Pothole Detection and Area Estimation
Pothole Segmentation and Area Estimation with Deep Neural Networks and Unmanned Aerial Vehicles is the penultimate product of two years worth of work. Co-authored with Dr. Sotirios Diamantas, it was presented at ISVC 2023 and printed in Springer. This work was a direct continuation of my graduate thesis, Detecting Potholes Using Deep Neural Networks with Unmanned Aerial Vehicles
How Does It Work?
Images are fed into our YOLOv8 model one-at-a-time. Our model is trained to recognize and find the outlines of potholes.
Videos and Screenshots
Videos of the model detecting potholes can be found on the Robot Perception channel, which is managed by co-author Dr. Sotirios Diamantas. The videos are also linked below:
