Multi-resolution reconstruction of longitudinal streambed footprints using embedded sparse convolutional autoencoders

•Embedded convolutional autoencoders (CAEs) for multi-resolution reconstruction of streambed footprints as heatmaps.•Three CAEs are trained for double-upsampling, enhancing spatial and data measurement resolution using transfer learning.•Convolutional block attention modules (CBAM) and adaptive trai...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Journal of hydrology (Amsterdam) Ročník 654; s. 132852
Hlavní autori: Yang, Yifan, Tang, Zihao, Shao, Dong, Xu, Zhonghou
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier B.V 01.06.2025
Predmet:
Convolutional autoencoder
data quality
geophysics
hydrology
Machine learning
Multi-resolution reconstruction
rivers
stream channels
Streambed footprint
Streambed footprint
Convolutional autoencoder
Multi-resolution reconstruction
Machine learning
ISSN:0022-1694
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:•Embedded convolutional autoencoders (CAEs) for multi-resolution reconstruction of streambed footprints as heatmaps.•Three CAEs are trained for double-upsampling, enhancing spatial and data measurement resolution using transfer learning.•Convolutional block attention modules (CBAM) and adaptive training for ensuring robust reconstruction of sparse data fields. This study introduces an embedded convolutional autoencoder (CAE) architecture designed for the multi-resolution reconstruction of longitudinal streambed footprints as sparse heatmaps. Three standalone but interrelated CAEs are trained to achieve double-upsampling, enhancing the heatmaps’ spatial resolution and data measurement resolution simultaneously. Transfer learning improves model training efficiency by incorporating a trained model into a larger model at the next level. Cascading the CAEs facilitates a direct pathway for enhancing data quality from the coarsest inputs and recovering fine-grained patterns. Systematic evaluations prove the CAEs’ reliability in working individually and collectively. Robustness analyses demonstrate the model’s ability to retain field reconstructive quality when subjected to various corrupted inputs, including bulk data loss and spiky noise interference with local measurements at different streambed sections. The model’s capacity benefitted from including attention mechanisms (convolutional block attention modules, CBAM) and the adaptive training strategy using crafted loss functions, ensuring efficient extraction and learning of sparse dense patterns and fast reconstruction of physically sound fields. The model architecture’s flexibility and scalability are highlighted, proving it suitable for more complex geophysical systems with higher dimensions. The proposed embedded CAE architecture provides a foundational tool for creating digital surrogates of river courses and similar entities, which often involve inherently sparsely distributive data in both spatial and temporal domains.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0022-1694
DOI:10.1016/j.jhydrol.2025.132852