Multi-modal temporal CNNs for live fuel moisture content estimation

Miller, Lynn; Zhu, Liujun; Yebra, Marta; Rüdiger, Christoph; Webb, Geoffrey I.

doi:10.1016/j.envsoft.2022.105467

Cited by 14 publications

(21 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extrinsic regression tasks are not as common as classification tasks, but several recent studies have investigated methods of estimating water content in vegetation, as measured by the variable Live Fuel Moisture Content (LFMC) [194][195][196][197]. Other regression tasks include estimating the wood volume of forests [198] by using a hybrid CNN-MLP model combining a time series of Sentinel-2 images with a single LiDAR image and crop yield [199] which uses a hybrid of CNN and LSTM.…”

Section: Satellite Earth Observationmentioning

confidence: 99%

“…Resnet & GRU Hybrid model Forest understory [193] 2022 2D-CNN & LSTM Ensemble model Road detection [190] 2020 U-Net & convLSTM Hybrid model Vegetation quality [192] 2017 LSTM; GRU Extrinsic regression tasks TempCNN-LFMC [195] 2021 1D-CNN Multi-tempCNN [196] 2022 1D-CNN Multi-modal, ensemble model LFMC estimation [194] 2020 LSTM Multi-modal LFMC estimation [197] 2022 1D-CNN & LSTM Multi-modal, hybrid, ensemble MLDL-net [199] 2020 2D-CNN & LSTM Hybrid model SSTNN [220] 2021 3D-CNN & LSTM Hybrid model MMFVE [198] 2022 2D-CNN Hybrid model More commonly, however, RNNs are combined with an attention layer to allow the model to focus on the most important time steps. The OD2RNN model [218], used separate GRU layers followed by attention layers to process Sentinel-1 and Sentinel-2 data, combining the features extracted by each source for the final fully-connected layers.…”

Section: Recurrent Neural Network (Rnns)mentioning

confidence: 99%

“…TASSEL [185], is an adaptation of TempCNN for OBIA classification, using TempCNN models to process features extracted from the objects, followed by an attention layer to weight the convolved features. TempCNN has also been adapted for extrinsic regression [195] and used for LFMC estimation [195][196][197].…”

Section: Convolutionalmentioning

confidence: 99%

See 2 more Smart Citations

Deep Learning for Time Series Classification and Extrinsic Regression: A Current Survey

Foumani¹,

Miller²,

Tan³

et al. 2023

Preprint

View full text Add to dashboard Cite

Time Series Classification and Extrinsic Regression are important and challenging machine learning tasks. Deep learning has revolutionized natural language processing and computer vision and holds great promise in other fields such as time series analysis where the relevant features must often be abstracted from the raw data but are not known a priori. This paper surveys the current state of the art in the fast-moving field of deep learning for time series classification and extrinsic regression. We review different network architectures and training methods used for these tasks and discuss the challenges and opportunities when applying deep learning to time series data. We also summarize two critical applications of time series classification and extrinsic regression, human activity recognition and satellite earth observation.

show abstract

Section: Satellite Earth Observationmentioning

confidence: 99%

Section: Recurrent Neural Network (Rnns)mentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for Time Series Classification and Extrinsic Regression: A Current Survey

Foumani¹,

Miller²,

Tan³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…39 Miller et al 2020. 40 Purtill 2021Miller et al 2022. 41 Ballard 2022 Australian Taxation Office 2021.…”

Section: Technologymentioning

confidence: 99%

Australian Politics and Policy junior edition

2023

View full text Add to dashboard Cite

show abstract

“…In recent years, the estimation of LFMC using deep learning has emerged. Deep learning provides new ideas for LFMC acquisition without prior knowledge and runs computationally in real time [35], [36].…”

mentioning

confidence: 99%

Assessing the Effects of Fuel Moisture Content on the 2018 Megafires in California

Kang

Quan

Lai

2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

In 2018, the megafire episodes on record occurred in California, causing a large number of civilian deaths and damages. As an important part of the "fire environment triangle," the fuel moisture content (FMC) of both live (LFMC) and dead (DFMC) vegetation were broadly accepted as the important drivers of wildfire ignition and spread, but their effects on the 2018 megafires in California were less explored. Here, we explored and compared the effects of LFMC and DFMC on the 2018 megafire in California, allowing for highlighting the role of different types of FMC in megafire risk assessment. The LFMC was collected from the global LFMC product. We used three indices obtained from the Canadian Forest Service Fire Weather Index Rating System as a proxy of DFMC products, including the fine fuel moisture code, duff moisture code (DMC), and drought code. We analyzed the long-term series (2001-2018) of these four indices in California to test whether these indices were indicative of the occurrence of the megafire, and which of the index was the most powerful driving the 2018 megafires. The results showed that all these indices were correlated with the fires in California. The LFMC showed the highest correlation with the fire occurrence between 2001 and 2018, whereas the DMC performed a major role in driving the 2018 megafire in California. This study presented the insights that the LFMC and DMC should be carefully considered in future operational fire risk assessments for megafire prescription, suppression, and response. Index Terms-Drought code (DC), duff moisture code (DMC), fine fuel moisture code (FFMC), fire risk, live fuel moisture content (LFMC), megafire, remote sensing. I. INTRODUCTION W ILDFIRE is a common natural disaster worldwide. However, as climate change progresses, it has led to Manuscript

show abstract

Multi-modal temporal CNNs for live fuel moisture content estimation

Cited by 14 publications

References 64 publications

Deep Learning for Time Series Classification and Extrinsic Regression: A Current Survey

Deep Learning for Time Series Classification and Extrinsic Regression: A Current Survey

Australian Politics and Policy junior edition

Assessing the Effects of Fuel Moisture Content on the 2018 Megafires in California

Contact Info

Product

Resources

About