Wildfire Smoke Detection System: Model Architecture, Training Mechanism, and Dataset

TL;DR

This paper introduces a novel smoke detection system combining a specialized transformer architecture with a new training mechanism and a large real-world wildfire dataset, significantly improving detection accuracy.

Contribution

It proposes the Cross Contrast Patch Embedding module and Separable Negative Sampling Mechanism, enhancing low-level feature extraction and training supervision in wildfire smoke detection.

Findings

Significant performance improvements over baseline models.

Introduction of the largest wildfire test dataset to date.

Enhanced detection accuracy on benchmark datasets.

Abstract

Vanilla Transformers focus on semantic relevance between mid- to high-level features and are not good at extracting smoke features as they overlook subtle changes in low-level features like color, transparency, and texture which are essential for smoke recognition. To address this, we propose the Cross Contrast Patch Embedding (CCPE) module based on the Swin Transformer. This module leverages multi-scale spatial contrast information in both vertical and horizontal directions to enhance the network's discrimination of underlying details. By combining Cross Contrast with Transformer, we exploit the advantages of Transformer in global receptive field and context modeling while compensating for its inability to capture very low-level details, resulting in a more powerful backbone network tailored for smoke recognition tasks. Additionally, we introduce the Separable Negative Sampling Mechanism (SNSM) to address supervision signal confusion during training and release the SKLFS-WildFire Test dataset, the largest real-world wildfire testset to date, for systematic evaluation. Extensive testing and evaluation on the benchmark dataset FIgLib and the SKLFS-WildFire Test dataset show significant performance improvements of the proposed method over baseline detection models. The code and data are available at github.com/WCUSTC/CCPE.