Recent years have seen the dramatic rise of the use of Unmanned Aerial Systems or drones by governments, consumers and – unfortunately – also by terrorists and criminals. Indeed, whereas there are a great number of very good applications for the use of drones, these new technological tools provide also a threat in the hands of people with bad intentions. Examples are the use of unmanned aerial robotic vehicles or drones for operations such as illegal observation and surveillance and drugs trafficking, or even as attack vector. Currently, it is very difficult for law enforcement and border management authorities to deal with these new threats, as the Radar Cross Section of these drones is too small to be detected by regular radar systems.
Several novel detection modalities are being researched to tackle this problem: RADAR, LIDAR, Acoustic Sensing, Radio Sensing, Thermal Sensing and Visual sensing. However, no individual sensing modality currently attains satisfying levels of accuracy. The European Commission noted this capability gap and decided to fund the H2020-SafeShore project, which has as a main goal to cover existing gaps in coastal border surveillance, increasing internal security by preventing cross-border crime such as trafficking in human beings and the smuggling of drugs.
The SafeShore core solution for detecting small targets that are flying at low altitude is to use a 3D LIDAR that scans the sky and creates above the protected area a virtual dome shield. 3D LIDAR will be integrated with passive acoustic sensors, passive radio detection and video analytics. With this 3D LIDAR technology, the SafeShore system is able to cover a vast area of the coastal border using mobile platforms. Each mobile platform covers its area with a dome-shaped virtual detection shield with a radius of about 250-300m. Overlapping multiple platforms creates a continuous detection shield along the shoreline.
Project Video Gallery
Project Publications
2019
- G. De Cubber, “Opportunities and threats posed by new technologies," in SciFi-IT, Ghent, Belgium, 2019.
[BibTeX] [Abstract] [Download PDF] [DOI]
The technological evolution is introducing in a fast pace new technologies in our everyday lives. As always, these new technologies can be applied for good causes and thereby give us the opportunity to do many interesting new things. Think for example about drones transporting blood samples between hospitals. However, like always, new technologies can also be applied for bad causes. Think for example about the same drones, but this time transporting bomb parcels instead of blood. In this paper, we focus on a number of novel technologies and discuss how security actors are currently doing their best to maximize the good use of these tools while minimizing the bad use. We will focus on research actions taken by Belgian Royal Military Academy in the domains of: – Augmented reality, and showcase how this technology can be used to improve surveillance operations. – Unmanned Aerial Systems (Drones), and showcase how the potential security threats posed by these systems can be mitigated by novel drone detection systems. – Unmanned Maritime Systems, and showcase how this technology can be used to increase the safety at sea. – Unmanned Ground Systems, and more specifically the autonomous cars, showcasing how to prevent potential cyber-attacks on these future transportation tools.
@InProceedings{de2019opportunities, author = {De Cubber, Geert}, booktitle = {SciFi-IT}, title = {Opportunities and threats posed by new technologies}, year = {2019}, abstract = {The technological evolution is introducing in a fast pace new technologies in our everyday lives. As always, these new technologies can be applied for good causes and thereby give us the opportunity to do many interesting new things. Think for example about drones transporting blood samples between hospitals. However, like always, new technologies can also be applied for bad causes. Think for example about the same drones, but this time transporting bomb parcels instead of blood. In this paper, we focus on a number of novel technologies and discuss how security actors are currently doing their best to maximize the good use of these tools while minimizing the bad use. We will focus on research actions taken by Belgian Royal Military Academy in the domains of: - Augmented reality, and showcase how this technology can be used to improve surveillance operations. - Unmanned Aerial Systems (Drones), and showcase how the potential security threats posed by these systems can be mitigated by novel drone detection systems. - Unmanned Maritime Systems, and showcase how this technology can be used to increase the safety at sea. - Unmanned Ground Systems, and more specifically the autonomous cars, showcasing how to prevent potential cyber-attacks on these future transportation tools.}, doi = {https://doi.org/10.5281/zenodo.2628758}, address = {Ghent, Belgium}, project = {MarSur,SafeShore}, url = {http://mecatron.rma.ac.be/pub/2019/Sci-Fi-It-2019-DeCubber (2).pdf}, unit= {meca-ras} } - G. De Cubber, “Explosive drones: How to deal with this new threat?," in International workshop on Measurement, Prevention, Protection and Management of CBRN Risks (RISE), Les Bon Villers, Belgium, 2019.
[BibTeX] [Abstract] [Download PDF] [DOI]
As the commercial and recreative use of small unmanned aerial vehicles or drones is booming, so are the military and criminals starting to use these systems more and more. Due to improvements in flight stability, autonomy and payload capacity it becomes possible to equip these drones with explosive charges, making them threat agents where traditional response mechanisms have few answers against. In this paper, we will discuss this new type of threat in detail, making the difference between the loitering munition, as used by regular armies and the traditional drones equipped with explosive charges, used in guerrilla warfare and by criminals. We will then discuss what research actions are currently being undertaken to provide answers to each of these threats and what countermeasures that are currently already available and which ones will be available in the near future.
@InProceedings{de2019explosive, author = {De Cubber, Geert}, booktitle = {International workshop on Measurement, Prevention, Protection and Management of CBRN Risks (RISE)}, title = {Explosive drones: How to deal with this new threat?}, year = {2019}, number = {9}, address = {Les Bon Villers, Belgium}, abstract = {As the commercial and recreative use of small unmanned aerial vehicles or drones is booming, so are the military and criminals starting to use these systems more and more. Due to improvements in flight stability, autonomy and payload capacity it becomes possible to equip these drones with explosive charges, making them threat agents where traditional response mechanisms have few answers against. In this paper, we will discuss this new type of threat in detail, making the difference between the loitering munition, as used by regular armies and the traditional drones equipped with explosive charges, used in guerrilla warfare and by criminals. We will then discuss what research actions are currently being undertaken to provide answers to each of these threats and what countermeasures that are currently already available and which ones will be available in the near future.}, doi = {10.5281/ZENODO.2628752}, project = {SafeShore}, url = {http://mecatron.rma.ac.be/pub/2019/Explosive drones - How to deal with this new threat.pdf}, unit= {meca-ras} } - I. Lahouli, Z. Chtourou, M. A. Ben Ayed, R. Haelterman, G. De Cubber, and R. Attia, “Pedestrian Detection and Trajectory Estimation in the Compressed Domain Using Thermal Images," in Computer Vision, Imaging and Computer Graphics Theory and Applications, Springer, 2019, p. 212–227.
[BibTeX] [Abstract] [Download PDF] [DOI]
Since a few decades, the Unmanned Aerial Vehicles (UAVs) are considered precious tools for different military applications such as the automatic surveillance in outdoor environments. Nevertheless, the onboard implementation of image and video processing techniques poses many challenges like the high computational cost and the high bandwidth requirements, especially on low-performance processing platforms like small or medium UAVs. A fast and efficient framework for pedestrian detection and trajectory estimation for outdoor surveillance using thermal images is presented in this paper. First, the detection process is based on a conjunction between contrast enhancement techniques and saliency maps as a hotspot detector, on Discrete Chebychev Moments (DCM) as a global image content descriptor and on a linear Support Vector Machine (SVM) as a classifier. Second, raw H.264/AVC compressed video streams with limited computational overhead are exploited to estimate the trajectories of the detected pedestrians. In order to simulate suspicious events, six different scenarios were carried out and filmed using a thermal camera. The obtained results show the effectiveness and the low computational requirements of the proposed framework which make it suitable for real-time applications and onboard implementation.
@InCollection{lahouli2019pedestrian, author = {Lahouli, Ichraf and Chtourou, Zied and Ben Ayed, Mohamed Ali and Haelterman, Rob and De Cubber, Geert and Attia, Rabah}, booktitle = {Computer Vision, Imaging and Computer Graphics Theory and Applications}, publisher = {Springer}, title = {Pedestrian Detection and Trajectory Estimation in the Compressed Domain Using Thermal Images}, year = {2019}, pages = {212--227}, abstract = {Since a few decades, the Unmanned Aerial Vehicles (UAVs) are considered precious tools for different military applications such as the automatic surveillance in outdoor environments. Nevertheless, the onboard implementation of image and video processing techniques poses many challenges like the high computational cost and the high bandwidth requirements, especially on low-performance processing platforms like small or medium UAVs. A fast and efficient framework for pedestrian detection and trajectory estimation for outdoor surveillance using thermal images is presented in this paper. First, the detection process is based on a conjunction between contrast enhancement techniques and saliency maps as a hotspot detector, on Discrete Chebychev Moments (DCM) as a global image content descriptor and on a linear Support Vector Machine (SVM) as a classifier. Second, raw H.264/AVC compressed video streams with limited computational overhead are exploited to estimate the trajectories of the detected pedestrians. In order to simulate suspicious events, six different scenarios were carried out and filmed using a thermal camera. The obtained results show the effectiveness and the low computational requirements of the proposed framework which make it suitable for real-time applications and onboard implementation.}, doi = {10.1007/978-3-030-26756-8_10}, project = {SafeShore}, url = {https://www.springerprofessional.de/en/pedestrian-detection-and-trajectory-estimation-in-the-compressed/16976092}, unit= {meca-ras} } - I. Lahouli, R. Haelterman, Z. Chtourou, G. De Cubber, and R. Attia, “Pedestrian Tracking in the Compressed Domain Using Thermal Images," in Representations, Analysis and Recognition of Shape and Motion from Imaging Data, Communications in Computer and Information Science, Springer International Publishing, 2019, vol. 842, p. 35–44.
[BibTeX] [Abstract] [Download PDF] [DOI]
The video surveillance of sensitive facilities or borders poses many challenges like the high bandwidth requirements and the high computational cost. In this paper, we propose a framework for detecting and tracking pedestrians in the compressed domain using thermal images. Firstly, the detection process uses a conjunction between saliency maps and contrast enhancement techniques followed by a global image content descriptor based on Discrete Chebychev Moments (DCM) and a linear Support Vector Machine (SVM) as a classifier. Secondly, the tracking process exploits raw H.264 compressed video streams with limited computational overhead. In addition to two, well-known, public datasets, we have generated our own dataset by carrying six different scenarios of suspicious events using a thermal camera. The obtained results show the effectiveness and the low computational requirements of the proposed framework which make it suitable for real-time applications and onboard implementation.
@InCollection{lahouli2019pedestriantracking, author = {Lahouli, Ichraf and Haelterman, Rob and Chtourou, Zied and De Cubber, Geert and Attia, Rabah}, booktitle = {Representations, Analysis and Recognition of Shape and Motion from Imaging Data, Communications in Computer and Information Science}, publisher = {Springer International Publishing}, title = {Pedestrian Tracking in the Compressed Domain Using Thermal Images}, year = {2019}, pages = {35--44}, volume = {842}, abstract = {The video surveillance of sensitive facilities or borders poses many challenges like the high bandwidth requirements and the high computational cost. In this paper, we propose a framework for detecting and tracking pedestrians in the compressed domain using thermal images. Firstly, the detection process uses a conjunction between saliency maps and contrast enhancement techniques followed by a global image content descriptor based on Discrete Chebychev Moments (DCM) and a linear Support Vector Machine (SVM) as a classifier. Secondly, the tracking process exploits raw H.264 compressed video streams with limited computational overhead. In addition to two, well-known, public datasets, we have generated our own dataset by carrying six different scenarios of suspicious events using a thermal camera. The obtained results show the effectiveness and the low computational requirements of the proposed framework which make it suitable for real-time applications and onboard implementation.}, doi = {10.1007/978-3-030-19816-9_3}, project = {SafeShore}, url = {https://app.dimensions.ai/details/publication/pub.1113953804}, unit= {meca-ras} } - D. Doroftei and G. De Cubber, “Using a qualitative and quantitative validation methodology to evaluate a drone detection system," ACTA IMEKO, vol. 8, iss. 4, p. 20–27, 2019.
[BibTeX] [Abstract] [Download PDF] [DOI]
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).
@Article{doroftei2019using, author = {Doroftei, Daniela and De Cubber, Geert}, journal = {{ACTA} {IMEKO}}, title = {Using a qualitative and quantitative validation methodology to evaluate a drone detection system}, year = {2019}, month = dec, number = {4}, pages = {20--27}, volume = {8}, abstract = {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).}, doi = {10.21014/acta_imeko.v8i4.682}, pdf = {https://acta.imeko.org/index.php/acta-imeko/article/view/IMEKO-ACTA-08%20%282019%29-04-05/pdf}, project = {SafeShore}, publisher = {{IMEKO} International Measurement Confederation}, url = {https://acta.imeko.org/index.php/acta-imeko/article/view/IMEKO-ACTA-08%20%282019%29-04-05/pdf}, unit= {meca-ras} } - A. Coluccia, A. Fascista, A. Schumann, L. Sommer, M. Ghenescu, T. Piatrik, G. De Cubber, M. Nalamati, A. Kapoor, M. Saqib, N. Sharma, M. Blumenstein, V. Magoulianitis, D. Ataloglou, A. Dimou, D. Zarpalas, P. Daras, C. Craye, S. Ardjoune, D. De la Iglesia, M. Mández, R. Dosil, and I. González, “Drone-vs-Bird Detection Challenge at IEEE AVSS2019," in 2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), 2019, pp. 1-7.
[BibTeX] [Abstract] [Download PDF] [DOI]
This paper presents the second edition of the “drone-vs-bird” detection challenge, launched within the activities of the 16-th IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS). The challenge’s goal is to detect one or more drones appearing at some point in video sequences where birds may be also present, together with motion in background or foreground. Submitted algorithms should raise an alarm and provide a position estimate only when a drone is present, while not issuing alarms on birds, nor being confused by the rest of the scene. This paper reports on the challenge results on the 2019 dataset, which extends the first edition dataset provided by the SafeShore project with additional footage under different conditions.
@INPROCEEDINGS{8909876, author={A. {Coluccia} and A. {Fascista} and A. {Schumann} and L. {Sommer} and M. {Ghenescu} and T. {Piatrik} and G. {De Cubber} and M. {Nalamati} and A. {Kapoor} and M. {Saqib} and N. {Sharma} and M. {Blumenstein} and V. {Magoulianitis} and D. {Ataloglou} and A. {Dimou} and D. {Zarpalas} and P. {Daras} and C. {Craye} and S. {Ardjoune} and D. {De la Iglesia} and M. {Mández} and R. {Dosil} and I. {González}}, booktitle={2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)}, title={Drone-vs-Bird Detection Challenge at IEEE AVSS2019}, year={2019}, volume={}, number={}, pages={1-7}, project = {SafeShore,MarSur}, doi = {10.1109/AVSS.2019.8909876}, abstract = {This paper presents the second edition of the “drone-vs-bird” detection challenge, launched within the activities of the 16-th IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS). The challenge's goal is to detect one or more drones appearing at some point in video sequences where birds may be also present, together with motion in background or foreground. Submitted algorithms should raise an alarm and provide a position estimate only when a drone is present, while not issuing alarms on birds, nor being confused by the rest of the scene. This paper reports on the challenge results on the 2019 dataset, which extends the first edition dataset provided by the SafeShore project with additional footage under different conditions.}, url = {https://ieeexplore.ieee.org/abstract/document/8909876}, unit= {meca-ras} }
2018
- I. Lahouli, R. Haelterman, Z. Chtourou, G. De Cubber, and R. Attia, “Pedestrian Detection and Tracking in Thermal Images from Aerial MPEG videos," in International Conference on Computer Vision Theory and Applications, Funchal, Portugal, 2018, p. 487–495.
[BibTeX] [Abstract] [Download PDF] [DOI]
Video surveillance for security and intelligence purposes has been a precious tool as long as the technology has been available but is computationally heavy. In this paper, we present a fast and efficient framework for pedestrian detection and tracking using thermal images. It is designed for automatic surveillance applications in an outdoor environment like preventing border intrusions or attacks on sensitive facilities using image and video processing techniques implemented on-board Unmanned Aerial Vehicles (UAV)s. The proposed framework exploits raw H.264 compressed video streams with limited computational overhead. Our work is driven by the fact that Motion Vectors (MV) are an integral part of any video compression technique, by day and night capabilities of thermal sensors and the distinguished thermal signature of humans. Six different scenarios were carried out and filmed using a thermal camera in order to simulate suspicious events. The obtained results show the effectiveness of the proposed framework and its low computational requirements which make it adequate for on-board processing and real-time applications.
@InProceedings{lahouli2018pedestrian, author = {Lahouli, Ichraf and Haelterman, Robby and Chtourou, Zied and De Cubber, Geert and Attia, Rabah}, booktitle = {International Conference on Computer Vision Theory and Applications}, title = {Pedestrian Detection and Tracking in Thermal Images from Aerial {MPEG} videos}, year = {2018}, organization = {DOI 10.5220/0006723704870495}, pages = {487--495}, publisher = {{SCITEPRESS} - Science and Technology Publications}, volume = {1}, abstract = {Video surveillance for security and intelligence purposes has been a precious tool as long as the technology has been available but is computationally heavy. In this paper, we present a fast and efficient framework for pedestrian detection and tracking using thermal images. It is designed for automatic surveillance applications in an outdoor environment like preventing border intrusions or attacks on sensitive facilities using image and video processing techniques implemented on-board Unmanned Aerial Vehicles (UAV)s. The proposed framework exploits raw H.264 compressed video streams with limited computational overhead. Our work is driven by the fact that Motion Vectors (MV) are an integral part of any video compression technique, by day and night capabilities of thermal sensors and the distinguished thermal signature of humans. Six different scenarios were carried out and filmed using a thermal camera in order to simulate suspicious events. The obtained results show the effectiveness of the proposed framework and its low computational requirements which make it adequate for on-board processing and real-time applications.}, doi = {10.5220/0006723704870495}, project = {SafeShore}, address = {Funchal, Portugal}, url = {https://www.scitepress.org/Papers/2018/67237/67237.pdf}, unit= {meca-ras} } - I. Lahouli, R. Haelterman, J. Degroote, M. Shimoni, G. De Cubber, and R. Attia, “Accelerating existing non-blind image deblurring techniques through a strap-on limited-memory switched Broyden method," IEICE TRANSACTIONS on Information and Systems, vol. 1, iss. 1, p. 8, 2018.
[BibTeX] [Abstract] [Download PDF] [DOI]
Video surveillance from airborne platforms can suffer from many sources of blur, like vibration, low-end optics, uneven lighting conditions, etc. Many different algorithms have been developed in the past that aim to recover the deblurred image but often incur substantial CPU-time, which is not always available on-board. This paper shows how a strap-on quasi-Newton method can accelerate the convergence of existing iterative methods with little extra overhead while keeping the performance of the original algorithm, thus paving the way for (near) real-time applications using on-board processing.
@Article{lahouli2018accelerating, author = {Lahouli, Ichraf and Haelterman, Robby and Degroote, Joris and Shimoni, Michal and De Cubber, Geert and Attia, Rabah}, journal = {IEICE TRANSACTIONS on Information and Systems}, title = {Accelerating existing non-blind image deblurring techniques through a strap-on limited-memory switched {Broyden} method}, year = {2018}, number = {1}, pages = {8}, volume = {1}, abstract = {Video surveillance from airborne platforms can suffer from many sources of blur, like vibration, low-end optics, uneven lighting conditions, etc. Many different algorithms have been developed in the past that aim to recover the deblurred image but often incur substantial CPU-time, which is not always available on-board. This paper shows how a strap-on quasi-Newton method can accelerate the convergence of existing iterative methods with little extra overhead while keeping the performance of the original algorithm, thus paving the way for (near) real-time applications using on-board processing.}, doi = {10.1587/transinf.2017mvp0022}, file = {:lahouli2018accelerating - Accelerating Existing Non Blind Image Deblurring Techniques through a Strap on Limited Memory Switched Broyden Method.PDF:PDF}, publisher = {The Institute of Electronics, Information and Communication Engineers}, project = {SafeShore}, url = {https://www.jstage.jst.go.jp/article/transinf/E101.D/5/E101.D_2017MVP0022/_pdf/-char/en}, unit= {meca-ras} } - I. Lahouli, E. Karakasis, R. Haelterman, Z. Chtourou, G. De Cubber, A. Gasteratos, and R. Attia, “Hot spot method for pedestrian detection using saliency maps, discrete Chebyshev moments and support vector machine," IET Image Processing, vol. 12, iss. 7, p. 1284–1291, 2018.
[BibTeX] [Abstract] [Download PDF] [DOI]
The increasing risks of border intrusions or attacks on sensitive facilities and the growing availability of surveillance cameras lead to extensive research efforts for robust detection of pedestrians using images. However, the surveillance of borders or sensitive facilities poses many challenges including the need to set up many cameras to cover the whole area of interest, the high bandwidth requirements for data streaming and the high-processing requirements. Driven by day and night capabilities of the thermal sensors and the distinguished thermal signature of humans, the authors propose a novel and robust method for the detection of pedestrians using thermal images. The method is composed of three steps: a detection which is based on a saliency map in conjunction with a contrast-enhancement technique, a shape description based on discrete Chebyshev moments and a classification step using a support vector machine classifier. The performance of the method is tested using two different thermal datasets and is compared with the conventional maximally stable extremal regions detector. The obtained results prove the robustness and the superiority of the proposed framework in terms of true and false positives rates and computational costs which make it suitable for low-performance processing platforms and real-time applications.
@Article{lahouli2018hot, author = {Lahouli, Ichraf and Karakasis, Evangelos and Haelterman, Robby and Chtourou, Zied and De Cubber, Geert and Gasteratos, Antonios and Attia, Rabah}, journal = {IET Image Processing}, title = {Hot spot method for pedestrian detection using saliency maps, discrete {Chebyshev} moments and support vector machine}, year = {2018}, number = {7}, pages = {1284--1291}, volume = {12}, abstract = {The increasing risks of border intrusions or attacks on sensitive facilities and the growing availability of surveillance cameras lead to extensive research efforts for robust detection of pedestrians using images. However, the surveillance of borders or sensitive facilities poses many challenges including the need to set up many cameras to cover the whole area of interest, the high bandwidth requirements for data streaming and the high-processing requirements. Driven by day and night capabilities of the thermal sensors and the distinguished thermal signature of humans, the authors propose a novel and robust method for the detection of pedestrians using thermal images. The method is composed of three steps: a detection which is based on a saliency map in conjunction with a contrast-enhancement technique, a shape description based on discrete Chebyshev moments and a classification step using a support vector machine classifier. The performance of the method is tested using two different thermal datasets and is compared with the conventional maximally stable extremal regions detector. The obtained results prove the robustness and the superiority of the proposed framework in terms of true and false positives rates and computational costs which make it suitable for low-performance processing platforms and real-time applications.}, doi = {10.1049/iet-ipr.2017.0221}, publisher = {IET Digital Library}, project = {SafeShore}, url = {https://ieeexplore.ieee.org/document/8387035}, unit= {meca-ras} } - I. Lahouli, R. Haelterman, G. De Cubber, Z. Chtourou, and R. Attia, “A fast and robust approach for human detection in thermal imagery for surveillance using UAVs," in 15th Multi-Conference on Systems, Signals and Devices, Hammamet, Tunisia, 2018.
[BibTeX] [Abstract] [Download PDF] [DOI]
The use of Unmanned Aerial Vehicles (UAV)s has spread in various fields such as surveillance and search and rescue purposes. This leads to many research efforts that are focusing on the detection of people using aerial images. However, these platforms have limited resources of power and bandwidth which cause many restrictions and challenges. The use of the thermal sensors offers the possibility to work day and night and the detection of the human bodies because of its distinguished thermal signature. In this paper, we propose a fast and efficient method for the detection of humans in outdoor scenes using thermal images taken from aerial platforms. We start by extracting the bright blobs based on a conjunction between a saliency map and a contrast enhancement techniques. Then, we use the Discrete Chebyshev Moments as a shape descriptor and finally, we classify the blobs into humans and non-humans. The proposed framework is first tested using a well-known thermal database that covers a wide range of lighting and weather conditions and further and then compared to an also well-known blob extractor which is the Maximally Stable Extremal Regions detector (MSER). The results highlight the effectiveness and even the superiority of the proposed method in terms of true positives, false alarms and processing time.
@InProceedings{lahouli2018fast, author = {Lahouli, Ichraf and Haelterman, Robby and De Cubber, Geert and Chtourou, Zied and Attia, Rabah}, booktitle = {15th Multi-Conference on Systems, Signals and Devices}, title = {A fast and robust approach for human detection in thermal imagery for surveillance using {UAVs}}, year = {2018}, volume = {1}, abstract = {The use of Unmanned Aerial Vehicles (UAV)s has spread in various fields such as surveillance and search and rescue purposes. This leads to many research efforts that are focusing on the detection of people using aerial images. However, these platforms have limited resources of power and bandwidth which cause many restrictions and challenges. The use of the thermal sensors offers the possibility to work day and night and the detection of the human bodies because of its distinguished thermal signature. In this paper, we propose a fast and efficient method for the detection of humans in outdoor scenes using thermal images taken from aerial platforms. We start by extracting the bright blobs based on a conjunction between a saliency map and a contrast enhancement techniques. Then, we use the Discrete Chebyshev Moments as a shape descriptor and finally, we classify the blobs into humans and non-humans. The proposed framework is first tested using a well-known thermal database that covers a wide range of lighting and weather conditions and further and then compared to an also well-known blob extractor which is the Maximally Stable Extremal Regions detector (MSER). The results highlight the effectiveness and even the superiority of the proposed method in terms of true positives, false alarms and processing time.}, doi = {10.1109/ssd.2018.8570637}, file = {:lahouli2018fast - A Fast and Robust Approach for Human Detection in Thermal Imagery for Surveillance Using UAVs.PDF:PDF}, project = {SafeShore}, address = {Hammamet, Tunisia}, url = {https://ieeexplore.ieee.org/document/8570637}, unit= {meca-ras} } - D. Doroftei and G. De Cubber, “Qualitative and quantitative validation of drone detection systems," in International Symposium on Measurement and Control in Robotics ISMCR2018, Mons, Belgium, 2018.
[BibTeX] [Abstract] [Download PDF] [DOI]
As drones are more and more entering our world, so comes the need to regulate the access to airspace for these systems. A necessary tool in order to do this is a means of detecting these drones. Numerous commercial and non-commercial 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, which 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 and an honest comparison between systems, it is therefore paramount that a stringent validation procedure is followed. Moreover, the validation methodology needs to find a compromise between the often contrasting requirements of end users (who want tests to be performed in operational conditions) and platform developers (who want tests to be performed that are statistically relevant). Therefore, we propose in this paper 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).
@InProceedings{doroftei2018qualitative, author = {Doroftei, Daniela and De Cubber, Geert}, booktitle = {International Symposium on Measurement and Control in Robotics ISMCR2018}, title = {Qualitative and quantitative validation of drone detection systems}, year = {2018}, volume = {1}, abstract = {As drones are more and more entering our world, so comes the need to regulate the access to airspace for these systems. A necessary tool in order to do this is a means of detecting these drones. Numerous commercial and non-commercial 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, which 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 and an honest comparison between systems, it is therefore paramount that a stringent validation procedure is followed. Moreover, the validation methodology needs to find a compromise between the often contrasting requirements of end users (who want tests to be performed in operational conditions) and platform developers (who want tests to be performed that are statistically relevant). Therefore, we propose in this paper 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).}, doi = {10.5281/ZENODO.1462586}, file = {:doroftei2018qualitative - Qualitative and Quantitative Validation of Drone Detection Systems.PDF:PDF}, keywords = {Unmanned Aerial Vehicles, Drones, Detection systems, Drone detection, Test and evaluation methods}, project = {SafeShore}, address = {Mons, Belgium}, url = {http://mecatron.rma.ac.be/pub/2018/Paper_Daniela.pdf}, unit= {meca-ras} }
2017
- G. De Cubber, R. Shalom, A. Coluccia, O. Borcan, R. Chamrád, T. Radulescu, E. Izquierdo, and Z. Gagov, “The SafeShore system for the detection of threat agents in a maritime border environment," in IARP Workshop on Risky Interventions and Environmental Surveillance, Les Bon Villers, Belgium, 2017.
[BibTeX] [Abstract] [Download PDF] [DOI]
This paper discusses the goals of the H2020-SafeShore project, which has as a main goal to cover existing gaps in coastal border surveillance, increasing internal security by preventing cross-border crime such as trafficking in human beings and the smuggling of drugs. It is designed to be integrated with existing systems and create a continuous detection line along the border
@InProceedings{de2017safeshore, author = {De Cubber, Geert and Shalom, Ron and Coluccia, Angelo and Borcan, Octavia and Chamr{\'a}d, Richard and Radulescu, Tudor and Izquierdo, Ebroul and Gagov, Zhelyazko}, booktitle = {IARP Workshop on Risky Interventions and Environmental Surveillance}, title = {The {SafeShore} system for the detection of threat agents in a maritime border environment}, year = {2017}, organization = {IARP}, abstract = {This paper discusses the goals of the H2020-SafeShore project, which has as a main goal to cover existing gaps in coastal border surveillance, increasing internal security by preventing cross-border crime such as trafficking in human beings and the smuggling of drugs. It is designed to be integrated with existing systems and create a continuous detection line along the border}, doi = {10.5281/zenodo.1115552}, keywords = {SafeShore, Counter UAV, Counter RPAS}, language = {en}, project = {Safeshore}, address = {Les Bon Villers, Belgium}, url = {http://mecatron.rma.ac.be/pub/2017/SafeShore Abstract RISE-2017_.pdf}, unit= {meca-ras} } - M. Buric and G. De Cubber, “Counter Remotely Piloted Aircraft Systems," MTA Review, vol. 27, iss. 1, 2017.
[BibTeX] [Abstract] [Download PDF] [DOI]
An effective Counter Remotely Aircraft System is a major objective of many researchers and industries entities. Their activity is strongly impelled by the operational requirements of the Law Enforcement Authorities and naturally follows both the course of the latest terrorist events and technological developments. The designing process of an effective Counter Remotely Aircraft System needs to benefit from a systemic approach, starting from the legal aspects, and ending with the technical ones. From a technical point of view, the system has to work according to the five “kill chain” model starting with the detection phase, going on with the classification, prioritization, tracking and neutralization of the targets and ending with the forensic phase.
@Article{buric2017counter, author = {Buric, Marian and De Cubber, Geert}, journal = {MTA Review}, title = {Counter Remotely Piloted Aircraft Systems}, year = {2017}, number = {1}, volume = {27}, abstract = {An effective Counter Remotely Aircraft System is a major objective of many researchers and industries entities. Their activity is strongly impelled by the operational requirements of the Law Enforcement Authorities and naturally follows both the course of the latest terrorist events and technological developments. The designing process of an effective Counter Remotely Aircraft System needs to benefit from a systemic approach, starting from the legal aspects, and ending with the technical ones. From a technical point of view, the system has to work according to the five “kill chain” model starting with the detection phase, going on with the classification, prioritization, tracking and neutralization of the targets and ending with the forensic phase.}, doi = {10.5281/zenodo.1115502}, keywords = {Counter Remotely Piloted Aircraft Systems, drone, drone detection tracking and neutralization, RPAS, SafeShore}, language = {en}, publisher = {Military Technical Academy Publishing House}, project = {SafeShore}, url = {http://mecatron.rma.ac.be/pub/2017/Counter Remotely Piloted Aircraft Systems.pdf}, unit= {meca-ras} } - A. Coluccia, M. Ghenescu, T. Piatrik, G. D. Cubber, A. Schumann, L. Sommer, J. Klatte, T. Schuchert, J. Beyerer, M. Farhadi, R. Amandi, C. Aker, S. Kalkan, M. Saqib, N. Sharma, S. Daud, K. Makkah, and M. Blumenstein, “Drone-vs-Bird detection challenge at IEEE AVSS2017," in 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), Lecce, Italy, 2017, p. 1–6.
[BibTeX] [Abstract] [Download PDF] [DOI]
Small drones are a rising threat due to their possible misuse for illegal activities, in particular smuggling and terrorism. The project SafeShore, funded by the European Commission under the Horizon 2020 program, has launched the drone-vs-bird detection challenge to address one of the many technical issues arising in this context. The goal is to detect a drone appearing at some point in a video where birds may be also present: the algorithm should raise an alarm and provide a position estimate only when a drone is present, while not issuing alarms on birds. This paper reports on the challenge proposal, evaluation, and results
@InProceedings{coluccia2017drone, author = {Angelo Coluccia and Marian Ghenescu and Tomas Piatrik and Geert De Cubber and Arne Schumann and Lars Sommer and Johannes Klatte and Tobias Schuchert and Juergen Beyerer and Mohammad Farhadi and Ruhallah Amandi and Cemal Aker and Sinan Kalkan and Muhammad Saqib and Nabin Sharma and Sultan Daud and Khan Makkah and Michael Blumenstein}, booktitle = {2017 14th {IEEE} International Conference on Advanced Video and Signal Based Surveillance ({AVSS})}, title = {Drone-vs-Bird detection challenge at {IEEE} {AVSS}2017}, year = {2017}, month = aug, organization = {IEEE}, pages = {1--6}, publisher = {{IEEE}}, abstract = {Small drones are a rising threat due to their possible misuse for illegal activities, in particular smuggling and terrorism. The project SafeShore, funded by the European Commission under the Horizon 2020 program, has launched the drone-vs-bird detection challenge to address one of the many technical issues arising in this context. The goal is to detect a drone appearing at some point in a video where birds may be also present: the algorithm should raise an alarm and provide a position estimate only when a drone is present, while not issuing alarms on birds. This paper reports on the challenge proposal, evaluation, and results}, doi = {10.1109/avss.2017.8078464}, project = {SafeShore}, address = {Lecce, Italy}, url = {http://mecatron.rma.ac.be/pub/2017/WOSDETCpaper (1).pdf}, unit= {meca-ras} } - I. Lahouli, R. Haelterman, Z. Chtourou, G. De Cubber, and R. Attia, “Pedestrian Tracking in the Compressed Domain Using Thermal Images," in VIIth International Workshop on Representation, analysis and recognition of shape and motion from Image data, Savoie, France, 2017.
[BibTeX] [Abstract] [Download PDF] [DOI]
The video surveillance of sensitive facilities or borders poses many challenges like the high bandwidth requirements and the high computational cost. In this paper, we propose a framework for detecting and tracking pedestrians in the compressed domain using thermal images. Firstly, the detection process uses a conjunction between saliency maps and contrast enhancement techniques followed by a global image content descriptor based on Discrete Chebychev Moments (DCM) and a linear Support Vector Machine (SVM) as a classifier. Secondly, the tracking process exploits raw H.264 compressed video streams with limited computational overhead. In addition to two, well-known, public datasets, we have generated our own dataset by carrying six different scenarios of suspicious events using a thermal camera. The obtained results show the effectiveness and the low computational requirements of the proposed framework which make it suitable for real-time applications and on-board implementation.
@InProceedings{lahouli2017pedestrian, author = {Lahouli, Ichraf and Haelterman, Robby and Chtourou, Zied and De Cubber, Geert and Attia, Rabah}, booktitle = {VIIth International Workshop on Representation, analysis and recognition of shape and motion from Image data}, title = {Pedestrian Tracking in the Compressed Domain Using Thermal Images}, year = {2017}, number = {1}, volume = {1}, abstract = {The video surveillance of sensitive facilities or borders poses many challenges like the high bandwidth requirements and the high computational cost. In this paper, we propose a framework for detecting and tracking pedestrians in the compressed domain using thermal images. Firstly, the detection process uses a conjunction between saliency maps and contrast enhancement techniques followed by a global image content descriptor based on Discrete Chebychev Moments (DCM) and a linear Support Vector Machine (SVM) as a classifier. Secondly, the tracking process exploits raw H.264 compressed video streams with limited computational overhead. In addition to two, well-known, public datasets, we have generated our own dataset by carrying six different scenarios of suspicious events using a thermal camera. The obtained results show the effectiveness and the low computational requirements of the proposed framework which make it suitable for real-time applications and on-board implementation.}, doi = {10.1007/978-3-030-19816-9_3}, project = {SafeShore}, address = {Savoie, France}, url = {http://mecatron.rma.ac.be/pub/2017/RFMI2017_LAHOULI.pdf}, unit= {meca-ras} }