Wildfire-Detection Method Using DenseNet and CycleGAN Data Augmentation-Based Remote Camera Imagery

Park, Minsoo; Tran, Dai Quoc; Jung, Dongmin; Park, Seunghee

doi:10.3390/rs12223715

Cited by 61 publications

(38 citation statements)

References 43 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of particular cases of fire detection based on remote imaging sensing systems and applying CNNs for object detection tasks, we identified two types of recent studies, the first ones classify examples of visible fire from very close distances, such as [39][40][41][42], while the second ones try to detect both fire and smoke columns visible in larger distances [43,44].…”

Section: State Of the Artmentioning

confidence: 99%

A Deep Learning Based Object Identification System for Forest Fire Detection

Guede-Fernández

Martins

Almeida

et al. 2021

Fire

View full text Add to dashboard Cite

Forest fires are still a large concern in several countries due to the social, environmental and economic damages caused. This paper aims to show the design and validation of a proposed system for the classification of smoke columns with object detection and a deep learning-based approach. This approach is able to detect smoke columns visible below or above the horizon. During the dataset labelling, the smoke object was divided into three different classes, depending on its distance to the horizon, a cloud object was also added, along with images without annotations. A comparison between the use of RetinaNet and Faster R-CNN was also performed. Using an independent test set, an F1-score around 80%, a G-mean around 80% and a detection rate around 90% were achieved by the two best models: both were trained with the dataset labelled with three different smoke classes and with augmentation; Faster R-CNNN was the model architecture, re-trained during the same iterations but following different learning rate schedules. Finally, these models were tested in 24 smoke sequences of the public HPWREN dataset, with 6.3 min as the average time elapsed from the start of the fire compared to the first detection of a smoke column.

show abstract

Section: State Of the Artmentioning

confidence: 99%

A Deep Learning Based Object Identification System for Forest Fire Detection

Guede-Fernández

Martins

Almeida

et al. 2021

Fire

View full text Add to dashboard Cite

show abstract

“…Before the rise in popularity of deep learning methods, computer vision algorithms leveraging hand-crafted features identified that the visual (e.g., color), spatial, and temporal (i.e., motion) qualities of smoke are essential for the machine detection of wildfires [3,[10][11][12]. More recently, deep learning approaches use a combination of convolutional neural networks (CNNs) [5][6][7][13][14][15][16][17], background subtraction [13,16,18], and object detection methods [4,8,17,19,20] to incorporate visual and spatial features. Long short-term memory (LSTM) networks [4,16] or optical flow [14,18,21] methods have been applied to incorporate temporal context from video sequences.…”

Section: Related Workmentioning

confidence: 99%

“…However, the task proves challenging in real-world scenarios given the transparent and amorphous nature of smoke; faint, small, or dissipating smoke plumes; and false positives from clouds, fog, and haze. While the idea of an automated wildfire smoke detection system has been previously explored, the difficulty of acquiring a large, labeled wildfire smoke dataset has limited researchers to using small or unbalanced datasets [3,4], manually searching for images online [4][5][6][7], or synthetically generating datasets [7][8][9].…”

Section: Introductionmentioning

confidence: 99%

FIgLib & SmokeyNet: Dataset and Deep Learning Model for Real-Time Wildland Fire Smoke Detection

et al. 2022

View full text Add to dashboard Cite

The size and frequency of wildland fires in the western United States have dramatically increased in recent years. On high-fire-risk days, a small fire ignition can rapidly grow and become out of control. Early detection of fire ignitions from initial smoke can assist the response to such fires before they become difficult to manage. Past deep learning approaches for wildfire smoke detection have suffered from small or unreliable datasets that make it difficult to extrapolate performance to real-world scenarios. In this work, we present the Fire Ignition Library (FIgLib), a publicly available dataset of nearly 25,000 labeled wildfire smoke images as seen from fixed-view cameras deployed in Southern California. We also introduce SmokeyNet, a novel deep learning architecture using spatiotemporal information from camera imagery for real-time wildfire smoke detection. When trained on the FIgLib dataset, SmokeyNet outperforms comparable baselines and rivals human performance. We hope that the availability of the FIgLib dataset and the SmokeyNet architecture will inspire further research into deep learning methods for wildfire smoke detection, leading to automated notification systems that reduce the time to wildfire response.

show abstract

“…Previous studies combined image data and artificial intelligence methods to improve the accuracy of forest fire detection or to minimize the factors that cause errors. Damage detection studies often face the problem of data imbalances [27], which previously relied only on images downloaded from the Web and social media platforms [28,29]. Online image databases, such as the Corsican Fire Database, have been used for binary classification as a useful test set for comparing computer vision algorithms [30] but are still not available in MLC.…”

Section: Related Workmentioning

confidence: 99%

Multilabel Image Classification with Deep Transfer Learning for Decision Support on Wildfire Response

et al. 2021

Self Cite

View full text Add to dashboard Cite

Given the explosive growth of information technology and the development of computer vision with convolutional neural networks, wildfire field data information systems are adopting automation and intelligence. However, some limitations remain in acquiring insights from data, such as the risk of overfitting caused by insufficient datasets. Moreover, most previous studies have only focused on detecting fires or smoke, whereas detecting persons and other objects of interest is equally crucial for wildfire response strategies. Therefore, this study developed a multilabel classification (MLC) model, which applies transfer learning and data augmentation and outputs multiple pieces of information on the same object or image. VGG-16, ResNet-50, and DenseNet-121 were used as pretrained models for transfer learning. The models were trained using the dataset constructed in this study and were compared based on various performance metrics. Moreover, the use of control variable methods revealed that transfer learning and data augmentation can perform better when used in the proposed MLC model. The resulting visualization is a heatmap processed from gradient-weighted class activation mapping that shows the reliability of predictions and the position of each class. The MLC model can address the limitations of existing forest fire identification algorithms, which mostly focuses on binary classification. This study can guide future research on implementing deep learning-based field image analysis and decision support systems in wildfire response work.

show abstract

Wildfire-Detection Method Using DenseNet and CycleGAN Data Augmentation-Based Remote Camera Imagery

Cited by 61 publications

References 43 publications

A Deep Learning Based Object Identification System for Forest Fire Detection

A Deep Learning Based Object Identification System for Forest Fire Detection

FIgLib & SmokeyNet: Dataset and Deep Learning Model for Real-Time Wildland Fire Smoke Detection

Multilabel Image Classification with Deep Transfer Learning for Decision Support on Wildfire Response

Contact Info

Product

Resources

About