tiny model, including adjustments to the sizes of prior boxes, incorporation of the InceptionNeXt module at the end of the neck section, and introduction of the SPPFCSPC-SR and Get-and-Send modules. These modifications aid in the preservation of details about small UAVs and heighten the models focus on them. The YOLOv7-GS model achieves commendable results on the DUT Anti-UAV and the Amateur Unmanned Air Vehicle Detection datasets and performs to be competitive against other mainstream algorithms.

Source

Integrated system for low-altitude drone response that enables rapid identification and mitigation of unauthorized drones. The system combines radar, camera, and communication capabilities to detect and analyze drone presence, determine threat classification, and initiate response actions. The system integrates with a central server to manage drone traffic, including automatic threat detection, tracking, and response capabilities.

Source

A proposed strategy for managing airspace and preventing illegal drones from compromising security involves the use of autonomous drones equipped with three key functionalities. Firstly, the implementation of YOLO-v5 technology allows for the identification of illegal drones and the establishment of a visual-servo system to determine their relative position to the autonomous drone. Secondly, an extended Kalman filter algorithm predicts the flight trajectory of illegal drones, enabling the autonomous drone to compensate in advance and significantly enhance the capture success rate. Lastly, to ensure system robustness and suppress interference from illegal drones, an adaptive fast nonsingular terminal sliding mode technique is employed. This technique achieves finite time convergence of the system state and utilizes delay estimation technology for the real-time compensation of unknown disturbances. The stability of the closed-loop system is confirmed through Lyapunov theory, and a model-based hardware-in-the-loop simulation strategy is adopted to streamline system development and impro... Read More

Source

Unmanned aerial vehicle (UAV) technology has witnessed widespread utilization in target surveillance activities. However, cooperative multiple UAVs for the identification of multiple targets poses a significant challenge due to the susceptibility of individual UAVs to false positive (FP) and false negative (FN) target detections. Specifically, the primary challenge addressed in this study stems from the weak discriminability of features in Synthetic Aperture Radar (SAR) imaging targets, leading to a high false alarm rate in SAR target detection. Additionally, the uncontrollable false alarm rate during electro-optical proximity detection results in an elevated false alarm rate as well. Consequently, a cumulative error propagation problem arises when SAR and electro-optical observations of the same target from different perspectives occur at different times. This paper delves into the target association problem within the realm of collaborative detection involving multiple unmanned aerial vehicles. We first propose an improved triplet loss function to effectively assess the similarity ... Read More

Source

A system for detecting, classifying, and tracking unmanned aerial vehicles (UAVs) across various frequency bands. The system employs advanced AI-driven analysis to automatically detect and classify UAVs, while simultaneously providing direction-of-arrival information. The system utilizes multiple receivers to capture and process RF signals, with AI-powered signal processing and classification algorithms that can handle multiple modulation schemes. The system also enables real-time direction-finding capabilities, including bearing estimation and velocity calculation, and provides location estimation and trajectory prediction.

Source

Aiming at the six-rotor unmanned aerial vehicles subjected to false data injection attacks, an event-triggered-based distributed intelligent learning control strategy is proposed in this paper. The reinforcement learning algorithm which is designed based on neural networks is introduced to achieve intelligent optimal control. Under the actor-critic-identifier frame, three kinds of neural networks are used to realize the control action, evaluate the system performance, and estimate the unknown dynamic, respectively. Then, an improved event-triggered strategy including a decreasing function of the consensus error is designed to reduce the waste of resources while decreasing the adverse impact on tracking control performance as far as possible, which also does not exhibit Zeno behavior. Furthermore, an adaptive compensation control scheme is given, which can effectively compensate for the negative impacts brought by the false data injection attacks existing in information interaction among UAVs. Despite the impact of lumped disturbances and false data injection attacks, the proposed dis... Read More

Source

The efficient detection of "low-altitude, small, slow" UAV is always a difficult problem to overcome. In this paper, we are aiming at the application and threat cases of "low-altitude, small, slow" UAV, analyze the basic principle and development status of several kinds of methods in the detection of this target, and then makes an outlook on the future development direction of "low-altitude, small, slow" UAV detection.

Source

Unmanned aerial vehicles operating in illicit activities, commonly referred to as "black flights," pose an imminent risk regarding public safety and present strategies for detect individuals with such a poor attitude have difficulty striking an appropriate compromise within highest standard of quality as well as quickness.However, its effectiveness lies essence, especially regarding minor components along with detailed contexts.In order to answer those real issues, it offer minimal convolutional neural network including information improvement which will recognize moving creatures at such a lower elevations with high precision while in current moment as well as provide guidance material to suppress black-flying UAVs.Three modules comprise the network to be it has been offered.An innovation and social processing module significantly improves the model's ability to extract low-level features, when a correct detection method integrates low-level and advanced features to boost multi -resolution detection accuracy in complex environments, particularly for tiny objects.Such modules serve t... Read More

Source

A multi-hypothesis tracking system for tracking aerial targets that enables efficient processing of multiple sensor data in high false alarm environments. The system employs a structured branching approach to MHT, where multiple association hypotheses are computed for each new detection, and the least probable associations are omitted. The system maintains multiple reasonable alternatives while postponing final association decisions, enabling selection of the most probable associations within a given time. The system also incorporates bias corrections, cluster tracking, and track segment links to enhance tracking accuracy and robustness.

Source

Detection and classification of small unmanned aerial systems (sUAS) using multi-modal neural networks. The system employs a deep learning approach that combines passive RF detection with spectrographic analysis and sensor fusion. It enables the detection and classification of sUAS based on their RF signatures, radar and acoustic spectrograms, and electro-optical/infrared images. The system also includes a threat assessment component that determines the intent of the sUAS and predicts its frequency hopping sequences. This multi-modal approach enables robust detection and classification of sUAS, particularly in complex environments with non-threat emitters.

Source

In recent years, unmanned aerial vehicles (UAVs) have gained widespread use in both military and civilian fields with the advancement of aviation technology and improved communication capabilities. However, the phenomenon of unauthorized UAV flights, or "black flying", poses a serious threat to the safe flight of aircraft in airspace and public safety. To effectively interfere with and attack UAV targets, it is crucial to enhance the detection and identification of "low, slow and small" UAVs. This study focuses on achieving high-precision and lightweight detection and identification of four-rotor, six-rotor, and fixed-wing UAVs in low-altitude complex environments. By combining deep learning target detection with superresolution feature enhancement, a lightweight UAV detection model is designed and field-tested for verification. To address the challenge of detecting small UAV targets with limited information, the feature fusion network is enhanced based on the traditional YOLOv4 algorithm to improve the detection ability of small targets via small target enhancement and candidate box... Read More

Source

To date, the problems associated with increasing the stability of the flight of an unmanned aerial vehicle (UAV) are becoming quite relevant. This need is due to the presence of many factors that affect the UAV during the flight. Often the UAV is in an aggressive environment, and is also exposed to threats to information security. This study proposes a method for assessing the state of the flight and, on its basis, developing a solution for adjusting the flight parameters. Together, the system will make it possible to make a decision that contributes either to returning to the starting point to rescue the UAV, or to continuing the flight to a given target. A pilot study has shown that when using this UAV system, the target achievement rate increases to 86% when attacking a drone. In this case, detection and response occur with a probability of one hundred percent. Thanks to a new method for estimating the direction of parameter change and a more accurate determination of the boundaries of confidence intervals, the number of false positives is minimized. Thus, the presented software m... Read More

Source

Recently, with the significant rise of drones, reinforcing and securing aerial security and privacy has become an urgent task. Their malicious use takes benefit from the malevolent deployment which leverages some existing gaps in Artificial Intelligence (AI) and cybersecurity. Anti-drone systems are the spotlighted security solution developed to ensure aerial safety and security against rogue drones. However, the anti-drone systems are constraints to accurate airborne target identification and real-time detection to neutralize the target properly without causing damages. In this paper, we have developed a real-time multi-target detection model based on Yolov7 aiming to detect, identify and locate the airborne target properly and rapidly using a varied dataset which is biased and imbalanced due to the differences between the targets. In order to develop a model with the best compromise between a high performance and fast speed, we have used a series of improvements by incorporating the CSPResNeXt module in the backbone, a transformer block with the C3TR attention mechanism and decoupl... Read More

Source

Unmanned Aerial Vehicles (UAVs) have been widely used in many areas, including transportation, surveillance, and military. However, their potential for safety and privacy violations is an increasing issue and highly limits their broader applications, underscoring the critical importance of UAV perception and defense (anti-UAV). Still, previous works have simplified such an anti-UAV task as a tracking problem, where the prior information of UAVs is always provided; such a scheme fails in real-world anti-UAV tasks (i.e. complex scenes, indeterminate-appear and -reappear UAVs, and real-time UAV surveillance). In this paper, we first formulate a new and practical anti-UAV problem featuring the UAVs perception in complex scenes without prior UAVs information. To benchmark such a challenging task, we propose the largest UAV dataset dubbed AntiUAV600 and a new evaluation metric. The AntiUAV600 comprises 600 video sequences of challenging scenes with random, fast, and small-scale UAVs, with over 723K thermal infrared frames densely annotated with bounding boxes. Finally, we develop a novel a... Read More

Source

GPS spoofing attacks are a severe threat to unmanned aerial vehicles. These attacks manipulate the true state of the unmanned aerial vehicles, potentially misleading the system without raising alarms. Several techniques, including machine learning, have been proposed to detect these attacks. Most of the studies applied machine learning models without identifying the best hyperparameters, using feature selection and importance techniques, and ensuring that the used dataset is unbiased and balanced. However, no current studies have discussed the impact of model parameters and dataset characteristics on the performance of machine learning models; therefore, this paper fills this gap by evaluating the impact of hyperparameters, regularization parameters, dataset size, correlated features, and imbalanced datasets on the performance of six most commonly known machine learning techniques. These models are Classification and Regression Decision Tree, Artificial Neural Network, Random Forest, Logistic Regression, Gaussian Nave Bayes, and Support Vector Machine. Thirteen features extracted fr... Read More

Source

In recent years, civilian Unmanned Aerial Vehicles(UAV) have been widely used in various fields, which bringing convenience to people while also having many negative effects. Illegal flying UAVs have affected people's daily life, and there is an urgent need to establish an efficient UAV countermeasure mechanism for critical places such as government, hospitals, universities, and research institutes. To this end, this paper proposes a UAV recognition and tracking method for complex environments, firstly pre-identify UAV targets by YOLOv5, then combine the motion trajectory of targets in continuous images to discard the error outputs, finally use Kernel Correlation Filter to improve the continuous tracking capability. Experiments show that the algorithm in this paper can effectively improve 9.13% precision and the 11.54% recall compared with YOLOv5s network, and the recognition speed can reach to 27.64fps. Tested on the UAV interception platform developed by our team, the experimental results show that the platform deployed with our algorithm can continuously track the targets and inte... Read More

Source

Drones are gaining high popularity for their beneficial use in civilian applications and smart cities. Capable of being structured in networks, they can be used to collect several types of data, such as images, and be sent to centers for further processing. At the same time, they also become a new target for multiple types of attacks, among them False Data Injection (FDI), Denial of Service, GPS Spoofing, etc. Therefore, designing new systems and defense mechanisms against these attacks becomes urgent and necessary. In this paper, we emphasize the dangerous nature of the so-called False Data Injection (FDI) and describe a method based on deep learning for its detection. Considered a severe and powerful attack, an injection of false data into the data (images) collected by the drones can considerably alter a final decision that the processing center may take. To fight against this attack, our proposal relies on image analysis and classification using a deep learning approach. After scaling the received image to fit the classifier, using nearest neighbor interpolation (NNI), we feed a ... Read More

Source

The simulation computer is the main body of the system, and the hardware and software of the system are designed on the basis of considering the two functional requirements of integrated UAV interception and counter-simulation. Firstly, the voices of unmanned aerial vehicles, birds and people within 100 m are collected, preprocessed and extracted with MFCC+GFCC eigenvalues, and their eigenvalues are used as data sets for learning and recognition by convolutional neural networks. Then, the interception model of UAV route planning based on convolutional neural network algorithm and the corresponding deception jamming counter-measure model are expounded. The model effect is simulated and evaluated by computer, and the convergence speed of the algorithm is discussed. The power and optimal deployment position and number of jammers are simulated under the condition that no base station is intercepted in 50 interception routes, thus proving the availability and accuracy of the jammer position planning model.

Source

Unmanned aerial vehicles (UAV), also known as drones have been used for a variety of reasons and the commercial drone market growth is expected to reach remarkable levels in the near future. However, some drone users can mistakenly or intentionally fly into flight paths at major airports, flying too close to commercial aircraft or invading people's privacy. In order to prevent these unwanted events, counter-drone technology is needed to eliminate threats from drones and hopefully they can be integrated into the skies safely. There are various counter-drone methods available in the industry. However, a counter-drone system supported by an artificial intelligence (AI) method can be an efficient way to fight against drones instead of human intervention. In this paper, a deep reinforcement learning (DRL) method has been proposed to counter a drone in a 3D space by using another drone. In a 2D space it is already shown that the deep reinforcement learning method is an effective way to counter a drone. However, countering a drone in a 3D space with another drone is a very challenging task ... Read More

Source

Unmanned aerial vehicles (UAVs) are widely used in aerial reconnaissance, mapping, sports events, etc. However, due to insufficient control technology technologies, small UAV "illegal flight" incidents occur from time to time, posing a serious threat to the safe flight of civil aviation and normal air training of troops, and automated monitoring technology of UAVs is of great significance to air defense security. In this paper, the algorithm for UAV target detection and tracking is studied and improved for scenarios where UAVs are vulnerable to obstruction, change in their own scale, and are difficult to be captured in time when they enter the field of view from the edge. In this paper, a target detection model based on YOLOv4 is trained for UAV samples to achieve automatic response to UAVs could be recognized in the field of view, and it is also called every specified number of frames to update target detection information during the subsequent Spatio-Temporal Context (STC) tracking process. A cross-validation mechanism is completed by combining the YOLOv4 recognition and STC tracki... Read More

Source

The increase in the volume of UAVs has been rapid in the past few years. The utilization of drones has increased considerably in the military and commercial setups, with UAVs of all sizes, shapes, and types being used for various applications, from recreational flying to purpose-driven missions. This development has come with challenges and has been identified as a potential source of operational disruptions leading to various security complications, including threats to Critical Infrastructures (CI). Thus, the need for developing fully autonomous antiUAV Defense Systems (AUDS) hasn't been more imminent than today. To attenuate and nullify the threat posed by the UAVs, either deliberately or otherwise, this paper presents the holistic design and operational prototype of drone detection technology based on visual detection using Digital Image Processing (DIP) and Machine Learning (ML) to detect, track and classify drones accurately. The proposed system uses a background-subtracted frame difference technique for detecting moving objects partnered with a Pan-Tilt tracking system powered... Read More

Source

With the growing use of Unmanned Aerial Vehicles (UAVs) in military and civilian applications, cyber-attacks are increasing significantly. Therefore, detection of attacks becomes indispensable for such systems. In this paper, we focus on the detection of False Data Injection (FDI) attacks in Unmanned Aerial Systems (UASs). Considered to be the most performed attack, an attacker injects fake data into the system in order to disrupt the final decision. To combat this threat, our proposal is built on image analysis and classification. First, we resize the received image in order to adapt it to feed the classifier using the Nearest Neighbor Interpolation (NNI). Second, we train, validate, and test a Convolutional Neural Network (CNN) to perform the image classification. Finally, we compare each classification result classes to a neighborhood using Euclidean distance. Numerical results on the VisDrone dataset demonstrate the efficiency of our proposal under a set of metrics.

Source

The illegal intrusion of drones into airports poses a constant threat to public safety, and existing anti-drone technologies suffer from problems like blind detection zones and target loss. To address the shortcomings of existing methods, this paper proposes a distributed anti-drone system based on YOLOv5. Combined with the characteristics of airport defense UAVs intrusion scenarios, the system implements functions such as automatically targeting and releasing jamming signals to intercept illegal UAVs. In this paper, the YOLO algorithm is used to optimize the system's detection of drones. The mechanical structure is used to achieve automatic targeting, effectively improving detection accuracy. The distributed cluster deployment is used to solve the defects of detection blind area and target loss. This paper provides a deployment idea for airport measures against lightweight UAV equipment through experimental validation, which provides theoretical guidance for future countermeasures against UAVs. The technology can be extended to railway stations and other infrastructures to ensure pu... Read More

Source

Autonomous anticollision of unmanned aerial vehicle (UAV) is one of the key technologies to realize intelligent decision-making and autonomous control, and it is of great significance to improve the flight safety and survivability of UAV in complex environment. Firstly, the UAV autonomous anticollision system configuration is constructed in this paper, and the UAV autonomous anticollision problem and related models are described. Then, the potential collision conflict prediction rules are defined, and a practical three-dimensional collision conflict prediction method is proposed. Finally, the UAV autonomous avoidance decision-making method is designed by using the optimization theory, and the corresponding simple and feasible flight control law is put forward. Numerical simulation results show that the proposed method can ensure the flight safety of UAV by relying on autonomous decision-making and control strategy, so as to realize the autonomous anticollision between a single UAV and non-cooperative dynamic obstacles in three-dimensional airspace.

Source

The autonomous flight of an unmanned aerial vehicle refers to creating a new flight route after self-recognition and judgment when an unexpected situation occurs during the flight. The unmanned aerial vehicle can fly at a high speed of more than 60 km/h, so obstacle recognition and avoidance must be implemented in real-time. In this paper, we propose to recognize objects quickly and accurately by effectively using the H/W resources of small computers mounted on industrial unmanned air vehicles. Since the number of pixels in the image decreases after the resizing process, filtering and object resizing were performed according to the altitude, so that quick detection and avoidance could be performed. To this end, objects up to 60 m in height were classified by subdividing them at 20 m intervals, and objects unnecessary for object detection were filtered with deep learning methods. In the 40 m to 60 m sections, the average speed of recognition was increased by 38%, without compromising the accuracy of object detection.

Source

Unmanned aerial vehicles (UAV) "black flight" incidents cause serious security risks and economic losses to airports. Additionally, existing UAV detection methods are mainly radar technology at airports. Whereas, it is unable to correctly identify the number of UAVs and visualise their size, which undoubtedly poses a serious security risk. Accordingly, an anchor-free UAV detection method HollowBox is proposed to supplement radar detection equipment for the airport "black flight" problem. It is inspired by the FoveaBox object detection method, the object detection feature layers are reset and the allocation ratio of positive and negative samples in the training phase are redefined, the HollowBox UAV detection is proposed according to the multi-size characteristics. Extensive experiments show that, the approach achieves 90.1% AP, 6% false detection rate and 17.2 FPS inference speed, which accomplished a satisfactory performance, as verified by the real-shot data collected of an airport in Tianjin. This work is of great significance for the application in airport UAV detection.

Source

The emergence of Unmanned Aerial Vehicles (UAVs) raised multiple concerns, given their potentially malicious misuse in unlawful acts. Vision-based counter-UAV applications offer a reliable solution compared to acoustic and radio frequency-based solutions because of their high detection accuracy in diverse weather conditions. The existing solutions work well on trained datasets, but their accuracy is relatively low for real-time detection. In this paper, we model deep learning-empowered solutions to improve the multiclass UAV's classification performance using single-shot object detection algorithms (YOLOv5 and YOLOv7). They efficiently and correctly differentiate between multirotor, fixed-wing, and single-rotor UAVs in challenging weather conditions. Experiments show that the suggested technique is reliable with an overall best average-classification precision of 86.7\%, 88.5\% average recall, 91.8\% average mAP, and 58.4\% average-IoU.

Source

The year 2022 showed an urgent need to improve the existing systems for recognizing objects in the aerial space, which is caused by the significant increase in the number of technical means (especially unmanned aerial vehicles) on the battlefield. Such a sharp increase in the number of objects that simultaneously take part in combat operations in the air requires the improvement of military object recognition systems, both qualitatively and quantitatively. This requires the development of appropriate new generation Friend-or-Foe algorithms for the objects recognition. The main requirements for recognition systems of aerial objects of civil application were determined. They includes: maximum com- patibility; support for a large number of objects; outdated recognition complexes support; support for alternative ways of recognition; support for alternative data entry methods; determining the coordinates of aerial objects in an emergency situation. Friend-or-foe recognition systems for military applications are also considered. In contrast to civilian systems, the following basic require... Read More

Source

Unmanned aerial vehicles (UAVs) are unmanned aircrafts operated by radio remote control and programmed control equipment. Due to their small size, low cost, and high flexibility, UAV s are widely used in military and civilian fields. Along with the broad applications of UAVs and the significant advancement of information technologies, they also face cyber threats. Among them, false spoofing is a typical cyber-attack. If this attack hits UAV s, it could result in damage to property and the release of private data or classified documents. In this paper, we propose the Double Shortcuts Zero-Bias Residual Network (Double Shortcuts ZB-ResNet) with small storage capacity and low time complexity for abnormal data detection of UAVs. It is designed by combining double shortcuts residual blocks and Zero-Bias (ZB) fully connected layer to anomaly detections. By comparing with the experimental results of the improved Convolutional Neural Networks with ZB layer, the detection accuracy of the Double Shortcuts ZB-ResNet model is improved by nearly two percentage points.

Source

Aiming at the difficulty of target location and classification in low altitude surveillance and anti Unmanned Aerial Vehicle(UAV) system. This paper mainly studies the recognition algorithm of low-altitude UAV. First, the UAV image is preprocessed, and candidate partial images of different sizes and positions are generated through a sliding window, and the moment invariant features of the image are extracted. Then, the neural network is used to train the image, and the support vector machine classifier is used to classify the aircraft, and then the recognition and classification algorithm of the aircraft target in the low-altitude airspace are finished. Based on the theoretical algorithm research, this paper uses MATLAB software to simulate and analyze the aircraft recognition algorithm, and the accuracy is more than 90%. The results show that the research algorithm can be used for UAV recognition and low-altitude aircraft classification.

Source

Advances in wireless networks and the rapid development of electronic components have actively contributed to the emergence of new communication and surveillance systems known as Unmanned Aerial Systems (UASs). In such systems, unmanned aerial vehicles (UAVs) can be used as a wireless ad hoc network and thus provide a communications infrastructure for diverse military or civil applications. For more efficiency, a swarm of drones can be deployed in an area of interest (e.g. disaster areas, battlefields) by forming a flying ad hoc network (FANET) capable of communicating wirelessly with a ground control station (GCS) in a more secure manner. In this work, we particularly focus on the detection of False Data Injection (FDI) attacks in Unmanned Aerial Systems. We propose a new approach based on mobile agents to collect data and an artificial neural network model to identify injected false data. Our approach is validated using realistic datasets, provided by the University of Minnesota UAS Laboratories, and our results show that our proposal outperforms the compared approach by demonstrat... Read More

Source

Unmanned aerial vehicles (UAVs) can provide great value in off-shore operations that require aerial surveillance, for example by detecting objects on the water surface. For efficient operations by autonomous aerial surveillance, a reliable automatic detection system must be in place: one that will limit the amount of false negatives, but not at the expense of too many false positives. In this paper, we assess multiple aspects of the detection system that may provide significant impact in off-shore aerial surveillance: First by assessing detection architectures based on convolutional neural networks, then by adding tracking algorithms to utilize temporal information, and finally by investigating the use of different imaging modalities. Through a comparison of several detection models, the experiments prove that misclassification of objects is a particular issue, where input resolution and size of objects influence the overall model performance. The use of a tracking algorithm allows for decreasing the confidence threshold, which results in fewer false negatives, without a significant ... Read More

Source

Detection, identification, and countermeasures for unmanned aerial vehicles (UAVs) employing novel communication protocols through a distributed sensor network. The system employs cognitive radio principles to analyze and adapt to unknown communication protocols, leveraging machine learning algorithms to generate profiles of detected UAVs. By analyzing sensor data and network behavior, the system identifies patterns indicative of specific communication protocols and generates countermeasures tailored to those protocols. The system can operate in both centralized and edge environments, enabling real-time detection and response capabilities against UAVs employing unconventional communication protocols.

Source

The combination of unmanned aerial vehicles (UAV) with deep learning models has the capacity to replace manned aircrafts for wildlife surveys. However, the scarcity of animals in the wild often leads to highly unbalanced, large datasets for which even a good detection method can return a large amount of false detections. Our objectives in this paper were to design a training method that would reduce training time, decrease the number of false positives and alleviate the fine-tuning effort of an image classifier in a context of animal surveys. We acquired two highly unbalanced datasets of deer images with a UAV and trained a Resnet-18 classifier using hard-negative mining and a series of recent techniques. Our method achieved sub-decimal false positive rates on two test sets (1 false positive per 19,162 and 213,312 negatives respectively), while training on small but relevant fractions of the data. The resulting training times were therefore significantly shorter than they would have been using the whole datasets. This high level of efficiency was achieved with little tuning effort an... Read More

Source

Unauthorized operations referred to as "black flights" of unmanned aerial vehicles (UAVs) pose a significant danger to public safety, and existing low-attitude object detection algorithms encounter difficulties in balancing detection precision and speed.Additionally, their accuracy is insufficient, particularly for small objects in complex environments.To solve these problems, we propose a lightweight feature-enhanced convolutional neural network able to perform detection with high precision detection for low-attitude flying objects in real time to provide guidance information to suppress black-flying UAVs.The proposed network consists of three modules.A lightweight and stable feature extraction module is used to reduce the computational load and stably extract more low-level feature, an enhanced feature processing module significantly improves the feature extraction ability of the model, and an accurate detection module integrates low-level and advanced features to improve the multiscale detection accuracy in complex environments, particularly for small objects.The proposed method a... Read More

Source

Abstract The number of unmanned aerial vehicles (UAVs, also known as drones) has increased dramatically in the airspace worldwide for tasks such as surveillance, reconnaissance, shipping and delivery. However, a small number of them, acting maliciously, can raise many security risks. Recent Artificial Intelligence (AI) capabilities for object detection can be very useful for the identification and classification of drones flying in the airspace and, in particular, are a good solution against malicious drones. A number of counter-drone solutions are being developed, but the cost of drone detection ground systems can also be very high, depending on the number of sensors deployed and powerful fusion algorithms. We propose a low-cost counter-drone solution composed uniquely by a guard-drone that should be able to detect, locate and eliminate any malicious drone. In this paper, a state-of-the-art object detection algorithm is used to train the system to detect drones. Three existing object detection models are improved by transfer learning and tested for real-time drone detection. Trainin... Read More

Source

The growing range of possibilities provided by the proliferation of commercial unmanned aerial vehicles, or drones, raises alarming safety and security threats. The efficient mitigation of these threats depends on authorities having defense systems to counter both accidentally trespassing and maliciously operated drones. To effectively counter such vehicles, defense systems must be able to detect a new drone entering a restricted airspace; locate its position; identify its purpose; and, should the identification procedure mark it as a threat, neutralize it.

Source

The rapid development of UAV has brought great convenience to various application fields. In the meanwhile, its extensive utilization has also resulted in many problems such as public safety hazards, personal security threats and personal privacy violations. UAV is difficult to capture in real-time because of its small scale and complex flight environment. In order to solve the above problems from the perspective of security protection, a low-cost UAV detection, distance measure and protection scheme are proposed based on deep learning in this paper. The influences of different loss functions and thresholds are studied on the detection accuracy of YOLOv4 to improve the detection performance of YOLOv4 on UAV. At the same time, in order to achieve more effective prevention and control of UAV, the monocular ranging method based on PnP is introduced to get the distance between camera and UAV. Finally, the study is applied in the real-world scene, and a good target detection and ranging effect have been achieved so that the proposed model is verified in the feasibility and effectiveness.

Source

Unmanned Aerial Vehicles (UAVs) have become a major part of everyday life, as well as an emerging research field, by establishing their versatility in a variety of applications. Nevertheless, this rapid spread of UAVs reputation has provoked serious security issues that can probably affect homeland security. Defence communities have started to investigate large field-of-view sensor-based methods to enable various civil protection applications, including the detection and localisation of flying threat objects. Counter-UAV (c-UAV) detection challenges may be granted from a fusion of sensors to enhance the confidence of flying threats identification. The real-time monitoring of the environment is absolutely rigorous and demands accurate methods to detect promptly the occurrence of harmful conditions. Deep learning (DL) based techniques are capable of tackling the challenges that are associated with generic objects detection and explicitly UAV identification. In this paper, we present a novel multimodal DL methodology that combines data from individual unimodal approaches that are associ... Read More

Source

In recent years, unmanned aerial vehicle technology has been continuously developed and matured, and the level of intelligence has been continuously improved. "Black Flying" brings instability to social security, while UAVs bring convenience to routine production and daily life. In addition, with the continuous development of UAV cluster technology, the future trend of air combat will gradually turn to cluster cooperation. UAV cluster combat will also be changed from concept to reality, from theory to practice. The outstanding performance of artificial intelligence technology in game tasks clearly shows that strategy based on human experience will be difficult to compete with intelligent algorithms in future confrontations. Considering the future demand for multi-UAVs prevention and control, this paper relies on advanced intelligent technologies such as genetic fuzzy trees, multiobjective particle swarm optimization algorithms, reinforcement learning and deep neural networks, aiming at the complex, dynamic, and strong interference, focusing on researching core issues such as the cons... Read More

Source

Nowadays, unmanned aerial vehicles (UAVs) have achieved massive improvement, which brings great convenience and advantage. Meanwhile, threats posed by them may damage public security and personal safety. This article proposes an architecture of intelligent anti-UAVs low-altitude defense system. To address the key problem of discovering UAVs, research based on multisensor information fusion is carried out. Firstly, to solve the problem of probing suspicious targets, a fusion method is designed, which combines radar and photoelectric information. Subsequently, single shot multibox detector model is introduced to identify UAV from photoelectric images. Moreover, improved spatially regularized discriminative correlation filters algorithm is used to elevate real-time and stability performance of system. Finally, experimental platform is constructed to demonstrate the effectiveness of the method. Results show better performance in range, accuracy, and success rate of surveillance.

Source

A radar system determines the presence of a UAV based on micro-Doppler signatures from its rotors, which exhibit distinct velocity offsets relative to the UAV's body. The system analyzes reflected signals from multiple components of the detected object to identify the characteristic velocity patterns associated with UAVs.

Source

Counter-drone technology by using artificial intelligence (AI) is an emerging technology and it is rapidly developing. Considering the recent advances in AI, counter-drone systems with AI can be very accurate and efficient to fight against drones. The time required to engage with the target can be less than other methods based on human intervention, such as bringing down a malicious drone by a machine-gun. Also, AI can identify and classify the target with a high precision in order to prevent a false interdiction with the targeted object. We believe that counter-drone technology with AI will bring important advantages to the threats coming from some drones and will help the skies to become safer and more secure. In this study, a deep reinforcement learning (DRL) architecture is proposed to counter a drone with another drone, the learning drone, which will autonomously avoid all kind of obstacles inside a suburban neighborhood environment. The environment in a simulator that has stationary obstacles such as trees, cables, parked cars, and houses. In addition, another non-malicious thi... Read More

Source

A method and device for detecting target signals in unmanned aerial vehicles (UAVs) that adaptively adjusts the signal threshold based on the UAV's flight altitude. When the UAV is at a high altitude, the threshold is lowered to reduce missing alarms from small objects, while at low altitudes, the threshold is raised to reduce false alarms from ground features. The method uses a sliding window detector to analyze neighboring signals and determine a specific threshold for each test signal, enabling more accurate detection of target objects while reducing false alarms.

Source

Two typical Unmanned Aerial Vehicles (UAV) countermeasures involve the detection and tracking of the UAV position, as well as of the human pilot; they are of critical importance before taking any countermeasure, and they already obtained strong attention from national security agencies in different countries.Recent advances in computer vision and artificial intelligence are already proposing many visual detection systems from an operating UAV, but they do not focus on the problem of the detection of the pilot of another approaching unauthorized UAV.In this work, a first attempt of proposing a full autonomous pipeline to process images from a flying UAV to detect the pilot of an unauthorized UAV entering a no-fly zone is introduced.A challenging video sequence has been created flying with a UAV in an urban scenario and it has been used for this preliminary evaluation.Experiments show very encouraging results in terms of recognition, and a complete dataset to evaluate artificial intelligence-based solution will be prepared.

Source

Since unmanned aerial vehicles (UAVs) can be controlled remotely in the absence of a unified means of identity authentication, they are quite vulnerable for illegal control by unidentifiable users. Only by tracing the identity of UAV itself, or providing management to pilots, current UAV identity authentication mechanism is far from achieving "single machine for single person." With the development of artificial intelligence, it is possible to achieve automatic UAV identification. Therefore, this article proposes a behavior-based intelligent UAV identification and security supervision. Based on location tracking and flying data acquisition provided by the airborne black box, the UAV's behavioral data are collected on real time. Then, a reliable identification of UAVs is completed through the behavioral modeling, and a warning is issued in the potential illegal cases. It provides the government with intelligent control and disposal decision basis for flying UAVs.

Source