Detection and localization of potholes in thermal images using deep neural networks
- PDF / 1,569,078 Bytes
- 20 Pages / 439.37 x 666.142 pts Page_size
- 33 Downloads / 206 Views
Detection and localization of potholes in thermal images using deep neural networks Saksham Gupta 1 & Paras Sharma 1 & Dakshraj Sharma 1 & Varun Gupta 1 & Nitigya Sambyal 2 Received: 3 October 2019 / Revised: 24 May 2020 / Accepted: 29 June 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
A pothole is a depression caused on roads due to seepage of water into soil structure or weight of continuously moving traffic. This not only damages the suspension of the vehicles but is also a prime reason for road accidents worldwide. This necessitates the need to develop an efficient automatic pothole detection system which can assist concerned authorities for timely repair and maintenance of the roads. This paper proposes a novel approach of bounding box based pothole localization from thermal images using deep neural networks. The modified ResNet34-single shot multibox detector gives an average precision of 74.53% whereas modified ResNet50-RetinaNet model provides 91.15% precision. The results obtained by the proposed modified ResNet50-RetinaNet model are the state-of-the-art results for localization of potholes using thermal images. In real-world scenarios such a system can assist relevant authorities to judge the severity of road damage and take appropriate effective measures accordingly. Keywords Pothole detection and localization . Thermal imaging . Residual networks . RetinaNet . Single shot multibox detector . Convolutional neural networks . Deep learning
1 Introduction A pothole is described as a cavity on the road surface caused mainly by subsidence or wear and tear. The water enters underlying soil and weakens its structure by contraction and expansion. This roadway material further breaks or gets displaced over time due to the weight of continuously
* Varun Gupta [email protected]
1
Department of Computer Science and Engineering, Chandigarh College of Engineering and Technology (Degree Wing), Chandigarh, India
2
Department of Computer Science and Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, India
Multimedia Tools and Applications
moving traffic over it. This ultimately results in fracture also known as pothole(s) being created on the roads. Potholes damage vehicles’ shock absorber and suspension system and in the worst case can cause fatal road accidents. According to a recent report by road transport ministry of India, potholes alone have been a prime cause of more than 3500 nationwide deaths in 2017 [26]. Potholes are also linked to the spread of vector-borne diseases like malaria and dengue due to stagnant water that often collects in it. Various damages caused by potholes necessitates the need for the development of an efficient automatic pothole detection system for real-world situations. Various potholes detection methods using hardware sensors, real-time image analysis and simulations have been proposed in the literature. However, these methods fail to perform efficiently in real-world scenarios and in unfavourable weather con
Data Loading...
