A STUDY ON COMBINATION OF DRONE AND DEEP LEARNING BASED IMAGE PROCESSING METHOD FOR WORK EFFICIENCY OF PENSTOK MAINTENANCE

Abstract

Hydroelectric power generates electricity with the potential energy of water flow. A penstock diameter over a range of several centimeters to several meters, one of the length over a range of several meters to several kilometers. A regular maintenance method of penstock needs a worker to approach penstock and check its surface by their eye. Maintenance workers must walk the distance from one end to the other of penstock to check the whole. This method needs a lot of time and cost. We propose an automatic maintenance method using a drone and an image processing method. This method uses the penstock picture taken by a drone to check the surface of the penstock. This picture is used as an input of a deep learning-based image processing method to extract the damaged area on the penstock. In this paper, we explain the overview of the proposed method, evaluate the damage extract method, discuss the remaining problems to realize the proposed method.