Experimental Large-Scale Jet Flames' Geometrical Features Extraction for Risk Management Using Infrared Images and Deep Learning Segmentation Methods

Abstract

Jet fires are relatively small and have the least severe effects among the diverse fire accidents that can occur in industrial plants; however, they are usually involved in a process known as the domino effect, that leads to more severe events, such as explosions or the initiation of another fire, making the analysis of such fires an important part of risk analysis. This research work explores the application of deep learning models in an alternative approach that uses the semantic segmentation of jet fires flames to extract the flame's main geometrical attributes, relevant for fire risk assessments. A comparison is made between traditional image processing methods and some state-of-the-art deep learning models. It is found that the best approach is a deep learning architecture known as UNet, along with its two improvements, Attention UNet and UNet++. The models are then used to segment a group of vertical jet flames of varying pipe outlet diameters to extract their main geometrical characteristics. Attention UNet obtained the best general performance in the approximation of both height and area of the flames, while also showing a statistically significant difference between it and UNet++. UNet obtained the best overall performance for the approximation of the lift-off distances; however, there is not enough data to prove a statistically significant difference between Attention UNet and UNet++. The only instance where UNet++ outperformed the other models, was while obtaining the lift-off distances of the jet flames with 0.01275 m pipe outlet diameter. In general, the explored models show good agreement between the experimental and predicted values for relatively large turbulent propane jet flames, released in sonic and subsonic regimes; thus, making these radiation zones segmentation models, a suitable approach for different jet flame risk management scenarios.

Computer Vision-based Characterization of Large-scale Jet Flames using a Synthetic Infrared Image Generation Approach

“…The third and final step in the pipeline of the proposed approach is the extraction of fire geometric characteristics from the obtained segmentation masks. This process is similar to the one described in Pérez-Guerrero et al [30], where the contour of the segmentation mask and a reference point to the nozzle, that serves as outlet of the flame, are used to obtain the jet flame area and the total height. An overall vi- sualization of this process can be observed in Fig.…”

“…The jet flame area and the distance between the fuel release point and the visible tip of the flame, described as the total height, are extracted from the segmentation masks following the procedure presented in Pérez-Guerrero et al [30], which is based on the segmentation mask contour, a reference point to the outlet of the flame, and a ratio of the pixels per metric of the image.…”

“…The evaluation of the extracted geometric characteristics is done in a similar way to Pérez-Guerrero et al [30], where the resulting measures are compared against the corresponding experimental data using two different error measures, the Mean Absolute Percentage Error (MAPE), and the Root Mean Square Percentage Error (RMSPE).…”

“…Convolutional Neural Networks (CNNs), for example, have shown notable performance in complex computer vision tasks, such as image recognition, object detection, and semantic segmentation [16]. A recent stride in the use of CNNs for the characterization of jet fires is explored in Pérez-Guerrero et al [30], where geometric features of the flames are extracted from segmentation masks of their radiation zones. The models explored, however, are heavily dependent on the amount of data and the quality of the labels.…”

“…Next, the Deep Learning (DL)-based jet flame segmentation and characterization system proposed by Pérez-Guerrero et al [30] is implemented on both real and artificial IR images. Two different CNNs, namely UNet and Attention UNet, are used to obtain a segmentation mask of the radiation zones within the flames.…”