Search

Your search keyword '"Zhong, Fei"' showing total 2,219 results
Start Over Author "Zhong, Fei"
2,219 results on '"Zhong, Fei"'

4. Fast vehicle detection algorithm based on lightweight YOLO7-tiny

Author

Li, Bo, Chen, YiHua, Xu, Hao, and Zhong, Fei

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The swift and precise detection of vehicles plays a significant role in intelligent transportation systems. Current vehicle detection algorithms encounter challenges of high computational complexity, low detection rate, and limited feasibility on mobile devices. To address these issues, this paper proposes a lightweight vehicle detection algorithm based on YOLOv7-tiny (You Only Look Once version seven) called Ghost-YOLOv7. The width of model is scaled to 0.5 and the standard convolution of the backbone network is replaced with Ghost convolution to achieve a lighter network and improve the detection speed; then a self-designed Ghost bi-directional feature pyramid network (Ghost-BiFPN) is embedded into the neck network to enhance feature extraction capability of the algorithm and enriches semantic information; and a Ghost Decouoled Head (GDH) is employed for accurate prediction of vehicle location and species; finally, a coordinate attention mechanism is introduced into the output layer to suppress environmental interference. The WIoU loss function is employed to further enhance the detection accuracy. Ablation experiments results on the PASCAL VOC dataset demonstrate that Ghost-YOLOv7 outperforms the original YOLOv7-tiny model. It achieving a 29.8% reduction in computation, 37.3% reduction in the number of parameters, 35.1% reduction in model weights, 1.1% higher mean average precision (mAP), the detection speed is higher 27FPS compared with the original algorithm. Ghost-YOLOv7 was also compared on KITTI and BIT-vehicle datasets as well, and the results show that this algorithm has the overall best performance.

Published
2023

9. Light-YOLOv5: A Lightweight Algorithm for Improved YOLOv5 in Complex Fire Scenarios

Author

Xu, Hao, Li, Bo, and Zhong, Fei

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Fire-detection technology is of great importance for successful fire-prevention measures. Image-based fire detection is one effective method. At present, object-detection algorithms are deficient in performing detection speed and accuracy tasks when they are applied in complex fire scenarios. In this study, a lightweight fire-detection algorithm, Light-YOLOv5 (You Only Look Once version five), is presented. First, a separable vision transformer (SepViT) block is used to replace several C3 modules in the final layer of a backbone network to enhance both the contact of the backbone network to global in-formation and the extraction of flame and smoke features; second, a light bidirectional feature pyramid network (Light-BiFPN) is designed to lighten the model while improving the feature extraction and balancing speed and accuracy features during a fire-detection procedure; third, a global attention mechanism (GAM) is fused into the network to cause the model to focus more on the global dimensional features and further improve the detection accuracy of the model; and finally, the Mish activation function and SIoU loss are utilized to simultaneously increase the convergence speed and enhance the accuracy. The experimental results show that compared to the original algorithm, the mean average accuracy (mAP) of Light-YOLOv5 increases by 3.3%, the number of parameters decreases by 27.1%, and the floating point operations (FLOPs) decrease by 19.1%. The detection speed reaches 91.1 FPS, which can detect targets in complex fire scenarios in real time.

Published
2022

22. Comprehensive analyses reveal the carcinogenic and immunological roles of ANLN in human cancers.

Author

Shi, Yanlong, Wang, Menglu, Lan, Sheng, Jian, Haokun, Wang, Yue, Wei, Qian, Zhong, Fei, and Ma, Xinyu

Subjects
Anillin, Biomarker, Cancer, Immune infiltration, Prognosis
Abstract

BACKGROUND: Anillin (ANLN) is an actin-binding protein that is essential for cell division and contributes to cell growth and migration. Although previous studies have shown that ANLN is related to carcinogenesis, no pan-cancer analyses of ANLN have been reported. Accordingly, in this study, we evaluated the carcinogenic roles of ANLN in various cancer types using online databases. METHODS: We evaluated the potential carcinogenic roles of ANLN using TIMER2 and Gene Expression Omnibus databases with 33 types of cancers. We further investigated the associations of ANLN with patient prognosis, genetic alterations, phosphorylation levels, and immune infiltration in multiple cancers using GEPIA2, cBioPortal, UACLAN, and TIMER2 databases. Additionally, the potential functions of ANLN were explored using Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses. Reverse transcription quantitative polymerase chain reaction and immunohistochemistry were used to determine ANLN mRNA and protein expression in colorectal cancer (CRC), gastric cancer (GC), and hepatocellular carcinoma (HCC) cell lines. RESULTS: ANLN was overexpressed in various tumor tissues compared with corresponding normal tissues, and significant correlations between ANLN expression and patient prognosis, genetic alterations, phosphorylation levels, and immune infiltration were noted. Moreover, enrichment analysis suggested that ANLN functionally affected endocytosis, regulation of actin cytoskeleton, and oxytocin signaling pathways. Importantly, ANLN mRNA and protein expression levels were upregulated in gastrointestinal cancers, including CRC, GC, and HCC. CONCLUSIONS: Our findings suggested that ANLN participated in tumorigenesis and cancer progression and may have applications as a promising biomarker of immune infiltration and prognosis in various cancers.

Published
2022

37. The Effect of Classroom-Based Physical Activity Elements on Academic Performance in Children and Adolescents: A Meta-Analysis.

Author

Xu, Yongjin, Lin, Nan, Wu, Chunchun, Wen, Xu, Zhong, Fei, Yu, Kehong, Shu, Li, and Huang, Cong

Subjects
PERFORMANCE in children, ACADEMIC achievement, PHYSICAL activity, RANDOM effects model, METABOLIC equivalent
Abstract

Purpose: We investigated the effects of classroom-based physical activity (CBPA) interventions on academic performance and assessed the impact of different CBPA elements on academic performance. Methods: Relevant experimental studies in four databases were searched from their inception to September 2022. Random effects models were used to compute standardized mean differences (SMDs) and 95% confidence intervals. Results: A total of 13 studies were included. CBPA interventions could improve academic performance in children and adolescents (SMD = 0.17, p =.02). For the CBPA elements, low frequency (no more than three times per week; SMD = 0.19, p =.01), moderate intensity (SMD = 0.19, p =.01), long time (more than 20 min in each intervention; SMD = 0.14, p =.05), integrated physical activity (SMD = 0.18, p =.04), no more than 400 metabolic equivalents of energy-min/week (SMD = 0.33, p =.001), and more than 8-week interventions (SMD = 0.19, p =.04) tended to improve academic performance. Conclusions: CBPA may improve academic performance in children and adolescents. CBPA elements also have an impact on academic performance. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results