Spatial Attention Mechanism for Weakly Supervised Fire and Traffic Accident Scene Classification

University

Missouri University of Science & Technology

Principal Investigator

Zhaozheng Yin (yinz@mst.edu)

Total Project Cost

$120,000

Funding Type

2016 USDOT

Start Date

8/1/2017

End Date

06/30/2019

Agency ID or Contract Number

69A3551747107

Abstract

During past ten years, on average there were near 16.5 thousands of hazardous materials (hazmat) transport incidents per year resulting in $82 millions of damages. Prompt, accurate, objective assessment on hazmat incidents is important for the first-responders to take appropriate actions timely, which will reduce the damage of hazmat incidents and protect the safety of people and environment. This multi-phase project aims to develop a method of processing and analyzing images captured by the crowdsourcing at incident scenes, which can aid the first-responders for hazmat incident assessment, decision-making and reporting. The outcome of the project will be a component of a smart safety enhancement system that improves the responsiveness and effectiveness of emergency responders and hospitals in dealing with hazardous material transportation incidents.

Deliverables

Download the Final Report

Final Data: https://doi.org/10.32873/unl.dr.20190809.2