Опубликована: Дек. 22, 2024
Язык: Английский
Computers and Electronics in Agriculture, Год журнала: 2025, Номер 232, С. 110104 - 110104
Опубликована: Фев. 20, 2025
Язык: Английский
Процитировано
1
Machines, Год журнала: 2025, Номер 13(1), С. 49 - 49
Опубликована: Янв. 13, 2025
This study explores the impact of transfer learning on enhancing deep models for detecting defects in aero-engine components. We focused metrics such as accuracy, precision, recall, and loss to compare performance VGG19 DeiT (data-efficient image transformer). RandomSearchCV was used hyperparameter optimization, we selectively froze some layers during training help better tailor our dataset. conclude that difference across all can be attributed adoption transformer-based architecture by model it does this well capturing complex patterns data. research demonstrates transformer hold promise improving accuracy efficiency defect detection within aerospace industry, which will, turn, contribute cleaner more sustainable aviation activities.
Язык: Английский
Процитировано
0
Electronics, Год журнала: 2025, Номер 14(5), С. 974 - 974
Опубликована: Фев. 28, 2025
Fault and defect detection are critical for ensuring the safety, reliability, quality of products infrastructure across various industries. As traditional manual inspection methods face limitations in efficiency accuracy, advancements artificial intelligence, particularly image segmentation, have paved way automated precise fault processes. A significant gap exists current research regarding integration comparative analysis classical modern segmentation approaches diverse application domains. This study addresses this by providing a systematic review that bridges techniques with cutting-edge deep learning methodologies. Unlike previous reviews focus solely on isolated or specific domains, paper offers holistic methodological innovations, breadth, emerging trends. Emphasis is placed models, hybrid approaches, like attention mechanisms lightweight architectures. Additionally, highlights challenges proposes future directions aimed at enhancing model scalability, robustness, adaptability. gaps field provides useful insights academia industry, making it key reference using segmentation.
Язык: Английский
Процитировано
0
The International Journal of Advanced Manufacturing Technology, Год журнала: 2025, Номер unknown
Опубликована: Янв. 14, 2025
Язык: Английский
Процитировано
0
Computer Science and Information Systems, Год журнала: 2025, Номер 22(1), С. 181 - 197
Опубликована: Янв. 1, 2025
This paper presents a glove defect classification method that integrates image enhancement techniques with lightweight model to enhance the efficiency and accuracy of in industrial manufacturing. A dataset comprising images five types gloves was collected, totaling 360 sample images, for training validation deep learning-based model. Image techniques, including super-pixels, exposure adjustment, blurring, limited contrast adaptive histogram equalization, increased diversity size, improving generalization. Based on MobileNetV2, improved by reducing number input channels through grayscale conversion optimizing loss function. Experimental results demonstrate MobileNetV2 achieved an average 97.85% both original enhanced datasets, effectively mitigated overfitting phenomena, exhibited significantly faster speed compared ResNet34 ResNet50 models.
Язык: Английский
Процитировано
0
Experimental Techniques, Год журнала: 2025, Номер unknown
Опубликована: Янв. 23, 2025
Язык: Английский
Процитировано
0
Next Materials, Год журнала: 2025, Номер 8, С. 100522 - 100522
Опубликована: Фев. 10, 2025
Язык: Английский
Процитировано
0
Solar, Год журнала: 2025, Номер 5(1), С. 6 - 6
Опубликована: Фев. 21, 2025
The reliable operation of photovoltaic (PV) systems is essential for sustainable energy production, yet their efficiency often compromised by defects such as bird droppings, cracks, and dust accumulation. Automated defect detection critical addressing these challenges in large-scale solar farms, where manual inspections are impractical. This study evaluates three YOLO object models—YOLOv5, YOLOv8, YOLOv11—on a comprehensive dataset to identify panel defects. YOLOv5 achieved the fastest inference time (7.1 ms per image) high precision (94.1%) cracked panels. YOLOv8 excelled recall rare defects, drops (79.2%), while YOLOv11 delivered highest mAP@0.5 (93.4%), demonstrating balanced performance across categories. Despite strong common like dusty panels (mAP@0.5 > 98%), drop posed due imbalances. These results highlight trade-offs between accuracy computational efficiency, providing actionable insights deploying automated enhance PV system reliability scalability.
Язык: Английский
Процитировано
0
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0
Energies, Год журнала: 2025, Номер 18(7), С. 1616 - 1616
Опубликована: Март 24, 2025
The operational efficiency of many industrial processes is greatly affected by condition monitoring, which has become more and important in the detection forecast electrical machine failures. Early identification possible problems prompt precise diagnosis reduce unscheduled downtime, lower maintenance costs, prevent catastrophic Traditional human-dependent diagnostic techniques are changing as a result advances artificial intelligence (AI), opening door to automated predictive plans. This paper provides detailed examination (AI) applications prediction device failures, with focus on such fuzzy systems, expert neural networks (ANNs), complex machine-learning algorithms. These methods use both historical present data identify predict allow timely actions. study looks at implementation challenges for AI-based including dependencies, processing demands, model interpretability, addition highlighting recent digital twins, explainable AI, IoT integration. review highlights revolutionary potential improving sustainability, efficiency, dependability especially context rotating machines, addressing existing constraints suggesting future research routes.
Язык: Английский
Процитировано
0