NSTL国家科技图书文献中心

1. Enhancing Agricultural Internet of Things Data Accuracy: Evaluating Kalman Filter-Based Sensor Denoising Techniques NSTL国家科技图书文献中心

Armando, Egas Jose | Hanyurwimfura, Damie...... - 《Journal of Biosystems Engineering》 - 2025,50(1) - 47～58 - 共12页

摘要： real-time sensor* denoising techniques using Kalman* | farmers aiming to optimize crop yields. However, sensor | filter models on a Raspberry Pi 4.MethodsA sensor* node* | of sensor* noise in real-time agricultural IoT* | PurposeWith the rapid adoption of the Internet

关键词： Real-time state prediction | Raspberry Pi 4 | Sensor denoising

2. Denoising graph attention wavelet network: an interpretable multi-sensor information fusion diagnostic method for rotating systems NSTL国家科技图书文献中心

Gao, Lei | Liu, Zhihao ... - 《Measurement Science & Technology》 - 2025,36(1) - 1～19 - 共19页

摘要： utilizing multi-sensor* data for fault diagnosis. To solve* | these issues, a denoising* graph attention wavelet* | network (DGAWN) is proposed for multi-sensor* information* | . Considering the spatial relationship of multi-sensor | topology of each sensor* data. The graph attention*

关键词： fault diagnosis | data fusion | graph wavelet convolution | attention mechanism | adaptive denoising

3. A filling method for missing soft measurement data based on a conditional denoising diffusion model NSTL国家科技图书文献中心

Jiang, Dongnian | Zhang, Shuai - 《Journal of computational science》 - 2025,85(Feb.) - 1.1～1.12 - 共12页

摘要： and filling method based on a conditional denoising | condition for the original denoising* diffusion model to* | the prediction accuracy of the soft sensor* model* | sensor model, reducing the mean squared error by | In complex industrial processes, incomplete

关键词： Conditional denoising diffusion model | Masking mechanism | Missing data | Multi-head self-attention mechanism | Soft sensor model

4. Fusing Transformer and diffusion for high-resolution prediction of daylight illuminance and glare based on sparse ceiling-mounted input NSTL国家科技图书文献中心

Huang Y. | Zeng T.... - 《Building and environment》 - 2025,267(Jan. Pt.A) - 1.1～1.15 - 共15页

摘要： prediction. The influence of sensor* number and placement* | © 2024 Elsevier LtdAccurate prediction of | workplane daylight illuminance and eye-height glare is | crucial for lighting control. Existing studies used | machine learning to predict illuminance at predetermined

关键词： Ceiling-mounted sensor | Daylight performance prediction | Denoising diffusion probabilistic model (DDPM) | Lighting control | Transformer model

5. An EMG Denoising Method Based on Flexible Wearable Sensors NSTL国家科技图书文献中心

ErXun Yan | Ping Lu ... - 《IEEE Design & test》 - 2025,42(2) - 7～16 - 共10页

摘要： denoising methods, the flexible electromyography sensor | denoising method that uses particle swarm optimization | Editors notes: This article proposes an EMG | with wearable sensors. Compared to traditional | offers superior fit and wearability. The article

关键词： Electromyography | Circuits | Noise reduction | Electrodes | Muscles | Interference | Particle swarm optimization | Wearable sensors

6. Multi-task deep learning for quantifying methane emissions from 2-D plume imagery with Low Signal-to-Noise Ratio NSTL国家科技图书文献中心

Qianhui Xu | Xiaojing Gu ... - 《International journal of remote sensing》 - 2025,46(1/2) - 704～727 - 共24页

摘要：, sensor* noise, wind speed, and sensor sensitivity. In* | denoising auxiliary task into the quantification network | by introducing an additional denoising* branch that* | network is trained with supervision using both denoising | inference phase, the denoising* branch is removed*

关键词： Methane emission | Methane quantification | Signal-to-Noise Ratio (SNR) | Deep learning | Multi-task learning

7. DualDn: Dual-Domain Denoising via Differentiable ISP NSTL国家科技图书文献中心

Ruikang Li | Yujin Wang ... - 《Computer Vision - ECCV 2024,Part LVIII》 - European Conference on Computer Vision - 2025, - 160～177 - 共18页

摘要：Image denoising* is a critical component in a* | -domain denoising* can be amplified by the subsequent ISP* | denoising works only for specific noise distributions and | denoising. Unlike previous single-domain denoising, DualDn | consists of two denoising* networks: one in the raw domain*

关键词： Image denoising | Dual-domain denoising | Image signal processing | End-to-end optimizing

8. Near-Infrared CO Sensor Based on Small Multi-Pass Gas Cell and GCRA-BiLSTM Model in Underground Parking NSTL国家科技图书文献中心

Minghao Cai | Guangzhao Cui ... - 《Microwave and optical technology letters》 - 2025,67(4) - e70200～ - 共9页

摘要： m. An improved wavelet thresholding denoising | adaptive noise (IWTD-CEEMDAN) denoising* algorithm was* | limit of detection (LoD) of the sensor* is 11.57 ppm at* | experiments are carried out on the sensor, and the | experimental results show that the sensor* has good linearity*

关键词： CO sensor | DFB | GCRA-BiLSTM | MPGC | TDLAS | Underground parking

9. Blind Extraction-Based Multichannel Speech Enhancement in Noisy and Reverberation Environments NSTL国家科技图书文献中心

Yuan Xie | Tao Zou ... - 《IEEE Sensors Letters》 - 2025,9(3) - 1～4 - 共4页

摘要： in sensor, hearing aids, robotics, and video | speech denoising* and dereverberation. First, a new* | sounds, achieving dereverberation and denoising | Speech enhancement has important applications | conferencing. However, the speech enhancement performance is

关键词： Speech enhancement | Reverberation | Reflection | Vectors | Noise reduction | Noise measurement | Feature extraction | Speech recognition | Additive noise | Signal to noise ratio

10. Real-Time Noise Suppression Algorithm for Electron Bombarded Active Pixel Sensor via Adaptive Spatio-Temporal Side Window Filter NSTL国家科技图书文献中心

Yilun Wang | Yunsheng Qian ... - 《Tenth Symposium on Novel Optoelectronic Detection Technology and Applications,Part Three of Three Parts》 - Symposium on Novel Optoelectronic Detection Technology and Applications - 2025, - 1351156.1～1351156.10 - 共10页

摘要：The electron bombardment active pixel sensor | -light image sensor. It uses a high-voltage electric | -sensitive CMOS image sensor* (ECIS), generating and* | , making conventional denoising* algorithms unable to* | filtering denoising* algorithm. In the spatial domain, the*

摘要： real-time sensor* denoising techniques using Kalman* | farmers aiming to optimize crop yields. However, sensor | filter models on a Raspberry Pi 4.MethodsA sensor* node* | of sensor* noise in real-time agricultural IoT* | PurposeWith the rapid adoption of the Internet

关键词： Real-time state prediction | Raspberry Pi 4 | Sensor denoising

摘要： utilizing multi-sensor* data for fault diagnosis. To solve* | these issues, a denoising* graph attention wavelet* | network (DGAWN) is proposed for multi-sensor* information* | . Considering the spatial relationship of multi-sensor | topology of each sensor* data. The graph attention*

关键词： fault diagnosis | data fusion | graph wavelet convolution | attention mechanism | adaptive denoising

摘要： and filling method based on a conditional denoising | condition for the original denoising* diffusion model to* | the prediction accuracy of the soft sensor* model* | sensor model, reducing the mean squared error by | In complex industrial processes, incomplete

关键词： Conditional denoising diffusion model | Masking mechanism | Missing data | Multi-head self-attention mechanism | Soft sensor model

摘要： prediction. The influence of sensor* number and placement* | © 2024 Elsevier LtdAccurate prediction of | workplane daylight illuminance and eye-height glare is | crucial for lighting control. Existing studies used | machine learning to predict illuminance at predetermined

关键词： Ceiling-mounted sensor | Daylight performance prediction | Denoising diffusion probabilistic model (DDPM) | Lighting control | Transformer model

摘要： denoising methods, the flexible electromyography sensor | denoising method that uses particle swarm optimization | Editors notes: This article proposes an EMG | with wearable sensors. Compared to traditional | offers superior fit and wearability. The article

关键词： Electromyography | Circuits | Noise reduction | Electrodes | Muscles | Interference | Particle swarm optimization | Wearable sensors

摘要：, sensor* noise, wind speed, and sensor sensitivity. In* | denoising auxiliary task into the quantification network | by introducing an additional denoising* branch that* | network is trained with supervision using both denoising | inference phase, the denoising* branch is removed*

关键词： Methane emission | Methane quantification | Signal-to-Noise Ratio (SNR) | Deep learning | Multi-task learning

摘要：Image denoising* is a critical component in a* | -domain denoising* can be amplified by the subsequent ISP* | denoising works only for specific noise distributions and | denoising. Unlike previous single-domain denoising, DualDn | consists of two denoising* networks: one in the raw domain*

关键词： Image denoising | Dual-domain denoising | Image signal processing | End-to-end optimizing

摘要： m. An improved wavelet thresholding denoising | adaptive noise (IWTD-CEEMDAN) denoising* algorithm was* | limit of detection (LoD) of the sensor* is 11.57 ppm at* | experiments are carried out on the sensor, and the | experimental results show that the sensor* has good linearity*

关键词： CO sensor | DFB | GCRA-BiLSTM | MPGC | TDLAS | Underground parking

摘要： in sensor, hearing aids, robotics, and video | speech denoising* and dereverberation. First, a new* | sounds, achieving dereverberation and denoising | Speech enhancement has important applications | conferencing. However, the speech enhancement performance is

关键词： Speech enhancement | Reverberation | Reflection | Vectors | Noise reduction | Noise measurement | Feature extraction | Speech recognition | Additive noise | Signal to noise ratio

摘要：The electron bombardment active pixel sensor | -light image sensor. It uses a high-voltage electric | -sensitive CMOS image sensor* (ECIS), generating and* | , making conventional denoising* algorithms unable to* | filtering denoising* algorithm. In the spatial domain, the*

关键词： Low light image | EBAPS | Denoise | Noise suppression

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： real-time sensor denoising techniques using Kalman | farmers aiming to optimize crop yields. However, sensor | filter models on a Raspberry Pi 4.MethodsA sensor node | of sensor noise in real-time agricultural IoT | PurposeWith the rapid adoption of the Internet

关键词： Real-time state prediction | Raspberry Pi 4 | Sensor denoising

摘要： utilizing multi-sensor data for fault diagnosis. To solve | these issues, a denoising graph attention wavelet | network (DGAWN) is proposed for multi-sensor information | . Considering the spatial relationship of multi-sensor | topology of each sensor data. The graph attention

关键词： fault diagnosis | data fusion | graph wavelet convolution | attention mechanism | adaptive denoising

摘要： and filling method based on a conditional denoising | condition for the original denoising diffusion model to | the prediction accuracy of the soft sensor model | sensor model, reducing the mean squared error by | In complex industrial processes, incomplete

关键词： Conditional denoising diffusion model | Masking mechanism | Missing data | Multi-head self-attention mechanism | Soft sensor model

摘要： prediction. The influence of sensor number and placement | © 2024 Elsevier LtdAccurate prediction of | workplane daylight illuminance and eye-height glare is | crucial for lighting control. Existing studies used | machine learning to predict illuminance at predetermined

关键词： Ceiling-mounted sensor | Daylight performance prediction | Denoising diffusion probabilistic model (DDPM) | Lighting control | Transformer model

摘要： denoising methods, the flexible electromyography sensor | denoising method that uses particle swarm optimization | Editors notes: This article proposes an EMG | with wearable sensors. Compared to traditional | offers superior fit and wearability. The article

关键词： Electromyography | Circuits | Noise reduction | Electrodes | Muscles | Interference | Particle swarm optimization | Wearable sensors

摘要：, sensor noise, wind speed, and sensor sensitivity. In | denoising auxiliary task into the quantification network | by introducing an additional denoising branch that | network is trained with supervision using both denoising | inference phase, the denoising branch is removed

关键词： Methane emission | Methane quantification | Signal-to-Noise Ratio (SNR) | Deep learning | Multi-task learning

摘要：Image denoising is a critical component in a | -domain denoising can be amplified by the subsequent ISP | denoising works only for specific noise distributions and | denoising. Unlike previous single-domain denoising, DualDn | consists of two denoising networks: one in the raw domain

关键词： Image denoising | Dual-domain denoising | Image signal processing | End-to-end optimizing

摘要： m. An improved wavelet thresholding denoising | adaptive noise (IWTD-CEEMDAN) denoising algorithm was | limit of detection (LoD) of the sensor is 11.57 ppm at | experiments are carried out on the sensor, and the | experimental results show that the sensor has good linearity

关键词： CO sensor | DFB | GCRA-BiLSTM | MPGC | TDLAS | Underground parking

摘要： in sensor, hearing aids, robotics, and video | speech denoising and dereverberation. First, a new | sounds, achieving dereverberation and denoising | Speech enhancement has important applications | conferencing. However, the speech enhancement performance is

关键词： Speech enhancement | Reverberation | Reflection | Vectors | Noise reduction | Noise measurement | Feature extraction | Speech recognition | Additive noise | Signal to noise ratio

摘要：The electron bombardment active pixel sensor | -light image sensor. It uses a high-voltage electric | -sensitive CMOS image sensor (ECIS), generating and | , making conventional denoising algorithms unable to | filtering denoising algorithm. In the spatial domain, the

关键词： Low light image | EBAPS | Denoise | Noise suppression

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： real-time sensor* denoising techniques using Kalman* | farmers aiming to optimize crop yields. However, sensor | filter models on a Raspberry Pi 4.MethodsA sensor* node* | of sensor* noise in real-time agricultural IoT* | PurposeWith the rapid adoption of the Internet

摘要： utilizing multi-sensor* data for fault diagnosis. To solve* | these issues, a denoising* graph attention wavelet* | network (DGAWN) is proposed for multi-sensor* information* | . Considering the spatial relationship of multi-sensor | topology of each sensor* data. The graph attention*

摘要： and filling method based on a conditional denoising | condition for the original denoising* diffusion model to* | the prediction accuracy of the soft sensor* model* | sensor model, reducing the mean squared error by | In complex industrial processes, incomplete

摘要： prediction. The influence of sensor* number and placement* | © 2024 Elsevier LtdAccurate prediction of | workplane daylight illuminance and eye-height glare is | crucial for lighting control. Existing studies used | machine learning to predict illuminance at predetermined

摘要：, sensor* noise, wind speed, and sensor sensitivity. In* | denoising auxiliary task into the quantification network | by introducing an additional denoising* branch that* | network is trained with supervision using both denoising | inference phase, the denoising* branch is removed*

摘要：Image denoising* is a critical component in a* | -domain denoising* can be amplified by the subsequent ISP* | denoising works only for specific noise distributions and | denoising. Unlike previous single-domain denoising, DualDn | consists of two denoising* networks: one in the raw domain*

摘要： m. An improved wavelet thresholding denoising | adaptive noise (IWTD-CEEMDAN) denoising* algorithm was* | limit of detection (LoD) of the sensor* is 11.57 ppm at* | experiments are carried out on the sensor, and the | experimental results show that the sensor* has good linearity*

摘要： in sensor, hearing aids, robotics, and video | speech denoising* and dereverberation. First, a new* | sounds, achieving dereverberation and denoising | Speech enhancement has important applications | conferencing. However, the speech enhancement performance is

摘要：The electron bombardment active pixel sensor | -light image sensor. It uses a high-voltage electric | -sensitive CMOS image sensor* (ECIS), generating and* | , making conventional denoising* algorithms unable to* | filtering denoising* algorithm. In the spatial domain, the*