STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Mechatronics and Automation

Smart Pothole Reporting Based on Deep-Learning

  1. Atul Gandhi

Abstract

In India, due to rage of rains and poor-quality construction the Indian state governments are facing the pothole and cracked road issues. The government is not able to assist faster to the people due to poor communication between people and govt. So, there exists a system which uses convolutional neural network (CNN) to detect the potholes. The peculiarity of this system is that it will process the pothole image reported by user directly to the stakeholders associated with that road. The system verifies the pothole image which has geographical location, that whether it is real or fake and if it is real then only it is further processed to the database. It is done using CNN. The model trained on number of images collected from a dataset. This system provides faster assistance to the people.

Keywords

Support