Power Lines Detection from a Stereo Vision System

Technological equipment for industrial applications requires to have different types of functionalities for control and communication, it is accomplished by having several modules working together, this is also a programming challenge because these modules can come from different sources and therefore technologies. Unmanned Aerial Vehicles (UAV) are excellent platforms for validation of new applications. This research presents the development of a stereo vision system for power line detection from UAV images. This technique allows generating a three-dimension point cloud with which is possible to identify shapes and objects in order to avoid collisions, guaranteeing the safety of surrounded people and goods, and vehicle integrity itself. A simulation environment is used for pre-flight planning and testing. Implementation requires a specific set up to detect power lines and an algorithm for online data processing to finally obtain a parametric equation that is used to guide the vehicle on the fly, 3D analysis is a post-operation process.