An Improved Lightweight Network Based on MobileNetV3 for Palmprint Recognition

Zhou, Kaijun; Deng, Kailai; Chen, Peng; YiLiang, Hu

doi:10.1007/978-3-031-18907-4_58

Cited by 4 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The research on the basic theory, functional devices and engineering applications of electromagnetic metamaterials has aroused extensive research interests in the fields of physics, information and materials (Baqir, 2019;Baqir, 2020;Pan et al, 2022a;Baqir and Choudhury, 2022;. Based on the designable electromagnetic parameters and distribution of metamaterials, researchers have developed various new devices such as metamaterial cloaks, lenses, and antennas (Li et al, 2021a;Wang et al, 2022a;Yang et al, 2022;Zhao and Wang, 2022;Wang et al, 2023a). According to the field localization and field enhancement characteristics of metamaterials (this effect is especially significant in surface plasmon metamaterials) and subwavelength scale characteristics (Li et al, 2020a;Wang et al, 2022b;Cao, 2022;Liu et al, 2022;Xie et al, 2023), a novel metamaterial sensing and imaging device has been developed, which can effectively improve the sensor's performance.…”

Section: Introductionmentioning

confidence: 99%

Design and performance evaluation of a novel metamaterial broadband THz filter for 6G applications

Althuwayb,

Rashid,

Elhamrawy

et al. 2023

Front. Mater.

View full text Add to dashboard Cite

Terahertz (THz) radiation, which has applications in the imaging of objects, non-destructive testing, satellite communication, medical diagnostics, and biosensing, has generated a great deal of attention due to its remarkable properties. This paper proposes a novel broadband filter for THz applications. The main idea is to overcome the insertion loss and bandwidth issues by modeling a frequency-domain finite difference method and guided-mode resonance (GMR). The optimal design scheme of the wideband pass filter based on the circular resonant ring is discussed by comparing the transmission parameters under various parameters. This scheme overcomes the restriction of the narrow passband bandwidth of the prior THz filters and achieves approximately 3 dB bandwidth of 0.54 THz. The proposed THz filter paper also has the advantages of a straightforward structure, low processing costs, and ease of conformal with other structures, and it can be used for stealth fighters, new communication technology, and precise instruments. In addition, when compared to existing models, the suggested filter offers higher 3 dB BW operation, increased transmittance, low insertion loss, and stable performance at various oblique angles.

show abstract

Section: Introductionmentioning

confidence: 99%

Design and performance evaluation of a novel metamaterial broadband THz filter for 6G applications

Althuwayb,

Rashid,

Elhamrawy

et al. 2023

Front. Mater.

View full text Add to dashboard Cite

show abstract

“…Experimental results demonstrate a notable enhancement in Top-1 accuracy of 0.8% compared to GhostNet, while requiring less computational resources. To further enhance the inference speed of the MobileNetV3 network, Zhao et al proposed a lightweight network based on MobileNetV3 [20]. Instead of utilizing the residual blocks found in MobileNetV3, they introduced a specially designed partial residual structure.…”

Section: Related Workmentioning

confidence: 99%

A Pedestrian Detection Algorithm Based on Improved YOLOv7-tiny

2024

JSSH

View full text Add to dashboard Cite

With the widespread development of embedded device applications based on deep learning, pedestrian detection using deep learning often requires the use of complex and large neural networks. However, these networks have a large memory footprint, which makes them difficult to deploy on embedded devices, and are computationally intensive, which makes it difficult to guarantee the speed of operation on embedded devices. To address these issues, a lightweight network, YOLO-Micro, is proposed in this paper. The model reduces the complexity of the model by replacing the original module in YOLOv7-tiny with the G-ELAN module proposed in this paper. Meanwhile, the attention mechanism and the activation function were introduced and replaced to enhance the feature extraction capability of the model. Finally, the model was successfully deployed to Android mobile. Experimental results on the PASCAL VOC Pedestrian dataset show that the inference speed of YOLO-Micro is 20.8% faster than the original model, and the mAP is improved from 77.95% to 79.42% of the original model.

show abstract