Vision based tracking system for unmanned aerial vehicle (UAV) with monocular camera

• Main Author: Li, Mingze
• Other Authors: Xie Lihua
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/64740
• Institution: Nanyang Technological University

Unmanned aerial vehicle (UAV) provides wide applications for military and special operations, and computer vision programming plays a significant role in the development of UAV studies. This project presents an image based system of moving object recognition and tracking in real time with monocular camera. The first part is to design a fast circle detector implemented with Hough Circle Transform algorithm and provide a general structure of computer vision design using OpenCV library. The second part presents the main program design for this topic. The system can capture the target image use camera and provide image recognition and tracking functions based on the video frames. Scale-invariant feature transform (SIFT) is an algorithm for image recognition to detect and represent local features in images. It is a new efficient algorithm and applied to the image processing part of the program to complete the image capture and recognition functions.