Segmentation of Industrial Burner Flames: A Comparative Study from Traditional Image Processing to Machine and Deep Learning
This comparative study provides guidance for researchers and practitioners in industrial processes like power generation and waste management, but it is incremental as it evaluates existing methods without introducing new ones.
The paper tackled the problem of binary segmentation of industrial burner flames by comparing traditional image processing, machine learning, and deep learning methods on a public benchmark dataset, finding that deep learning achieved the highest accuracy while traditional methods offered fast and simple solutions.
In many industrial processes, such as power generation, chemical production, and waste management, accurately monitoring industrial burner flame characteristics is crucial for safe and efficient operation. A key step involves separating the flames from the background through binary segmentation. Decades of machine vision research have produced a wide range of possible solutions, from traditional image processing to traditional machine learning and modern deep learning methods. In this work, we present a comparative study of multiple segmentation approaches, namely Global Thresholding, Region Growing, Support Vector Machines, Random Forest, Multilayer Perceptron, U-Net, and DeepLabV3+, that are evaluated on a public benchmark dataset of industrial burner flames. We provide helpful insights and guidance for researchers and practitioners aiming to select an appropriate approach for the binary segmentation of industrial burner flames and beyond. For the highest accuracy, deep learning is the leading approach, while for fast and simple solutions, traditional image processing techniques remain a viable option.