Fire | Special Issue : Intelligent Forest Fire Prediction and Detection

Research

30 pages, 6517 KiB

Open AccessArticle

Wildfire Smoke Detection Enhanced by Image Augmentation with StyleGAN2-ADA for YOLOv8 and RT-DETR Models

by Ganghyun Park and Yangwon Lee

Fire 2024, 7(10), 369; https://doi.org/10.3390/fire7100369 - 17 Oct 2024

Viewed by 1225

Wildfires pose significant environmental and societal threats, necessitating improved early detection methods. This study investigates the effectiveness of integrating real-time object detection deep learning models (YOLOv8 and RT-DETR) with advanced data augmentation techniques, including StyleGAN2-ADA, for wildfire smoke detection. We evaluated model performance [...] Read more.

Wildfires pose significant environmental and societal threats, necessitating improved early detection methods. This study investigates the effectiveness of integrating real-time object detection deep learning models (YOLOv8 and RT-DETR) with advanced data augmentation techniques, including StyleGAN2-ADA, for wildfire smoke detection. We evaluated model performance on datasets enhanced with fundamental transformations and synthetic images, focusing on detection accuracy. YOLOv8X demonstrated superior overall performance with [email protected] of 0.962 and [email protected] of 0.900, while RT-DETR-X excelled in small object detection with a 0.983 detection rate. Data augmentation, particularly StyleGAN2-ADA, significantly enhanced model performance across various metrics. Our approach reduced average detection times to 1.52 min for YOLOv8X and 2.40 min for RT-DETR-X, outperforming previous methods. The models demonstrated robust performance under challenging conditions, like fog and camera noise, providing reassurance of their effectiveness. While false positives remain a challenge, these advancements contribute significantly to early wildfire smoke detection capabilities, potentially mitigating wildfire impacts through faster response times. This research establishes a foundation for more effective wildfire management strategies and underscores the potential of deep learning applications in environmental monitoring. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

22 pages, 30137 KiB

Open AccessArticle

Satellite Image Cloud Automatic Annotator with Uncertainty Estimation

by Yijiang Gao, Yang Shao, Rui Jiang, Xubing Yang and Li Zhang

Fire 2024, 7(7), 212; https://doi.org/10.3390/fire7070212 - 25 Jun 2024

Cited by 1 | Viewed by 1369

Abstract

In satellite imagery, clouds obstruct the ground information, directly impacting various downstream applications. Thus, cloud annotation/cloud detection serves as the initial preprocessing step in remote sensing image analysis. Recently, deep learning methods have significantly improved in the field of cloud detection, but training [...] Read more.

In satellite imagery, clouds obstruct the ground information, directly impacting various downstream applications. Thus, cloud annotation/cloud detection serves as the initial preprocessing step in remote sensing image analysis. Recently, deep learning methods have significantly improved in the field of cloud detection, but training these methods necessitates abundant annotated data, which requires experts with professional domain knowledge. Moreover, the influx of remote sensing data from new satellites has further led to an increase in the cost of cloud annotation. To address the dependence on labeled datasets and professional domain knowledge, this paper proposes an automatic cloud annotation method for satellite remote sensing images, CloudAUE. Unlike traditional approaches, CloudAUE does not rely on labeled training datasets and can be operated by users without domain expertise. To handle the irregular shapes of clouds, CloudAUE firstly employs a convex hull algorithm for selecting cloud and non-cloud regions by polygons. When selecting convex hulls, the cloud region is first selected, and points at the edges of the cloud region are sequentially selected as polygon vertices to form a polygon that includes the cloud region. Then, the same selection is performed on non-cloud regions. Subsequently, the fast KD-Tree algorithm is used for pixel classification. Finally, an uncertainty method is proposed to evaluate the quality of annotation. When the confidence value of the image exceeds a preset threshold, the annotation process terminates and achieves satisfactory results. When the value falls below the threshold, the image needs to undergo a subsequent round of annotation. Through experiments on two labeled datasets, HRC and Landsat 8, CloudAUE demonstrates comparable or superior accuracy to deep learning algorithms, and requires only one to two annotations to obtain ideal results. An unlabeled self-built Google Earth dataset is utilized to validate the effectiveness and generalizability of CloudAUE. To show the extension capabilities in various fields, CloudAUE also achieves desirable results on a forest fire dataset. Finally, some suggestions are provided to improve annotation performance and reduce the number of annotations. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

17 pages, 6150 KiB

Open AccessArticle

Deep Learning-Based Forest Fire Risk Research on Monitoring and Early Warning Algorithms

by Dongfang Shang, Fan Zhang, Diping Yuan, Le Hong, Haoze Zheng and Fenghao Yang

Fire 2024, 7(4), 151; https://doi.org/10.3390/fire7040151 - 22 Apr 2024

Cited by 2 | Viewed by 2011

Abstract

With the development of image processing technology and video analysis technology, forest fire monitoring technology based on video recognition is more and more important in the field of forest fire prevention and control. The objects currently applied to forest fire video image monitoring [...] Read more.

With the development of image processing technology and video analysis technology, forest fire monitoring technology based on video recognition is more and more important in the field of forest fire prevention and control. The objects currently applied to forest fire video image monitoring system monitoring are mainly flames and smoke. This paper proposes a forest fire risk monitoring and early warning algorithm, which integrates a deep learning model, infrared monitoring and early warning, and forest fire weather index. The algorithm first obtains the current visible image and infrared image of the same forest area, utilizing a smoke detection model based on deep learning to detect smoke in the visible image, and obtains the confidence level of the occurrence of fire in said visible image. Then, it determines whether the local temperature value of said infrared image exceeds a preset warning value, and obtains a judgment result based on the infrared image. It calculates again a current FWI based on environmental data, and determines a current fire danger level based on the current FWI. Finally, it determines whether or not to carry out a fire warning based on said fire danger level, said confidence level of the occurrence of fire in said visible image, and said judgment result based on the infrared image. The experimental results show that the accuracy of the algorithm in this paper reaches 94.12%, precision is 96.1%, recall is 93.67, and F1-score is 94.87. The algorithm in this paper can improve the accuracy of smoke identification at the early stage of forest fire danger occurrence, especially by excluding the interference caused by clouds, fog, dust, and so on, thus improving the fire danger warning accuracy. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

19 pages, 6775 KiB

Open AccessArticle

FFYOLO: A Lightweight Forest Fire Detection Model Based on YOLOv8

by Bensheng Yun, Yanan Zheng, Zhenyu Lin and Tao Li

Fire 2024, 7(3), 93; https://doi.org/10.3390/fire7030093 - 16 Mar 2024

Cited by 5 | Viewed by 3213

Abstract

Forest is an important resource for human survival, and forest fires are a serious threat to forest protection. Therefore, the early detection of fire and smoke is particularly important. Based on the manually set feature extraction method, the detection accuracy of the machine [...] Read more.

Forest is an important resource for human survival, and forest fires are a serious threat to forest protection. Therefore, the early detection of fire and smoke is particularly important. Based on the manually set feature extraction method, the detection accuracy of the machine learning forest fire detection method is limited, and it is unable to deal with complex scenes. Meanwhile, most deep learning methods are difficult to deploy due to high computational costs. To address these issues, this paper proposes a lightweight forest fire detection model based on YOLOv8 (FFYOLO). Firstly, in order to better extract the features of fire and smoke, a channel prior dilatation attention module (CPDA) is proposed. Secondly, the mixed-classification detection head (MCDH), a new detection head, is designed. Furthermore, MPDIoU is introduced to enhance the regression and classification accuracy of the model. Then, in the Neck section, a lightweight GSConv module is applied to reduce parameters while maintaining model accuracy. Finally, the knowledge distillation strategy is used during training stage to enhance the generalization ability of the model and reduce the false detection. Experimental outcomes demonstrate that, in comparison to the original model, FFYOLO realizes an mAP_0.5 of 88.8% on a custom forest fire dataset, which is 3.4% better than the original model, with 25.3% lower parameters and 9.3% higher frames per second (FPS). Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

21 pages, 4842 KiB

Open AccessArticle

Intelligent Methods for Forest Fire Detection Using Unmanned Aerial Vehicles

by Nikolay Abramov, Yulia Emelyanova, Vitaly Fralenko, Vyacheslav Khachumov, Mikhail Khachumov, Maria Shustova and Alexander Talalaev

Fire 2024, 7(3), 89; https://doi.org/10.3390/fire7030089 - 15 Mar 2024

Cited by 1 | Viewed by 2795

Abstract

This research addresses the problem of early detection of smoke and open fire on the observed territory by unmanned aerial vehicles. We solve the tasks of improving the quality of incoming video data by removing motion blur and stabilizing the video stream; detecting [...] Read more.

This research addresses the problem of early detection of smoke and open fire on the observed territory by unmanned aerial vehicles. We solve the tasks of improving the quality of incoming video data by removing motion blur and stabilizing the video stream; detecting the horizon line in the frame; and identifying fires using semantic segmentation with Euclidean–Mahalanobis distance and the modified convolutional neural network YOLO. The proposed horizon line detection algorithm allows for cutting off unnecessary information such as cloud-covered areas in the frame by calculating local contrast, which is equivalent to the pixel informativeness indicator of the image. Proposed preprocessing methods give a delay of no more than 0.03 s due to the use of a pipeline method for data processing. Experimental results show that the horizon clipping algorithm improves fire and smoke detection accuracy by approximately 11%. The best results with the neural network were achieved with YOLO 5m, which yielded an F1 score of 76.75% combined with a processing speed of 45 frames per second. The obtained results differ from existing analogs by utilizing a comprehensive approach to early fire detection, which includes image enhancement and alternative real-time video processing methods. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

15 pages, 2257 KiB

Open AccessArticle

Vegetation Classification and a Biomass Inversion Model for Wildfires in Chongli Based on Remote Sensing Data

by Feng Xu, Wenjing Chen, Rui Xie, Yihui Wu and Dongming Jiang

Fire 2024, 7(2), 58; https://doi.org/10.3390/fire7020058 - 17 Feb 2024

Cited by 2 | Viewed by 1598

Abstract

Vegetation classification, biomass assessment, and wildfire dynamics are interconnected wildfire-ecosystem components. The Chongli District, located in Zhangjiakou City, was the venue for skiing at the 2022 Winter Olympics. Its high mountains and dense forests create a unique environment. The establishment of alpine ski [...] Read more.

Vegetation classification, biomass assessment, and wildfire dynamics are interconnected wildfire-ecosystem components. The Chongli District, located in Zhangjiakou City, was the venue for skiing at the 2022 Winter Olympics. Its high mountains and dense forests create a unique environment. The establishment of alpine ski resorts highlighted the importance of comprehensive forest surveys. Understanding vegetation types and their biomass is critical to assessing the distribution of local forest resources and predicting the likelihood of forest fires. This study used satellite multispectral data from the Sentinel-2B satellite to classify the surface vegetation in the Chongli District through K-means clustering. By combining this classification with a biomass inversion model, the total biomass of the survey area can be calculated. The biomass inversion equation established based on multispectral remote sensing data and terrain information in the Chongli area have a strong correlation (shrub forest R² = 0.811, broadleaf forest R² = 0.356, coniferous forest R² = 0.223). These correlation coefficients are key indicators for our understanding of the relationship between remote sensing data and actual vegetation biomass, reflecting the performance of the biomass inversion model. Taking shrubland as an example, a correlation coefficient as high as 0.811 shows the model’s ability to accurately predict the biomass of this type of vegetation. In addition, through multiple linear regression, the optimal shrub, broadleaf, and coniferous forest biomass models were obtained, with the overall accuracy reaching 93.58%, 89.56%, and 97.53%, respectively, meeting the strict requirements for survey accuracy. This study successfully conducted vegetation classification and biomass inversion in the Chongli District using remote sensing data. The research results have important implications for the prevention and control of forest fires. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

15 pages, 9273 KiB

Open AccessArticle

Influence of Terrain Slope on Sub-Surface Fire Behavior in Boreal Forests of China

by Yanlong Shan, Bo Gao, Sainan Yin, Diankun Shao, Lili Cao, Bo Yu, Chenxi Cui and Mingyu Wang

Fire 2024, 7(2), 55; https://doi.org/10.3390/fire7020055 - 14 Feb 2024

Viewed by 1675

Abstract

In recent years, the influence of extreme weather patterns has led to an alarming increase in the frequency and severity of sub-surface forest fires in boreal forests. The Ledum palustre-Larix gmelinii forests of the Daxing’an Mountains of China have emerged as a hotspot [...] Read more.

In recent years, the influence of extreme weather patterns has led to an alarming increase in the frequency and severity of sub-surface forest fires in boreal forests. The Ledum palustre-Larix gmelinii forests of the Daxing’an Mountains of China have emerged as a hotspot for sub-surface fires, and terrain slope has been recognized as a pivotal factor shaping forest fire behavior. The present study was conducted to (1) study the effect of terrain slope on the smoldering temperature and spread rate using simulated smoldering experiments and (2) establish occurrence probability prediction model of the sub-surface fires’ smoldering with different slopes based on the random forest model. The results showed that all the temperatures with different slopes were high, and the highest temperature was 947.91 °C. The spread rates in the horizontal direction were higher than those in the vertical direction, and the difference increased as the slope increased. The influence of slope on the peak temperature was greater than that of spread rate. The peak temperature was extremely positively correlated with the slope, horizontal distance and vertical depth. The spread rate was extremely positively correlated with the slope. The spread rate in the vertical direction was strongly positively correlated with the depth, but was strongly negatively correlated with the horizontal distance; the horizontal spread rate was opposite. The prediction equations for smoldering peak temperature and spread rate were established based on slope, horizontal distance, and vertical depth, and the model had a good fit (p < 0.01). Using random forest model, we established the occurrence prediction models for different slopes based on horizontal distance, vertical depth, and combustion time. The models had a good fit (AUC > 0.9) and high prediction accuracy (accuracy > 80%). The study proved the effect of slope on the characteristics of sub-surface fire smoldering, explained the variation in peak temperature and spread rate between different slopes, and established the occurrence prediction model based on the random forest model. The selected models had a good fit, and prediction accuracy met the requirement of the sub-surface fire prediction. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

17 pages, 11471 KiB

Open AccessEditor’s ChoiceArticle

CNTCB-YOLOv7: An Effective Forest Fire Detection Model Based on ConvNeXtV2 and CBAM

by Yiqing Xu, Jiaming Li, Long Zhang, Hongying Liu and Fuquan Zhang

Fire 2024, 7(2), 54; https://doi.org/10.3390/fire7020054 - 12 Feb 2024

Cited by 6 | Viewed by 2834

Abstract

In the context of large-scale fire areas and complex forest environments, the task of identifying the subtle features and aspects of fire can pose a significant challenge for the deep learning model. As a result, to enhance the model’s ability to represent features [...] Read more.

In the context of large-scale fire areas and complex forest environments, the task of identifying the subtle features and aspects of fire can pose a significant challenge for the deep learning model. As a result, to enhance the model’s ability to represent features and its precision in detection, this study initially introduces ConvNeXtV2 and Conv2Former to the You Only Look Once version 7 (YOLOv7) algorithm, separately, and then compares the results with the original YOLOv7 algorithm through experiments. After comprehensive comparison, the proposed ConvNeXtV2-YOLOv7 based on ConvNeXtV2 exhibits a superior performance in detecting forest fires. Additionally, in order to further focus the network on the crucial information in the task of detecting forest fires and minimize irrelevant background interference, the efficient layer aggregation network (ELAN) structure in the backbone network is enhanced by adding four attention mechanisms: the normalization-based attention module (NAM), simple attention mechanism (SimAM), global attention mechanism (GAM), and convolutional block attention module (CBAM). The experimental results, which demonstrate the suitability of ELAN combined with the CBAM module for forest fire detection, lead to the proposal of a new method for forest fire detection called CNTCB-YOLOv7. The CNTCB-YOLOv7 algorithm outperforms the YOLOv7 algorithm, with an increase in accuracy of 2.39%, recall rate of 0.73%, and average precision (AP) of 1.14%. Full article

(This article belongs to the Special Issue Intelligent Forest Fire Prediction and Detection)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Intelligent Forest Fire Prediction and Detection

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI