End-to-end networks for detection and tracking of micro unmanned aerial vehicles

Date

2018

Author

Aker, Cemal

Metadata

Show full item record

Item Usage Stats

21
views

8
downloads

Cite This

As the number of micro unmanned aerial vehicles (mUAV) increases, several threats arise. Hence, there is a need for a system that can detect and track them. In this thesis, an object detection model based on convolutional neural networks for mUAV detection, and a novel end-to-end object tracking architecture are proposed. To solve the scarce data problem for training the detection network, an algorithm for creating an extensive artificial dataset by combining background-subtracted real images is proposed. It has been shown that the created dataset is adequate for training well performing networks and that the system can detect and track various types of mUAVs in challenging environments.

Subject Keywords

Micro air vehicles., Tracking (Engineering)., Vehicles, Remotely piloted., Convolutions (Mathematics)., Neural networks (Computer science).

URI

http://etd.lib.metu.edu.tr/upload/12622688/index.pdf
https://hdl.handle.net/11511/27574

Collections

Graduate School of Natural and Applied Sciences, Thesis