Abstract From the simplest vision architectures in insects to the extremely complex cortical hierarchy in primates, it is fascinating to observe how biology found efficient solutions to solve vision problems, which may stimulate the emergence of new ideas for the researchers in computer vision. Biological computer vision is an excellent resolution that serves as a low-cost and information-rich source complementing the sensor suite for unmanned aerial vehicle (UAV). For fully autonomous UAV, the capability of autonomous target recognition and visual navigation is of vital importance for a completing a mission in case of GPS signal lost. This chapter mainly focuses on target recognition, image matching, and autonomous visual tracking and landing by taking advantage of the bio-inspired computation, with the aim of dealing with vision-based surveillance and navigation. An artificial bee colony (ABC) optimized edge potential function (EPF) approach is presented to accomplish the target recognition task for low-altitude aircraft. Then a chaotic quantum-behaved particle swarm optimization (PSO) based on lateral inhibition is proposed for image matching. Moreover, implementation of autonomous visual tracking and landing for a type of low-cost UAV (quadrotor) is conducted.
Biological Vision-Based Surveillance and Navigation
2013-09-30
32 pages
Article/Chapter (Book)
Electronic Resource
English
Color Night Vision for Navigation and Surveillance
Online Contents | 2000
|Color Night Vision for Navigation and Surveillance
Transportation Research Record | 2000
|ArXiv | 2025
|AIAA | 2008
|