FARSA: Fully Automated Roadway Safety Assessment
Authors:
- Weilian SongWSWeilian Song
- Scott WorkmanSWScott Workman
- Armin HadzicAHArmin Hadzic
- Xu ZhangXZXu Zhang
- Eric GreenEGEric Green
- Mei ChenMCMei Chen
- Reginald SouleyretteRSReginald Souleyrette
- Nathan JacobsNJNathan Jacobs
- 1 March 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. pp, 521-529
Abstract
This paper addresses the task of road safety assessment. An emerging approach for conducting such assessments in the United States is through the US Road Assessment Program (usRAP), which rates roads from highest risk (1 star) to lowest (5 stars). Obtaining these ratings requires manual, fine-grained labeling of roadway features in street-level panoramas, a slow and costly process. We propose to automate this process using a deep convolutional neural network that directly estimates the star rating from a street-level panorama, requiring milliseconds per image at test time. Our network also estimates many other road-level attributes, including curvature, roadside hazards, and the type of median. To support this, we incorporate task-specific attention layers so the network can focus on the panorama regions that are most useful for a particular task. We evaluated our approach on a large dataset of real-world images from two US states. We found that incorporating additional tasks, and using a semi-supervised training approach, significantly reduced overfitting problems, allowed us to optimize more layers of the network, and resulted in higher accuracy.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- Understanding and Mapping Natural Beauty2017
- A Unified Model for Near and Remote Sensing2017
- Deep Learning the City: Quantifying Urban Perception at a Global Scale2016
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016
- HD Maps: Fine-Grained Road Segmentation by Parsing Ground and Aerial Images2016
- Multi-band Polarization Imaging and Applications2016
- Learning deep representations for ground-to-aerial geolocalization2015
- Predicting Geo-informative Attributes in Large-Scale Image Collections Using Convolutional Neural Networks2015
- Streetscore -- Predicting the Perceived Safety of One Million Streetscapes2014
- Learning High-Level Judgments of Urban Perception2014