Deep Neural Network-Based Data Reconstruction for Landslide Detection

Utomo, Darmawan; Hu, Liang-Cheng; Hsiung, Pao-Ann

doi:10.1109/igarss39084.2020.9323124

Cited by 5 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The initial step was to identify the best and most popular machine learning algorithm for forecasting sensor data. The discovery of LSTM networks, which are widely known and effective at processing time series data, marked the achievement of this milestone [18]. Since only time series data were employed in this thesis, LSTM was a wise choice.…”

Section: Literature Overviewmentioning

confidence: 99%

Predictive Modelling for Environmental Telemetry Data and Damage Prediction

2022

IRJMETS

View full text Add to dashboard Cite

Big data is being produced by the IoT more and more. Realizing this hinges on being able to extract relevant information from huge data. Big data may be utilized to optimize processes. This may be done using neural networks and machine learning, which also makes it possible to create intelligent apps that exploit this data. Sensory boxes were used to collect data from rooms, and many LSTM-based neural networks were employed to forecast the sensor values in the future. Calculations were made to determine the absolute mean error and standard deviation of the forecasts. A standard deviation-adjusted absolute mean figure was used to gauge how long it took to create a forecast. The performance and prediction accuracy of the LSTM models were then assessed. Smart healthcare platforms can continually monitor the health and activities of patients and the elderly, as well as the security and safety of the environment, with the use of environmental sensors and wearable technology. When combined with a person's medical history, this constant flow of data can provide tailored diagnosis and treatment and automate crucial duties like medical data archiving and assessing the efficacy of medical therapies. Recent developments in big data processing and artificial intelligence (AI) technology have made it possible to construct extremely dependable, precise, and strong infrastructures for data capture and processing. The main aim of the author is to predict the damage happening in the sensors that are used to detect the environmental data and also the damage prediction of the sensors using random forest as the technique for the prediction.

show abstract

Section: Literature Overviewmentioning

confidence: 99%

Predictive Modelling for Environmental Telemetry Data and Damage Prediction

2022

IRJMETS

View full text Add to dashboard Cite

show abstract

“…For landslide sensor data reconstruction, Utomo et al analyzed the situation of abnormal missing landslide data. For example, sensor data failure, external interference, or other environmental factors may be lost, and they predicted the missing data value through a long shortterm memory neural network (LSTM), which shows great performance, even in the case of 90% data loss [6]. In displacement prediction, Liu et al explored algorithms for the prediction of landslide displacements, and the results showed that LSTM and gated recurrent units (GRUs) perform with encouraging results [7].…”

Section: Introductionmentioning

confidence: 99%

Landslide Risk Prediction Model Using an Attention-Based Temporal Convolutional Network Connected to a Recurrent Neural Network

Zhang

Yang

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Landslide risk assessment is an important component of the landslide research field. For the problem of landslide assessment indicators, we utilize the TOPSIS-Entropy method to assess the risk situation of landslide occurrences, which is easy to obtain directly from sensor data. By using the TOPSIS-Entropy method in landslide datasets, the instability margins of landslide risk are obtained, reflecting the current instability probability of the landslide body. For the landslide prediction issue, deep neural networks are used to predict the corresponding landslide instability margins (LIMs). Attention mechanism-based (Attn) temporal convolutional networks (TCN) connected with recurrent neural network (RNN) models for landslide risk prediction are proposed, including TCN-Attn-RNN and RNN-Attn-TCN, which both use an encoder-decoder architecture. The encoder in the first model uses the temporal convolutional network (TCN), and the decoder uses a neural network with an RNN architecture, including long short-term memory (LSTM) networks, gated recurrent units (GRUs), and their derivative algorithms. In the second model, the encoder uses a neural network with an RNN architecture, and the decoder uses a TCN. Combining the TOPSIS-Entropy method with TCN-Attn-RNN and RNN-Attn-TCN, reliable prediction models of landslide risk are proposed. By building a landslide simulation platform, we obtained landslide data. Compared to their counterparts, the proposed prediction models of landslide risk instability margins have better predictive effects.

show abstract

Deep Learning in Landslide Studies: A Review

Hemalatha

2023

Progress in Landslide Research and Technology, Volume 1 Issue 2, 2022

View full text Add to dashboard Cite

This study reviews deep learning architectures and techniques used in the landslide domain. This study aims to understand the state of the art, challenges, and opportunities of applying deep learning to landslide research. Every paper discussed in this article is reviewed for the deep learning approach employed, the study area where it is implemented, additional benchmark algorithms implemented, model assessment metrics, the best model that is selected, and the limitations mentioned by the authors. This review increases visibility into (1) various deep learning methodologies as implemented in real-world landslide mapping, detection, monitoring, and prediction case studies, (2) projects constraints of applying deep learning to landslide research (3) provides recommendations and breakthroughs that must be established in certain areas of landslide studies.

show abstract

Deep Neural Network-Based Data Reconstruction for Landslide Detection

Cited by 5 publications

References 14 publications

Predictive Modelling for Environmental Telemetry Data and Damage Prediction

Predictive Modelling for Environmental Telemetry Data and Damage Prediction

Landslide Risk Prediction Model Using an Attention-Based Temporal Convolutional Network Connected to a Recurrent Neural Network

Deep Learning in Landslide Studies: A Review

Contact Info

Product

Resources

About