2019 7th International Conference on Robot Intelligence Technology and Applications (RiTA) 2019
DOI: 10.1109/ritapp.2019.8932740
|View full text |Cite
|
Sign up to set email alerts
|

A deep learning approach for early wildfire detection from hyperspectral satellite images

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
23
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
6
1
1

Relationship

0
8

Authors

Journals

citations
Cited by 38 publications
(23 citation statements)
references
References 16 publications
0
23
0
Order By: Relevance
“…Some of the above-mentioned systems have been using traditional detection methods since their first implementation, but the use of Machine Learning techniques has started to capture more interest in recent years, with Deep Learning techniques gaining traction in the last two years, as one of the main tools used in the automatic recognition of forest fires [2,11]. The most common models that are implemented for smoke and fire detection and classification in images are based on either CNN (Convolutional Neural Networks) [15,21], Faster R-CNN [21], Fully Convolutional Neural networks [22] or Spatiospectral Deep Neural Networks [23]. Some of the most recent studies in fire detection systems have also been changing their traditional Deep learning approaches to object-based detection systems [24,25], which has also been rising in popularity in the Industry.…”
Section: State Of the Artmentioning
confidence: 99%
“…Some of the above-mentioned systems have been using traditional detection methods since their first implementation, but the use of Machine Learning techniques has started to capture more interest in recent years, with Deep Learning techniques gaining traction in the last two years, as one of the main tools used in the automatic recognition of forest fires [2,11]. The most common models that are implemented for smoke and fire detection and classification in images are based on either CNN (Convolutional Neural Networks) [15,21], Faster R-CNN [21], Fully Convolutional Neural networks [22] or Spatiospectral Deep Neural Networks [23]. Some of the most recent studies in fire detection systems have also been changing their traditional Deep learning approaches to object-based detection systems [24,25], which has also been rising in popularity in the Industry.…”
Section: State Of the Artmentioning
confidence: 99%
“…Finally, the work proposed by Toan et al [39] is, to the best of our knowledge, the only DL model that leverages multispectral images to perform wildfire segmentation. The authors propose a DCNN that incorporates both spectral and spatial information that they obtain through the GOES-16 satellite.…”
Section: Deep Learning-based Wildfire Segmentationmentioning
confidence: 99%
“…The authors propose a DCNN that incorporates both spectral and spatial information that they obtain through the GOES-16 satellite. As spectral images have an additional dimension of spectral bands with partial dependencies between them [39], Toan et al propose a 3D version of a convolutional layer, in which each neuron in the following layer is connected only to a cube of neurons in the previous layer [39]. The model proposed by Toan et al employs three of the mentioned 3D convolutional layers.…”
Section: Deep Learning-based Wildfire Segmentationmentioning
confidence: 99%
See 1 more Smart Citation
“…By leveraging low-level representations of the input data, deep learning models are capable of making robust classification in active fire detection. In [22], a deep learning algorithm based on convolutional neural network is adopted using GOES-16 images as the input to detect wildfires. The proposed method utilize spatial and spectral information to classify the center pixel of image patches.…”
Section: Introductionmentioning
confidence: 99%