Sematic segmentation of loess landslides with STAPLE mask and fully connected conditional random field

Li, Huajin; He, Yusen; Liu, Fangzhou; Deng, Jiahao; Li, Weile; Wei, Yong; Zhou, Jingren

doi:10.1007/s10346-022-01983-8

Cited by 19 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As we know, U-Net [25] and its variants have been widely used in deep learning models for image segmentation and other fields [26][27][28][29]. Our proposed network is based on the M-net [16], which originated from the field of image segmentation and has been proven to be effective in the domain of denoising.…”

Section: Reconstruction Networkmentioning

confidence: 99%

Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization

Deng,

Sun,

Zhuang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Anomaly detection has a wide range of applications and is especially important in industrial quality inspection. Currently, many top-performing anomaly detection models rely on feature embedding-based methods. However, these methods do not perform well on datasets with large variations in object locations. Reconstruction-based methods use reconstruction errors to detect anomalies without considering positional differences between samples. In this study, a reconstruction-based method using the noise-to-norm paradigm is proposed, which avoids the invariant reconstruction of anomalous regions. Our reconstruction network is based on M-net and incorporates multiscale fusion and residual attention modules to enable end-to-end anomaly detection and localization. Experiments demonstrate that the method is effective in reconstructing anomalous regions into normal patterns and achieving accurate anomaly detection and localization. On the MPDD and VisA datasets, our proposed method achieved more competitive results than the latest methods, and it set a new state-of-the-art standard on the MPDD dataset.

show abstract

Section: Reconstruction Networkmentioning

confidence: 99%

Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization

Deng,

Sun,

Zhuang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Publications related to landslide deformation prediction could be classified into deterministic methods and probabilistic methods based on the methodological perspective. In scholarly research, considerable effort has been dedicated to exploring a range of deterministic models for predicting landslide deformations (Feng et al, 2022a;Li et al, 2023). Wang H. et al (2023a) performed landslide displacement prediction using on-site data based on ARIMA model.…”

Section: Introductionmentioning

confidence: 99%

Generative probabilistic prediction of precipitation induced landslide deformation with variational autoencoder and gated recurrent unit

Cai,

Lan,

Huang

et al. 2024

Front. Earth Sci.

Self Cite

View full text Add to dashboard Cite

Landslides, prevalent in mountainous areas, are typically triggered by tectonic movements, climatic changes, and human activities. They pose catastrophic risks, especially when occurring near settlements and infrastructure. Therefore, detecting, monitoring, and predicting landslide deformations is essential for geo-risk mitigation. The mainstream of the previous studies have often focused on deterministic models for immediate landslide prediction. However, in most of them, the aspect of prediction uncertainties are not sufficiently addressed. This paper introduces an innovative probabilistic prediction method using a Variational Autoencoder (VAE) combined with Gated Recurrent Unit (GRU) to forecast landslide deformations from a generative standpoint. Our approach consists of two main elements: firstly, training the VAE-GRU model to maximize the variational lower bound on the likelihood of historical precipitation data; secondly, using the learned approximated posterior distribution to predict imminent deformations from a generative angle. To assess the prediction quality, we use four widely-used metrics: Prediction Interval Coverage Probability (PICP), Prediction Interval Normalized Average Width (PINAW), Coverage Width-Based Criterion (CWC), and Prediction Interval Normalized Root Mean Square Width (PINRW). The results demonstrate that our proposed VAE-GRU framework surpasses traditional state-of-the-art (SOTA) probabilistic deformation prediction algorithms in terms of accuracy and reliability.

show abstract

“…Toppling failure has been reported in more and more situations with the construction of mines, highways, hydropower stations, and so on. Many scholars have carried out research on the evolution mechanism of toppling failure and identification of instability failure (Huang, 2008;Huang et al, 2017;Tao et al, 2019;Bao et al, 2022;Nie et al, 2022;Zhu et al, 2022;Li et al, 2022a). The cantilever beam theory was introduced to study the rock mass with slab fracture structure (Goodman and Bray, 1976;Sun and Zhang, 1985;Aydan and Kawamoto, 1992;Amini et al, 2009Amini et al, , 2012Liu H. J. et al, 2016;Alejano et al, 2018;Zheng et al, 2018;Cai et al, 2022a).…”

Section: Introductionmentioning

confidence: 99%

Mechanical mechanism of rock mass slabbing aggravating toppling failure

Cai

Lu²,

Li³

et al. 2023

Front. Ecol. Evol.

View full text Add to dashboard Cite

Many slabbing rock masses have emerged in hydropower slopes and underground engineering, with the construction of basic engineering and resource development projects along the zone of the Belt and Road. The anti-dip slabbing rock mass is prone to toppling and the degree of slabbing controls the development of toppling deformation. There are a few reports on the mechanical mechanism of rock mass toppling deformation after slabbing. Based on the analysis of the genetic conditions of rock mass slabbing, the influence of rock mass after slabbing on toppling deformation was explored by means of the mechanics method. The toppling bending deflection (TBD) and the toppling fracture depth (TFD) were selected as the analysis indexes, and the response regularity of slabbing on toppling rock mass was analyzed with examples. The results show that the width and thickness of the slabbing rock mass become narrower and thinner, the toppling bending deflection (TBD) increases, the toppling fracture depth (TFD) decreases, and the toppling deformation and failure intensify. The TBD is independent of the width of rock mass slabbing under self-weight, and the change of TBD is slow when the slab beam slabbing number (n) of thickness is <4 and fast when the slabbing number is above 4. The first TFD decreases fast when w is <2.0 m and it tends to be stable when w is above 2.0 m. The first TFD reduces relatively fast with the decrease in the thickness (t) of the slab beam. The result of this study can provide a reference for the treatment and evaluation of slabbing rock mass toppling deformation.

show abstract

Sematic segmentation of loess landslides with STAPLE mask and fully connected conditional random field

Cited by 19 publications

References 39 publications

Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization

Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization

Generative probabilistic prediction of precipitation induced landslide deformation with variational autoencoder and gated recurrent unit

Mechanical mechanism of rock mass slabbing aggravating toppling failure

Contact Info

Product

Resources

About