Forest fire segmentation on U-net using data augmentation and attention mechanisms

Forest fires are sudden, destructive natural disasters that are challenging to manage and rescue, causing significant harm and loss to forests, forest ecosystems, and human lives. In recent years, deep learning (DL) methods have been researched for detecting and locating forest fires due to their excellent performance in object detection and recognition. This paper addresses the challenge of recognizing small forest fire targets by augmenting the dataset, improving the structure of the U-net image segmentation network to enhance accuracy, reducing the required training dataset, and designing a U-net-based forest fire detection algorithm. This algorithm incorporates attention mechanism modules, improving the model’s feature selection capability and capturing comprehensive information in forest fire images. This allows for more precise segmentation of small forest fire targets within the images. Additionally, by proportionally blending two sample label data pairs, the model generates new sample label data through mixup, thereby enhancing its generalization capabilities.

• Publication year

  2025

• Venue

  Conference on Computer Graphics, Artificial Intelligence, and Data Processing

• Publication date

  2025-04-10

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1117/12.3061200
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Advances in the study of global forest wildfires

  2023cited by this paper
• Attention mechanisms in computer vision: A survey

  2021cited by this paper
• A survey on Image Data Augmentation for Deep Learning

  2019cited by this paper
• PyTorch: An Imperative Style, High-Performance Deep Learning Library

  2019cited by this paper
• mixup: Beyond Empirical Risk Minimization

  2017cited by this paper
• Squeeze-and-Excitation Networks

  2017cited by this paper
• TensorFlow: A system for large-scale machine learning

  2016cited by this paper
• U-Net: Convolutional Networks for Biomedical Image Segmentation

  2015cited by this paper

• No citing papers are available for this paper.