NSTL国家科技图书文献中心

1. Relationship between Training Sample Size and Rice Mapping Accuracy Using Sentinels 1 and 2 NSTL国家科技图书文献中心

Barideh, Rahman | Nasimi, Fereshteh - 《Journal of the Indian Society of Remote Sensing》 - 2025,53(3) - 923～931 - 共9页

摘要： number of required training* samples. Therefore, this* | relationship between the training* sample size and the* | classification accuracy* of rice and non-rice covers. For this* | runs were performed with varying training* sample* | distinguish rice from non-rice covers with high accuracy. So

关键词： Google earth engine | Radar satellite | Random forest | Rice mapping | Time series

2. Vanilla Feature Distillation for Improving the Accuracy-Robustness Trade-Off in Adversarial Training NSTL国家科技图书文献中心

Guodong Cao | Zhibo Wang ... - 《IEEE transactions on dependable and secure computing》 - 2025,22(1) - 664～676 - 共13页

摘要：Adversarial training* has been widely explored* | accuracy degradation. To achieve a better trade-off | between robustness and accuracy, we propose the Vanilla | Feature Distillation Adversarial Training* (VFDAT), which* | (optimized towards high accuracy) to guide adversarial

关键词： Training | Robustness | Feature extraction | Perturbation methods | Computational modeling | Predictive models | Adaptation models

3. Estimating Per-Class Statistics for Label Noise Learning NSTL国家科技图书文献中心

Wenshui Luo | Shuo Chen ... - 《IEEE Transactions on Pattern Analysis and Machine Intelligence》 - 2025,47(1) - 305～322 - 共18页

摘要： mislead the training* algorithm and result in degraded* | classification accuracy* when compared with state-of-the-art* | Real-world data may contain a considerable | amount of noisily labeled examples, which usually | classification performance on test data. Therefore, Label Noise

关键词： Noise measurement | Estimation | Noise | Training | Accuracy | Classification algorithms | Matrix converters

4. Spectral sensor-based device for real-time detection and severity estimation of groundnut bud necrosis virus in tomato NSTL国家科技图书文献中心

Karishma Kumari | Roaf Parray ... - 《Journal of Field Robotics》 - 2025,42(1) - 5～19 - 共15页

摘要： high accuracy* (95.01% training accuracy and 93.65* | % testing accuracy). The model identified the near | used to validate the accuracy* of the device's* | A machine learning-based approach was utilized | to develop a device for groundnut bud necrosis

关键词： decision tree | groundnut bud necrosis virus disease severity | machine learning | RT-PCR | spectral sensor | training accuracy

5. Evaluating multi-channel interaction design for enhancing Pose accuracy in yoga training among visually impaired individuals NSTL国家科技图书文献中心

Zhu, Xiaohan | Zhao, Xuandong ... - 《Disability and rehabilitation..Assistive technology.》 - 2025,20(3) - 621～634 - 共14页

摘要： design strategy to improve the accuracy* of yoga* | accuracy and product usability were verified using the | Assisted Accuracy* Scale and the SUS Scale* | interaction design significantly improved the accuracy* and* | improves the accuracy* of yoga for the visually impaired*

关键词： Multisensory channels | visually impaired groups | yoga assistance | design and research | action accuracy

6. EfficientNet-EA for Visual Location Recognition in Natural Scenes NSTL国家科技图书文献中心

Heng Zhang | Yanchao Chen ... - 《IEEE robotics and automation letters》 - 2025,10(1) - 596～603 - 共8页

摘要： accuracy. During the training phase, the model adeptly | recognition often experiences reduced accuracy* because of* | EfficientNet and by using MultiSimilarityLoss for training | positive samples, which in turn enhances its training | In natural scenarios, the visual location

关键词： Feature extraction | Visualization | Training | Accuracy | Image recognition | Semantics | Meteorology | Lighting | Convolutional neural networks | Computational modeling

7. Magnetotelluric Inversion Based on Double-Layer Convolutional Neural Network NSTL国家科技图书文献中心

Han Lin | Jianfeng Jin ... - 《IEEE geoscience and remote sensing letters》 - 2025,22 - 1～5 - 共5页

摘要： introduced to improve the accuracy* of 2.5-D magnetotelluric* | to generate the dataset for the training* of the* | convolutional neural network (CNN). The training* input* | accuracy of the inversion results when employing the | A double-layer neural network combining a

关键词： Conductivity | Neural networks | Training | Accuracy | Data models | Convolutional neural networks | Magnetic resonance imaging | Receivers | Electromagnetics | Resistance

8. Optimized Synthetic Data Integration With Transformer’s DGA Data for Improved ML-Based Fault Identification NSTL国家科技图书文献中心

Atul Jaysing Patil | Ram Naresh ... - 《IEEE transactions on dielectrics and electrical insulation》 - 2025,32(1) - 598～607 - 共10页 - 被引量：1

摘要： forest (RF)that maximizes the accuracy* of these data* | -driven algorithms without training* with excessive data* | instances resulting in overfitting on the training* dataset* | model training, the study introduces a novel two-step | accuracy of the KNN model increased from 79.33% to 90.26

关键词： Fault diagnosis | Power transformer insulation | Training | Accuracy | Testing | Oil insulation | Data models

9. ISQ: Intermediate-Value Slip Quantization for Accumulator-Aware Training NSTL国家科技图书文献中心

Chi Zhang | Xu Yang ... - 《IEEE signal processing letters》 - 2025,32 - 976～980 - 共5页

摘要： accuracy of the model. Therefore, there is a challenge in | balancing accuracy* and low-bit accumulators. In this* | letter, we propose a novel method applying for training | accuracy under the 16-bit accumulator can be restored to | accuracy can be achieved with the lowest hardware

关键词： Quantization (signal) | Training | Accuracy | Hardware | Neural networks | Optimization | Backpropagation | Field programmable gate arrays | Data mining | Convolution

10. A Weakly Supervised Learning Approach for Sea Ice Stage of Development Classification From AI4Arctic Sea Ice Challenge Dataset NSTL国家科技图书文献中心

Xinwei Chen | Muhammed Patel ... - 《IEEE Transactions on Geoscience and Remote Sensing》 - 2025,63 - 1～15 - 共15页

摘要：-annotated ice charts. During training, region-based loss | , for model training* and evaluation. Visual* | accuracy. This methodology marks a critical step forward | Deep learning (DL)-based fully supervised | approaches have demonstrated remarkable performance in sea

关键词： Google earth engine | Radar satellite | Random forest | Rice mapping | Time series

摘要：Adversarial training* has been widely explored* | accuracy degradation. To achieve a better trade-off | between robustness and accuracy, we propose the Vanilla | Feature Distillation Adversarial Training* (VFDAT), which* | (optimized towards high accuracy) to guide adversarial

关键词： Training | Robustness | Feature extraction | Perturbation methods | Computational modeling | Predictive models | Adaptation models

摘要： mislead the training* algorithm and result in degraded* | classification accuracy* when compared with state-of-the-art* | Real-world data may contain a considerable | amount of noisily labeled examples, which usually | classification performance on test data. Therefore, Label Noise

关键词： Noise measurement | Estimation | Noise | Training | Accuracy | Classification algorithms | Matrix converters

摘要： high accuracy* (95.01% training accuracy and 93.65* | % testing accuracy). The model identified the near | used to validate the accuracy* of the device's* | A machine learning-based approach was utilized | to develop a device for groundnut bud necrosis

关键词： decision tree | groundnut bud necrosis virus disease severity | machine learning | RT-PCR | spectral sensor | training accuracy

摘要： design strategy to improve the accuracy* of yoga* | accuracy and product usability were verified using the | Assisted Accuracy* Scale and the SUS Scale* | interaction design significantly improved the accuracy* and* | improves the accuracy* of yoga for the visually impaired*

关键词： Multisensory channels | visually impaired groups | yoga assistance | design and research | action accuracy

摘要： accuracy. During the training phase, the model adeptly | recognition often experiences reduced accuracy* because of* | EfficientNet and by using MultiSimilarityLoss for training | positive samples, which in turn enhances its training | In natural scenarios, the visual location

关键词： Feature extraction | Visualization | Training | Accuracy | Image recognition | Semantics | Meteorology | Lighting | Convolutional neural networks | Computational modeling

摘要： introduced to improve the accuracy* of 2.5-D magnetotelluric* | to generate the dataset for the training* of the* | convolutional neural network (CNN). The training* input* | accuracy of the inversion results when employing the | A double-layer neural network combining a

关键词： Conductivity | Neural networks | Training | Accuracy | Data models | Convolutional neural networks | Magnetic resonance imaging | Receivers | Electromagnetics | Resistance

关键词： Fault diagnosis | Power transformer insulation | Training | Accuracy | Testing | Oil insulation | Data models

摘要： accuracy of the model. Therefore, there is a challenge in | balancing accuracy* and low-bit accumulators. In this* | letter, we propose a novel method applying for training | accuracy under the 16-bit accumulator can be restored to | accuracy can be achieved with the lowest hardware

关键词： Quantization (signal) | Training | Accuracy | Hardware | Neural networks | Optimization | Backpropagation | Field programmable gate arrays | Data mining | Convolution

摘要：-annotated ice charts. During training, region-based loss | , for model training* and evaluation. Visual* | accuracy. This methodology marks a critical step forward | Deep learning (DL)-based fully supervised | approaches have demonstrated remarkable performance in sea

关键词： Sea ice | Radar polarimetry | Training | Accuracy | Sentinel-1 | Data models | Codes | Visualization | Scattering | Electronic mail

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： number of required training samples. Therefore, this | relationship between the training sample size and the | classification accuracy of rice and non-rice covers. For this | runs were performed with varying training sample | distinguish rice from non-rice covers with high accuracy. So

关键词： Google earth engine | Radar satellite | Random forest | Rice mapping | Time series

摘要：Adversarial training has been widely explored | accuracy degradation. To achieve a better trade-off | between robustness and accuracy, we propose the Vanilla | Feature Distillation Adversarial Training (VFDAT), which | (optimized towards high accuracy) to guide adversarial

关键词： Training | Robustness | Feature extraction | Perturbation methods | Computational modeling | Predictive models | Adaptation models

摘要： mislead the training algorithm and result in degraded | classification accuracy when compared with state-of-the-art | Real-world data may contain a considerable | amount of noisily labeled examples, which usually | classification performance on test data. Therefore, Label Noise

关键词： Noise measurement | Estimation | Noise | Training | Accuracy | Classification algorithms | Matrix converters

摘要： high accuracy (95.01% training accuracy and 93.65 | % testing accuracy). The model identified the near | used to validate the accuracy of the device's | A machine learning-based approach was utilized | to develop a device for groundnut bud necrosis

关键词： decision tree | groundnut bud necrosis virus disease severity | machine learning | RT-PCR | spectral sensor | training accuracy

摘要： design strategy to improve the accuracy of yoga | accuracy and product usability were verified using the | Assisted Accuracy Scale and the SUS Scale | interaction design significantly improved the accuracy and | improves the accuracy of yoga for the visually impaired

关键词： Multisensory channels | visually impaired groups | yoga assistance | design and research | action accuracy

摘要： accuracy. During the training phase, the model adeptly | recognition often experiences reduced accuracy because of | EfficientNet and by using MultiSimilarityLoss for training | positive samples, which in turn enhances its training | In natural scenarios, the visual location

关键词： Feature extraction | Visualization | Training | Accuracy | Image recognition | Semantics | Meteorology | Lighting | Convolutional neural networks | Computational modeling

摘要： introduced to improve the accuracy of 2.5-D magnetotelluric | to generate the dataset for the training of the | convolutional neural network (CNN). The training input | accuracy of the inversion results when employing the | A double-layer neural network combining a

关键词： Conductivity | Neural networks | Training | Accuracy | Data models | Convolutional neural networks | Magnetic resonance imaging | Receivers | Electromagnetics | Resistance

摘要： forest (RF)that maximizes the accuracy of these data | -driven algorithms without training with excessive data | instances resulting in overfitting on the training dataset | model training, the study introduces a novel two-step | accuracy of the KNN model increased from 79.33% to 90.26

关键词： Fault diagnosis | Power transformer insulation | Training | Accuracy | Testing | Oil insulation | Data models

摘要： accuracy of the model. Therefore, there is a challenge in | balancing accuracy and low-bit accumulators. In this | letter, we propose a novel method applying for training | accuracy under the 16-bit accumulator can be restored to | accuracy can be achieved with the lowest hardware

关键词： Quantization (signal) | Training | Accuracy | Hardware | Neural networks | Optimization | Backpropagation | Field programmable gate arrays | Data mining | Convolution

摘要：-annotated ice charts. During training, region-based loss | , for model training and evaluation. Visual | accuracy. This methodology marks a critical step forward | Deep learning (DL)-based fully supervised | approaches have demonstrated remarkable performance in sea

关键词： Sea ice | Radar polarimetry | Training | Accuracy | Sentinel-1 | Data models | Codes | Visualization | Scattering | Electronic mail

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Adversarial training* has been widely explored* | accuracy degradation. To achieve a better trade-off | between robustness and accuracy, we propose the Vanilla | Feature Distillation Adversarial Training* (VFDAT), which* | (optimized towards high accuracy) to guide adversarial

摘要： mislead the training* algorithm and result in degraded* | classification accuracy* when compared with state-of-the-art* | Real-world data may contain a considerable | amount of noisily labeled examples, which usually | classification performance on test data. Therefore, Label Noise

摘要： high accuracy* (95.01% training accuracy and 93.65* | % testing accuracy). The model identified the near | used to validate the accuracy* of the device's* | A machine learning-based approach was utilized | to develop a device for groundnut bud necrosis

摘要： design strategy to improve the accuracy* of yoga* | accuracy and product usability were verified using the | Assisted Accuracy* Scale and the SUS Scale* | interaction design significantly improved the accuracy* and* | improves the accuracy* of yoga for the visually impaired*

摘要： accuracy. During the training phase, the model adeptly | recognition often experiences reduced accuracy* because of* | EfficientNet and by using MultiSimilarityLoss for training | positive samples, which in turn enhances its training | In natural scenarios, the visual location

摘要： introduced to improve the accuracy* of 2.5-D magnetotelluric* | to generate the dataset for the training* of the* | convolutional neural network (CNN). The training* input* | accuracy of the inversion results when employing the | A double-layer neural network combining a

摘要： accuracy of the model. Therefore, there is a challenge in | balancing accuracy* and low-bit accumulators. In this* | letter, we propose a novel method applying for training | accuracy under the 16-bit accumulator can be restored to | accuracy can be achieved with the lowest hardware

摘要：-annotated ice charts. During training, region-based loss | , for model training* and evaluation. Visual* | accuracy. This methodology marks a critical step forward | Deep learning (DL)-based fully supervised | approaches have demonstrated remarkable performance in sea