The usage of small unmanned aerial systems (sUAS) has experienced significant growth since the coming of this decade. FAA estimated there is at least 1.94 million hobbyist drone in 2017. This large number and ease of access to owning a sUAS has contributed significantly to the number of unlawful and unsafe violation of airspace. Most notoriously, criminal has been using them to access and move contrabands across borders and prison perimeters. Through work with North Carolinian Prison System, we have also identified intrusion from land, which we plan to add to our project as a integrated system of detection, offering multiple levels of protection against intrusion.
The end goal of this project is to deliver an application where a monitoring administrator can be alerted in real time where and how likely there is a drone or human trespassing. In terms of thermal camera imaging, we will also send the infrared picture to the administrator so he/she can make a decision and react swiftly to threat. We aim to have a detection rate of over 90% in the area we cover and significantly reduce the ability of criminals and other person with nefarious intent to carry about their objectives.
Drones are detected using acoustic signature emitted from the drones.
Thermal Cameras are used at specific entry points known to be accessed by trespassers. A motion detector and classifier are then applied within the local digital processor, which decide whether to send a warning to the central Server.
The code for acoustic detection of drones, including the preliminary infrastructure, can be found here
The code base for thermal imaging, movement detection and classifying logic can be found here
The code for server end logic can be found here
The code for mobile application can be found here here