A Deep Learning-Based Experiment on Forest Wildfire Detection in Machine Vision Course

Abstract

As an interdisciplinary course, Machine Vision combines AI and digital image processing methods. This paper develops a comprehensive machine vision experiment on forest wildfire detection that organically integrates digital image processing, machine learning and deep learning technologies. Although the research on wildfire detection has made great progress, many experiments are not suitable for students to operate. Also, the detection with high accuracy is still a big challenge. In this paper, we divide the task of forest wildfire detection into two modules, which are wildfire image classification and wildfire region detection. We propose a novel wildfire image classification algorithm based on Reduce-VGGnet, and a wildfire region detection algorithm based on the optimized CNN with the combination of spatial and temporal features. The experimental results show that the proposed Reduce-VGGNet model can reach 91.20% in accuracy, and the optimized CNN model with the combination of spatial and temporal features can reach 97.35% in accuracy. Our framework is a novel way to combine research and teaching. It can achieve good detection performance and can be used as a comprehensive experiment for Machine Vision course, which can provide the support for talent cultivation in machine vision area.

Development of a deep learning-based surveillance system for forest fire detection and monitoring using UAV

“…However, adapting the framework for satellite images, known for their data complexity and noise, remains challenging. Future work involves exploring alternative CNN-based methods, refining pre-processing techniques, and incorporating multi-sensor data for enhanced wildfire detection in noisy satellite images [ 16 ].…”

SWIFT: Simulated Wildfire Images for Fast Training Dataset

“…It was tested on various datasets, achieving an accuracy of 98.42% better than the InceptionResNetV2, VGG16, MobileNetV2, InceptionV3, and DenseNet201 models. Wong et al [48] proposed a novel wildland fire image classification method, namely, Reduce-VGGNet. This method is a modified VGG16 model, replacing the three fully connected layers with two fully connected layers and using the softmax method.…”