<![CDATA[Wildfires pose a threat to ecosystems and human safety, necessitating the development of effective monitoring techniques. Detecting forest fires based on images of forest conditions could be a breakthrough. However, the model built from imbalanced data yields low accuracy. This research addresses the challenge of class imbalance in multiclass classification for forest fire detection using the EfficientNet-B1 model. This research examines the implementation of class weighting to improve model performance, with a particular focus on minority classes, specifically Fire and Smoke. A dataset of 7,331 training images was categorized into four classes. The results showed that employing the class weighting method achieved an accuracy of 90%. The training duration of 14 minutes and 45 seconds outperforms the data augmentation method in terms of time efficiency. This study contributes to the development of more effective methods for forest fire monitoring and provides insights for future research in machine learning applications in environmental contexts.]]>
Copyrights © 2025
