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

During the past ten years, on average there were nearly 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 the environment. Therefore, one of the most important steps is to automatically detect transport incidents, such as fire and traffic accidents. In this work, the authors introduce a simple and yet effective framework that integrates the convolutional feature maps of deep Convolutional Neural Network with a spatial attention mechanism for fire and traffic accident scene classification. Their spatial attention model learns to highlight the most discriminative convolutional features, which is related to the regions of interest in the input image. The authors train their network in a weakly supervised way. In other words, without the requirement of precise bounding box annotating the exact location of fire or traffic accidents in the image, their network can be learned from the only image-level label. In addition to the image-based traffic scene classification, the model is also applied on a set of collected videos for real-world applications. The proposed model, a simple end-to-end architecture, achieves promising performance on fire scene classification from images, and traffic accident scene classification from both images and videos.


  • English

Media Info

  • Media Type: Digital/other
  • Edition: Final Report
  • Features: Figures; Photos; References; Tables;
  • Pagination: 32p

Subject/Index Terms

Filing Info

  • Accession Number: 01714330
  • Record Type: Publication
  • Report/Paper Numbers: 25-1121-0005-137-1, MATC-MS&T: 137-1
  • Contract Numbers: 69A3551747107
  • Created Date: Aug 15 2019 3:01PM