Comparing Machine Learning based Segmentation Models on Jet Fire Radiation Zones

TL;DR

This study compares traditional and deep learning segmentation models, especially UNet, for identifying radiation zones in jet fire images, aiming to improve risk assessment and fire characterization.

Contribution

It evaluates various segmentation approaches and loss functions on jet fire data, highlighting UNet with Weighted Cross-Entropy Loss as the most effective model.

Findings

UNet with Weighted Cross-Entropy Loss achieved the best segmentation performance.

Hausdorff Distance and Adjusted Random Index correlated highly with expert rankings.

Deep learning models can effectively segment radiation zones in jet fire images.

Abstract

Risk assessment is relevant in any workplace, however there is a degree of unpredictability when dealing with flammable or hazardous materials so that detection of fire accidents by itself may not be enough. An example of this is the impingement of jet fires, where the heat fluxes of the flame could reach nearby equipment and dramatically increase the probability of a domino effect with catastrophic results. Because of this, the characterization of such fire accidents is important from a risk management point of view. One such characterization would be the segmentation of different radiation zones within the flame, so this paper presents an exploratory research regarding several traditional computer vision and Deep Learning segmentation approaches to solve this specific problem. A data set of propane jet fires is used to train and evaluate the different approaches and given the difference in the distribution of the zones and background of the images, different loss functions, that seek to alleviate data imbalance, are also explored. Additionally, different metrics are correlated to a manual ranking performed by experts to make an evaluation that closely resembles the expert's criteria. The Hausdorff Distance and Adjusted Random Index were the metrics with the highest correlation and the best results were obtained from the UNet architecture with a Weighted Cross-Entropy Loss. These results can be used in future research to extract more geometric information from the segmentation masks or could even be implemented on other types of fire accidents.