Deep Learning-Based Instance Segmentation for Indoor Fire Load Recognition has been published in the journal IEEE Access.
This article uses deep learning-based instance segmentation to automatically detect indoor fire loads. Accurate fire load (combustible objects) information is crucial for safety design and resilience assessment of buildings. Traditional fire load acquisition methods are relatively time consuming, tedious, and error-prone, failed to adapt to dynamic changed indoor scenes. Thus, this research proposes a computer vision-based method to automatically detect indoor fire loads using deep learning-based instance segmentation. First, indoor elements are classified into different categories according to their material composition. Next, an image dataset of indoor scenes with instance annotations is developed. Finally, a deep learning model, based on Mask R-CNN, is developed to detect fire loads in images. Experimental results show that our model achieves promising accuracy and proves the method has high performance characteristics. This research contributes to the body of knowledge 1) a novel method of high accuracy and efficiency for automated fire load recognition in indoor environments based on instance segmentation; 2) training techniques for a deep learning model in a relatively small dataset of indoor images; 3) an image dataset with annotations of indoor fire loads. Although instance segmentation has been applied for several years, this is a pioneering research on using it for automated indoor fire load recognition, which paves the foundation for automatic fire load estimation and resilience assessment for the built environment.
Note: The IEEE Access is an important journal in the field of engineering and technology and belongs to the SCI journal. The research results was supported in part by the National Natural Science Foundation of China.
