Infrared imaging segmentation employing an explainable deep neural network DOI Creative Commons

XINFEI LIAO,

Dan Wang, Zairan Li

и другие.

TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES, Год журнала: 2023, Номер 31(6), С. 1021 - 1038

Опубликована: Окт. 7, 2023

Explainable AI (XAI) improved by a deep neural network (DNN) of residual (ResNet) and long short-term memory networks (LSTMs), termed XAIRL, is proposed for segmenting foot infrared imaging datasets. First, an sensor dataset acquired device preprocessed. The image features are then defined extracted with XAIRL being applied to segment the dataset. This paper compares discusses our results XAIRL. Evaluation indices perform various measurements segmentation including accuracy, precision, recall, F1 score, intersection over union (IoU), Dice similarity coefficient, mean union, boundary displacement error (BDE), Hausdorff distance, receiver operating characteristic (ROC). Compared from literature, shows highest overall performance, achieving accuracy 0.93, precision 0.91, recall 0.95, score 0.93. also displays IoU, ROC curve lowest BDE distance. Although U-Net performs well most metrics, Mask R-CNN slightly worse performance but still outperforms random forest support vector machine algorithms. By building high-quality dataset, learning-based algorithms can accurately analyze temperature pressure distribution. These models be used customize shoes individual wearers, improving their comfort reducing risk injuries, particularly those high blood pressure.

Язык: Английский

Deep learning for water quality DOI
Wei Zhi, Alison P. Appling, Heather E. Golden

и другие.

Nature Water, Год журнала: 2024, Номер 2(3), С. 228 - 241

Опубликована: Март 12, 2024

Язык: Английский

Процитировано

66
A coupled model to improve river water quality prediction towards addressing non-stationarity and data limitation DOI
Shengyue Chen, Jinliang Huang, Peng Wang

и другие.

Water Research, Год журнала: 2023, Номер 248, С. 120895 - 120895

Опубликована: Ноя. 20, 2023

Язык: Английский

Процитировано

27
HDTO-DeepAR: A novel hybrid approach to forecast surface water quality indicators DOI Creative Commons
Rosysmita Bikram Singh, Kanhu Charan Patra, Biswajeet Pradhan

и другие.

Journal of Environmental Management, Год журнала: 2024, Номер 352, С. 120091 - 120091

Опубликована: Янв. 15, 2024

Water is a vital resource supporting broad spectrum of ecosystems and human activities. The quality river water has declined in recent years due to the discharge hazardous materials toxins. Deep learning machine have gained significant attention for analysing time-series data. However, these methods often suffer from high complexity forecasting errors, primarily non-linear datasets hyperparameter settings. To address challenges, we developed an innovative HDTO-DeepAR approach predicting indicators. This proposed compared with standalone algorithms, including DeepAR, BiLSTM, GRU XGBoost, using performance metrics such as MAE, MSE, MAPE, NSE. NSE hybrid ranges between 0.8 0.96. Given value's proximity 1, model appears be efficient. PICP values (ranging 95% 98%) indicate that highly reliable Experimental results reveal close resemblance model's predictions actual values, providing valuable insights future trends. comparative study shows suggested surpasses all existing, well-known models.

Язык: Английский

Процитировано

12
The Use of Attention-Enhanced CNN-LSTM Models for Multi-Indicator and Time-Series Predictions of Surface Water Quality DOI
Minhao Zhang, Zhiyu Zhang, Xuan Wang

и другие.

Water Resources Management, Год журнала: 2024, Номер 38(15), С. 6103 - 6119

Опубликована: Авг. 9, 2024

Язык: Английский

Процитировано

8
An Interpretable Parallel Spatial CNN-LSTM Architecture for Fault Diagnosis in Rotating Machinery DOI
Qianyu Zhou, Jiong Tang

IEEE Internet of Things Journal, Год журнала: 2024, Номер 11(19), С. 31730 - 31744

Опубликована: Июль 4, 2024

In the evolving landscape of prognostics and health management (PHM) enhanced by Internet Things (IoT), diagnosing machinery system faults is critical for ensuring operational efficiency safety across various industries. This research introduces a novel, interpretable deep learning architecture designed to overcome key limitations in existing fault detection methods, such as high demand extensive training data lack transparency feature extraction. Our model uniquely integrates dual branches: one processing raw time-series through spatially transformed convolutional neural network another incorporating wavelet transform coefficients. dual-branch approach not only maximizes effective use limited but also significantly enhances interpretability, eliminating need engineering manual selection. The significance this lies its innovative methodology, which bridges gap between advanced techniques practical applicability industrial settings. By leveraging IoT sensors real-time processing, our exemplifies application PHM. proposed algorithm rigorously evaluated on experimental gearbox further validated publicly available bearing set, demonstrating generalizability scalability. Through comprehensive parametric investigations, we elucidate impact robustness physics-integrated parallel architecture, showcasing potential improve diagnosis accuracy diverse conditions. study advances state-of-the-art provides framework developing more efficient models applications.

Язык: Английский

Процитировано

7
Which riverine water quality parameters can be predicted by meteorologically-driven deep learning? DOI
Huang Sheng, Yueling Wang, Jun Xia

и другие.

The Science of The Total Environment, Год журнала: 2024, Номер 946, С. 174357 - 174357

Опубликована: Июнь 28, 2024

Язык: Английский

Процитировано

6
Deep learning for spatiotemporal forecasting in Earth system science: a review DOI Creative Commons
Manzhu Yu, Qunying Huang, Zhenlong Li

и другие.

International Journal of Digital Earth, Год журнала: 2024, Номер 17(1)

Опубликована: Авг. 19, 2024

Deep learning (DL) has demonstrated strong potential in addressing key challenges spatiotemporal forecasting across various Earth system science (ESS) domains. This review examines 69 studies applying DL to tasks within climate modeling and weather prediction, disaster management, air quality modeling, hydrological renewable energy forecasting, oceanography, environmental monitoring. We summarize commonly used architectures for ESS, technical innovations, the latest advancements predictive applications. While have proven capable of handling data, remain tackling complexities specific such as complex scale dependencies, model interpretability, integration physical knowledge. Recent innovations demonstrate growing efforts integrate knowledge, improve explainability, adapt domain-specific needs, quantify uncertainties. Finally, this highlights future directions, including (1) developing more interpretable hybrid models that synergize traditional approaches, (2) extending generalizability through techniques like domain adaptation transfer learning, (3) advancing methods uncertainty quantification missing data handling.

Язык: Английский

Процитировано

5
Hybrid deep learning based prediction for water quality of plain watershed DOI

K. H. Wang,

Lei Liu,

Xuechen Ben

и другие.

Environmental Research, Год журнала: 2024, Номер 262, С. 119911 - 119911

Опубликована: Сен. 2, 2024

Язык: Английский

Процитировано

4
An efficient data fusion model based on Bayesian model averaging for robust water quality prediction using deep learning strategies DOI
Meysam Alizamir,

Kayhan Moradveisi,

Kaywan Othman Ahmed

и другие.

Expert Systems with Applications, Год журнала: 2024, Номер 261, С. 125499 - 125499

Опубликована: Окт. 15, 2024

Язык: Английский

Процитировано

3
A novel predictive framework for water quality assessment based on socio-economic indicators and water leaving reflectance DOI
Hao Chen,

Ali P. Yunus

Groundwater for Sustainable Development, Год журнала: 2025, Номер unknown, С. 101405 - 101405

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0