With the exponential rise in the use of drones anywhere anytime, malicious use by outlaws is increasing as well. This calls for protective, detective, preventive measures to counter these attacks. ...This paper aims to review literature on drone detection and classification that utilizes a myriad of modalities ranging from using thermal infrared sensors to radar detections. In addition, there is a summary of a detailed discussion on drone classification along with recent progress and development in drone detection using machine learning, all of which is performed in an attempt to identify means to thwart such attacks. Furthermore, some future research directions, related to this new field of study, are discussed.
With the widespread adoption of drones in daily life, next-generation smart cities need to establish highways, i.e., trajectories where drones can fly and operate safely. However, due to the ...untrusted nature of their ecosystem, drones might misbehave and take disallowed trajectories, e.g., to reduce the time to fly to a destination, reduce energy consumption, visit unauthorized areas, or disrupt operations of sensitive sites. In this paper, we address the cited problem by proposing ORION, a new framework for online drone trajectory verification. ORION requires one or more receivers distributed in a given area capable of receiving and analyzing standard Remote Identification (RID) messages emitted by operational drones. ORION compares the locations reported in such messages with the closest set of coordinates in the allowed trajectory. It raises an alarm if the distance between such locations exceeds a threshold calibrated offline. We validate the performance of ORION through data collected from both a real drone flight in Amsterdam (Netherlands) and taxi trajectories in Porto (Portugal), achieving a True Positive Ratio (correct detection of disallowed trajectories) up to 0.95 and a False Positive Ratio (incorrect detection of disallowed trajectories) up to 0.04. Our solution significantly outperforms existing approaches used for drone detection or time-series analysis. Finally, we also release the gathered data as open-source to foster future research.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Now that the use of drones is becoming more common, the need to regulate the access to airspace for these systems is becoming more pressing. A necessary tool in order to do this is a means of ...detecting drones. Numerous parties have started the development of such drone detection systems. A big problem with these systems is that the evaluation of the performance of drone detection systems is a difficult operation that requires the careful consideration of all technical and non-technical aspects of the system under test. Indeed, weather conditions and small variations in the appearance of the targets can have a huge difference on the performance of the systems. In order to provide a fair evaluation, it is therefore paramount that a validation procedure that finds a compromise between the requirements of end users (who want tests to be performed in operational conditions) and platform developers (who want statistically relevant tests) is followed. Therefore, we propose in this article a qualitative and quantitative validation methodology for drone detection systems. The proposed validation methodology seeks to find this compromise between operationally relevant benchmarking (by providing qualitative benchmarking under varying environmental conditions) and statistically relevant evaluation (by providing quantitative score sheets under strictly described conditions).
Detecting drones in infrared videos is highly desired in many realistic scenarios, e.g., unauthorized drone monitoring around airports. Nevertheless, automated drone detection is rather challenging ...when the targets appear as tiny objects (≤10×10 pixels) against complex backgrounds. Conventional object detection algorithms, which mainly use static visual features, can hardly distinguish tiny objects from undesired artefacts in complex backgrounds. To alleviate this problem, we learn from the early biological visual pathway (including the parvocellular and magnocellular pathways), which process static and motion information simultaneously. Therefore, we propose a magnocellular inspired method for video tiny-object detection (Magno-VTOD) that integrates both static and motion visual information. The Magno-VTOD firstly employs a retinal magnocellular computation model to extract the motion strength of moving objects. The motion responses are then used to enhance the areas of the flying tiny drones effectively and efficiently, thereby facilitating the subsequent target detection procedure. We implement the video tiny-object detection method based on the widely adopted deep neural networks guided by the magnocellular computation model. Experimental results obtained on the large-scale Anti-UAV dataset (304451 video frames) validate that the proposed Magno-VTOD method significantly outperforms the competing state-of-the-art object detection methods on the tiny drone detection task. Particularly, the AP value is increased by 15.4% for tiny object detection, and by 17.1%/13.7% against wood/mountain backgrounds.
•Drone detection is desired but current methods underperform against complex backgrounds.•A bio-inspired drone detection method utilizing static and motion information is designed.•A retinal magnocellular computation model is used to extract the motion strength.•The motion strength is employed to guide the deep networks for tiny object detection.•The proposed method significantly outperforms the competing state-of-the-art methods.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
The fast development of unmanned aerial vehicles (UAVs), commonly known as drones, has brought a unique set of opportunities and challenges to both the civilian and military sectors. While drones ...have proven useful in sectors such as delivery, agriculture, and surveillance, their potential for abuse in illegal airspace invasions, privacy breaches, and security risks has increased the demand for improved detection and classification systems. This state-of-the-art review presents a detailed overview of current improvements in drone detection and classification techniques: highlighting novel strategies used to address the rising concerns about UAV activities. We investigate the threats and challenges faced due to drones' dynamic behavior, size and speed diversity, battery life, etc. Furthermore, we categorize the key detection modalities, including radar, radio frequency (RF), acoustic, and vision-based approaches, and examine their distinct advantages and limitations. The research also discusses the importance of sensor fusion methods and other detection approaches, including wireless fidelity (Wi-Fi), cellular, and Internet of Things (IoT) networks, for improving the accuracy and efficiency of UAV detection and identification.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Drone object detection is one of the main applications of image processing technology and pattern recognition using deep learning. However, the limited drone image data that can be accessed for ...training detection algorithms is a challenge in the development of drone object detection technology. Therefore, many studies have been conducted to increase the amount of drone image data using data augmentation techniques. This study aims to evaluate the effect of data augmentation on deep learning accuracy in drone object detection using the YOLOv5 algorithm. The methods used in this research include collecting drone image data, augmenting data with rotate, crop and cutout, training the YOLOv5 algorithm with and without data augmentation, as well as testing and analyzing training results.The results of the study show that data augmentation can't improve the accuracy of the YOLOv5 algorithm in drone object detection. Evidenced by the decreasing value of precision and mAP@0.5 and the relatively constant value of recall and F-1 score. This is caused by too much augmentation can cause loss of important information in the data and improper augmentation can cause noise or distortion in the data.
In recent years, popularity of unmanned air vehicles enormously increased due to their autonomous moving capability and applications in various domains. This also results in some serious security ...threats, that needs proper investigation and timely detection of the amateur drones (ADr) to protect the security sensitive institutions. In this paper, we propose the novel machine learning (ML) framework for detection and classification of ADr sounds out of the various sounds like bird, airplanes, and thunderstorm in the noisy environment. To extract the necessary features from ADr sound, Mel frequency cepstral coefficients (MFCC), and linear predictive cepstral coefficients (LPCC) feature extraction techniques are implemented. After feature extraction, support vector machines (SVM) with various kernels are adopted to accurately classify these sounds. The experimental results verify that SVM cubic kernel with MFCC outperform LPCC method by achieving around 96.7% accuracy for ADr detection. Moreover, the results verified that the proposed ML scheme has more than 17% detection accuracy, compared with correlation-based drone sound detection scheme that ignores ML prediction.
This paper presents a comprehensive survey on anti-drone systems. After drones were released for non-military usages, drone incidents in the unarmed population are gradually increasing. However, it ...is unaffordable to construct a military grade anti-drone system for every private or public facility due to installation and operation costs, and regulatory restrictions. We focus on analyzing anti-drone system that does not use military weapons, investigating a wide range of anti-drone technologies, and deriving proper system models for reliable drone defense. We categorized anti-drone technologies into detection, identification, and neutralization, and reviewed numerous studies on each. Then, we propose a hypothetical anti-drone system that presents the guidelines for adaptable and effective drone defense operations. Further, we discuss drone-side safety and security schemes that could nullify current anti-drone methods, and propose future solutions to resolve these challenges.
This paper investigates the time-frequency correlation of spectrogram samples when overlapping windows are considered in the computation of the short-time Fourier transform (STFT). Specifically, a ...general closed-form formula is derived and its properties analyzed. Moreover, the practical usefulness is demonstrated on both simulated and real data, for the problem of small-drones detection based on FMCW radar. Results show that a pre-whitening filter compensating for the correlation can significantly improve the detection performance.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
This paper presents a comprehensive review of current literature on drone detection and classification using machine learning with different modalities. This research area has emerged in the last few ...years due to the rapid development of commercial and recreational drones and the associated risk to airspace safety. Addressed technologies encompass radar, visual, acoustic, and radio-frequency sensing systems. The general finding of this study demonstrates that machine learning-based classification of drones seems to be promising with many successful individual contributions. However, most of the performed research is experimental and the outcomes from different papers can hardly be compared. A general requirement-driven specification for the problem of drone detection and classification is still missing as well as reference datasets which would help in evaluating different solutions.