Unveiling air pollution patterns in Yemen: a spatial–temporal functional data analysis

Environmental Science and Pollution Research, Journal Year: 2023, Volume and Issue: 30(17), P. 50067 - 50095

Published: Feb. 15, 2023

Language: Английский

---

Empowering human-robot interaction using sEMG sensor: Hybrid deep learning model for accurate hand gesture recognition

Results in Engineering, Journal Year: 2023, Volume and Issue: 20, P. 101639 - 101639

Published: Nov. 28, 2023

In this paper, a novel approach using Henry Gas Solubility-based Stacked Convolutional Neural Network (HGS-SCNN) for hand gesture recognition surface electromyography (sEMG) sensors is proposed. The stacked architecture of the CNN model helps to capture both low-level and high-level features, enabling effective representation learning. To begin, we generated dataset comprising 600 samples gestures. Next, applied Discrete Wavelet Transform (DWT) technique extract features from filtered sEMG signal. This step allowed us spatial frequency information, enhancing discriminative power extracted features. Extensive experiments are conducted evaluate performance proposed HGS-SCNN model. addition, obtained results compared with state-of-the-art techniques, namely AOA-SCNN, GWO-SCNN, WOA-SCNN. comparative analysis demonstrates that outperforms these existing methods, achieving an impressive accuracy 99.3%. experimental validate effectiveness our in accurately detecting combination DWT-based feature extraction offers robust reliable recognition, thereby opening new possibilities intuitive human-machine interaction applications requiring gesture-based control.

Language: Английский

---

Prediction and detection of harvesting stage in cotton fields using deep adversarial networks

Soft Computing, Journal Year: 2024, Volume and Issue: 28(2), P. 1819 - 1831

Published: Jan. 1, 2024

Language: Английский

---

Enhancing the Performance and Accuracy in Real-Time Football and Player Detection Using Upgraded YOLOv5 Architecture

International Journal of Computational Intelligence Systems, Journal Year: 2024, Volume and Issue: 17(1)

Published: June 24, 2024

Abstract The study presents a significantly improved version of the YOLOv5 real-time object detection model for football player recognition. proposed technique includes feature-tuning and hyper-parameter optimization methods that have been carefully selected to enhance both speed accuracy, resulting in superior performance architecture. Furthermore, incorporates SimSPPF module enables multi-scale feature extraction with less computational power, making it highly efficient effective solution. We GhostNet reduce complexity Slim scale layer precise bounding box prediction. Our tests, conducted recordings multiple matches, demonstrate our accurately detects players even complex scenarios occlusions dynamic illumination. suggested method outperforms original YOLOv5n terms precision, recall, mean average precision at 0.5 IoU. It is also more computationally efficient. This has potential applications live broadcasting, monitoring, sports analytics. upgraded demonstrates accuracy efficiency compared previous rely on traditional image processing techniques or two-stage detectors. makes suitable practical, real-world deployments.

Language: Английский

---

Main challenges (generation and returned energy) in a deep intelligent analysis technique for renewable energy applications.

Iraqi Journal for Computer Science and Mathematics, Journal Year: 2023, Volume and Issue: unknown, P. 34 - 47

Published: June 11, 2023

In recent years, there has been an increasing demand for Renewable Energy (RE), which refers to energy generated from natural sources such as solar and wind power. Consequently, numerous scientific studies have conducted explore various approaches controlling this type of energy. This work aims highlight the main challenges associated with generation return RE by employing intelligent data analysis techniques, specifically deep learning. These are examined different perspectives, including pre-processing, methodology techniques used in learning, evaluation measures employed. Some research area is focused on predicting highest amount that can be at a particular time location, while others aim predict largest electrical returned electricity grid optimize use surplus resources maximize their benefits. efforts crucial ensure effective continuous operation grid. However, despite efficiency high accuracy these models, they hindered complex calculations require considerable produce desired outcomes. Additionally, employed evaluate models' performance, assessing completion rate, quality results, efficiency, error feasibility investing RE, network.

Language: Английский

---