Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/4366
Title: Pothole Detection using Smartphone: A Driver Assistant
Authors: Kumar, Deepak
Kamalapuri, Amar
Choudhury, Nehal
Mukherjee, Shyamapada
Keywords: Potholes
Detection
Cloud
Alert-System
Dimen­sion
Machine-Learning
Image Processing
MobileNet
Issue Date: Dec-2023
Citation: IEEE International Conference on Modelling Simulation & Intelligent Computing (MoSICom 2023), BITS Pilani, Dubai, 07-09 December 2023
Abstract: India has an extensive road network that becomes degraded over time, causing potholes and other hazardous condi­tions for drivers. Potholes can cause physical injury and vehicle damage, so their detection is crucial. While some existing systems help drivers identify potholes, they lack precise localization and are location-dependent. In this paper, a cost-effective solution that utilizes mobile phone cameras and a deep neural network (Single Shot MultiBox Detector with MobileNet) for real-time pothole detection and localization is presented. The system alerts drivers about detected potholes instantly and achieved good mean Average Precision. Our solution aims to improve road safety and minimize vehicle damage.
Description: Copyright belongs to proceeding publisher
URI: http://hdl.handle.net/2080/4366
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2023_MOSICOM_SMukherjee_Pothole.pdf2.76 MBAdobe PDFView/Open    Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.