An Efficient Large Kernel Convolution Network Designed for Neural Processing Unit DOI
Jiawen Wang, Chenfei Liao, Dewei Li

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 142, С. 109887 - 109887

Опубликована: Дек. 28, 2024

Язык: Английский

Z-YOLOv8s-based approach for road object recognition in complex traffic scenarios DOI Creative Commons
Ruixin Zhao, Sai Hong Tang, Eris Elianddy Supeni

и другие.

Alexandria Engineering Journal, Год журнала: 2024, Номер 106, С. 298 - 311

Опубликована: Июль 13, 2024

Язык: Английский

Процитировано

11

A comprehensive survey of visible and infrared imaging in complex environments: Principle, degradation and enhancement DOI
Yuanbo Li, Ping Zhou, Gongbo Zhou

и другие.

Information Fusion, Год журнала: 2025, Номер unknown, С. 103036 - 103036

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

1

Key-Exchange Convolutional Auto-Encoder for Data Augmentation in Early Knee Osteoarthritis Classification DOI
Zhe Wang, Aladine Chetouani, Rachid Jennane

и другие.

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

7

Disease detection on exterior surfaces of buildings using deep learning in China DOI Creative Commons
You Chen, Dayao Li

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 12, 2025

Urban infrastructure, particularly in ageing cities, faces significant challenges maintaining building aesthetics and structural integrity. Traditional methods for detecting diseases on exteriors, such as manual inspections, are often inefficient, costly, prone to errors, leading incomplete assessments delayed maintenance actions. This study explores the application of advanced deep learning techniques accurately detect exterior surfaces buildings urban environments, aiming enhance detection efficiency accuracy while providing a real-time monitoring solution that can be widely implemented infrastructure health management. The research model improves feature extraction by integrating DenseNet blocks Swin-Transformer prediction heads, trained validated using dataset 289 high-resolution images collected from diverse environments China. Data augmentation improved model's robustness against varying conditions. proposed achieved high rate 84.42%, recall 77.83%, an F1 score 0.81, with speed 55 frames per second. These metrics demonstrate effectiveness identifying complex damage patterns, minute cracks, even within noisy significantly outperforming traditional methods. highlights potential transform strategies offering practical ultimately enhancing contributing practices timely interventions.

Язык: Английский

Процитировано

0

A multi-objective dynamic detection model in autonomous driving based on an improved YOLOv8 DOI
Chaoran Li,

Yinghui Zhu,

Min Zheng

и другие.

Alexandria Engineering Journal, Год журнала: 2025, Номер 122, С. 453 - 464

Опубликована: Март 18, 2025

Язык: Английский

Процитировано

0

Fast shallow multi-subnet detector for real-time object detection DOI
Yuan Li, Min Song, Ke Hu

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 156, С. 111090 - 111090

Опубликована: Май 24, 2025

Язык: Английский

Процитировано

0

A Shared-Road-Rights Driving Strategy Based on Resolution Guidance for Right-of-Way Conflicts DOI Open Access
Mei Li,

Guisheng Li,

Chuan Sun

и другие.

Electronics, Год журнала: 2024, Номер 13(16), С. 3214 - 3214

Опубликована: Авг. 14, 2024

In addressing the critical issue of right-of-way conflicts in mixed-traffic environments, this paper introduces a novel shared driving strategy that encompasses two guiding frameworks for resolution. The first framework applies to active lane changing. Before changing occurs, allocates right way autonomous vehicles (AVs). Based on allocated way, AVs decide whether send request relevant vehicles. To enhance lane-changing comfort, vehicle assesses variance roll and lateral acceleration exceeds preset threshold, ultimately deciding proceed with change. second pertains passive After detecting an obstacle, way. calculate based their speed distance from using information determine change lanes or decelerate order avoid obstacle. If is chosen, further evaluation necessary. improve compare pitch longitudinal acceleration, then they proposed has been validated various scenarios, including high-speed (105 km/h), low (13 general scenarios obstacles at 125 m. results show effectively functions both low-speed scenarios.

Язык: Английский

Процитировано

1

BABE: Backdoor attack with bokeh effects via latent separation suppression DOI
Junjian Li, Honglong Chen, Yudong Gao

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 138, С. 109462 - 109462

Опубликована: Окт. 22, 2024

Процитировано

0

An Efficient Large Kernel Convolution Network Designed for Neural Processing Unit DOI
Jiawen Wang, Chenfei Liao, Dewei Li

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 142, С. 109887 - 109887

Опубликована: Дек. 28, 2024

Язык: Английский

Процитировано

0