Few-shot image recognition based on multi-scale features prototypical network

被引：0

作者：

Liu, Jiatong ^{[1
]}

Duan, Yong ^{[1
]}

机构：

[1] School of Information Science Engineering, Shenyang University of Technology, Shenyang 110870, P. R. China Shenyang Key Laboratory of Advanced Computing and Application Innovation, Shenyang 110870, P. R. China

来源：

High Technology Letters | 2024年 / 30卷 / 03期

关键词：

In order to improve the model’s capability in expressing features during few-shot learning; a multi-scale features prototypical network (MS-PN) algorithm is proposed. The metric learning algorithm is employed to extract image features and project them into a feature space; thus evaluating the similarity between samples based on their relative distances within the metric space. To sufficiently extract feature information from limited sample data and mitigate the impact of constrained data volume; a multi-scale feature extraction network is presented to capture data features at various scales during the process of image feature extraction. Additionally; the position of the prototype is fine-tuned by assigning weights to data points to mitigate the influence of outliers on the experiment. The loss function integrates contrastive loss and label-smoothing to bring similar data points closer and separate dissimilar data points within the metric space. Experimental evaluations are conducted on small-sample datasets mini-ImageNet and CUB200-2011. The method in this paper can achieve higher classification accuracy. Specifically; in the 5-way 1-shot experiment; classification accuracy reaches 50. 13% and 66. 79% respectively on these two datasets. Moreover; in the 5-way 5-shot experiment; accuracy of 66. 79% and 85. 91% are observed; respectively. © 2024 Inst. of Scientific and Technical Information of China. All rights reserved;

D O I：

10.3772/j.issn.1006-6748.2024.03.007

中图分类号：

学科分类号：

摘要：

引用

页码：280 / 289

共 50 条

[41] Few-shot wildlife detection based on multi-scale context extraction
Liu, Ke
Lin, Shanling
Shi, Xinyu
Lin, Jianpu
Lu, Shanhong
Lin, Zhixian
Guo, Tailiang
CHINESE JOURNAL OF LIQUID CRYSTALS AND DISPLAYS, 2025, 40 (03) : 516 - 526
[42] Boosting Few-shot Image Recognition via Domain Alignment Prototypical Networks
Lu, Jiang
Cao, Zhong
Wu, Kailun
Zhang, Gang
Zhang, Changshui
2018 IEEE 30TH INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE (ICTAI), 2018, : 260 - 264
[43] Multi-scale feature self-enhancement network for few-shot learning
Dong, Bowen
Wang, Ronggui
Yang, Juan
Xue, Lixia
MULTIMEDIA TOOLS AND APPLICATIONS, 2021, 80 (25) : 33865 - 33883
[44] MARANet: Multi-scale Adaptive Region Attention Network for Few-Shot Learning
Chen, Jia
Li, Xiyang
Ou, Yangjun
Hu, Xinrong
Peng, Tao
ADVANCES IN COMPUTER GRAPHICS, CGI 2023, PT I, 2024, 14495 : 415 - 426
[45] Multi-scale attentional similarity guidance network for few-shot semantic segmentation
Ze-yu Liu
Jian-wei Liu
Neural Computing and Applications, 2022, 34 : 18895 - 18915
[46] Multi-scale feature self-enhancement network for few-shot learning
Bowen Dong
Ronggui Wang
Juan Yang
Lixia Xue
Multimedia Tools and Applications, 2021, 80 : 33865 - 33883
[47] A multi-scale hierarchical node graph neural network for few-shot learning
Zhang, Yan
Zhou, Xudong
Wang, Ke
Wang, Nian
Li, Zenghui
MULTIMEDIA TOOLS AND APPLICATIONS, 2023, 83 (20) : 58201 - 58223
[48] Multi-Scale Decision Network With Feature Fusion and Weighting for Few-Shot Learning
Wang, Xiaoru
Ma, Bing
Yu, Zhihong
Li, Fu
Cai, Yali
IEEE ACCESS, 2020, 8 : 92172 - 92181
[49] Multi-scale attentional similarity guidance network for few-shot semantic segmentation
Liu, Ze-yu
Liu, Jian-wei
NEURAL COMPUTING & APPLICATIONS, 2022, 34 (21): : 18895 - 18915
[50] Fish Image Recognition Based on Residual Network and Few-shot Learning
Yuan P.
Song J.
Xu H.
Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery, 2022, 53 (02): : 282 - 290

← 1 2 3 4 5 →