in conjunction with the “International workshop on small-drone surveillance, detection and counteraction techniques” of IEEE AVSS 2017, August 29th, Lecce, Italy.
Motivation and description
Small drones are a rising threat due to their possible misuse for illegal activities such as smuggling of drugs as well as for terrorism attacks using explosives or chemical weapons. Several surveillance and detection technologies are under investigation at the moment, with different trade-offs in complexity, range, and capabilities. The “International workshop on small-drone surveillance, detection and counteraction techniques”, part of the 14th edition of IEEE AVSS, will bring together researchers from both academia and industry, to share recent advances in this field. In conjunction with this event, there will take place a Drone-vs-bird detection challenge. Indeed, given their characteristics, drones can be easily confused with birds, which makes the surveillance tasks even more challenging especially in maritime areas where bird populations may be massive. The use of video analytics can solve the issue, but effective algorithms are needed able to operate also under unfavorable conditions, namely weak contrast, long range, reduced visibility, etc.
The challenge aims at attracting research efforts to identify novel solutions to the problem outlined above, i.e., discrimination between birds and drones, by providing a video dataset that may be difficult to obtain (drone flying require special conditions and permissions, and shore areas are needed for the considered problem). The challenge goal is to detect a drone appearing at some time in a short video sequence where birds are 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.
Participation and joint workshop
The challenge is organized in conjunction with the International workshop on small-drone surveillance, detection and counteraction techniques (WOSDETC), which will guarantee publication of the paper associated with the best proposals. All the participants to the challenge must submit score files with their results, as explained below, together with a companion paper (up to 5 pages) describing the applied methodology.
The winner will receive as prize a TX2 platform offered by Nvidia. The best paper will be published in the conference proceedings and may be also presented in the workshop day session; other interesting papers might also be accepted for publication.
Moreover, a paper resuming the overall detection/tracking results will be published in the main conference including the name of all participant to the challenge.
Participation to the workshop is of course possible independently of the challenge, following the standard submission and peer-review process.
Challenge organization and evaluation
For this challenge the following dataset is made available, upon request and after signing a data user agreement: a collection of 5 MPEG4-coded videos where a drone enters the scene at some point. Annotation is provided in separate files in terms of frame number and bounding box of the target ([x y width height]) only when the drone is present.
Submission and evaluation procedure
A few days before the challenge deadline, a few video sequences will be provided for testing.
Authors should submit by the deadline one file for each test video, in the same format as the annotation file, providing the frame number and estimated bounding box ([x y width height]) only for the frames where the algorithm detects the presence of the drone. For frames not reported in the files, no detection is assumed.
The algorithm should try to put the bounding box as close as possible to the target.
A penalty will be computed frame-by-frame as the area (in pixels) of the smallest box that includes both the true and estimated bounding boxes, normalized by the area of the target’s bounding box in order to be meaningfully averaged over all frames. Two examples are:
For frames with no target a bounding box [0 0 1 1] is used, i.e., located at the origin with 1 pixel area.
The final score is obtained as the average penalty, hence the best (smallest) possible score is 1.
As mentioned, authors should also submit a companion paper presenting their methodology and results.
Authors achieving the best performance (minimum score) will receive as prize a TX2 platform offered by Nvidia, and their paper will be published in the conference proceedings. Other papers with interesting results, although not winning the challenge, will be also accepted for publication. The winner will be asked to run the algorithms at the conference venue.
Result and paper submission
The result must be submitted through the CMT web site activated for the workshop. See the submission page.
Submission deadline: June 7th, 2017
Author notification: June 15th, 2017
Angelo Coluccia, University of Salento, Lecce, Italy
Geert De Cubbert, Royal Military Academy, Belgium
Tomas Piatrik, Queen Mary University, London, UK
Marian Ghenescu, UTI Grup, Romania
Alessio Fascista, University of Salento, Lecce, Italy