Video-based Concrete Road Damage Assessment Using JetRacer Kit

Abstract

In the work, an analysis of defects that arose during the operation is performed based on two sections of a concrete road. Defects arising during operation are constantly exposed to moisture and mechanical loads. Visually calculating damage to concrete structures can be quite difficult. The use of computer vision makes it possible to estimate the area of cracks and their number on a certain section of the road based on number of key points using Oriented FAST and Rotated BRIEF algorithm. The work contains a more comprehensive approach to the determination of damaged sections of concrete roads using autonomous robot-machine JetRacer Kit.