A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification
Feature map reconstruction networks (FRN) have demonstrated significant potential by leveraging feature reconstruction. However, the typical process of FRN gives rise to two notable issues. First, FRN exhibits high sensitivity to noise, particularly ambient noise, which can lead to substantial recon...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/7/1098 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849730292456095744 |
|---|---|
| author | Meijia Wang Boyuan Zheng Guochao Wang Junpo Yang Jin Lu Weichuan Zhang |
| author_facet | Meijia Wang Boyuan Zheng Guochao Wang Junpo Yang Jin Lu Weichuan Zhang |
| author_sort | Meijia Wang |
| collection | DOAJ |
| description | Feature map reconstruction networks (FRN) have demonstrated significant potential by leveraging feature reconstruction. However, the typical process of FRN gives rise to two notable issues. First, FRN exhibits high sensitivity to noise, particularly ambient noise, which can lead to substantial reconstruction errors and hinder the network’s ability to extract meaningful features. Second, FRN is particularly vulnerable to changes in data distribution. Owing to the fine-grained nature of the training data, the model is highly susceptible to overfitting, which may compromise its ability to extract effective feature representations when confronted with new classes. To address these challenges, this paper proposes a novel main feature selection module (MFSM), which suppresses feature noise interference and enhances the discriminative capacity of feature representations through principal component analysis (PCA). Extensive experiments validate the effectiveness of MFSM, revealing substantial improvements in classification accuracy for few-shot fine-grained image classification (FSFGIC) tasks. |
| format | Article |
| id | doaj-art-ebf7764ee7d64fd6a93d984e7f24d008 |
| institution | DOAJ |
| issn | 2227-7390 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-ebf7764ee7d64fd6a93d984e7f24d0082025-08-20T03:08:55ZengMDPI AGMathematics2227-73902025-03-01137109810.3390/math13071098A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image ClassificationMeijia Wang0Boyuan Zheng1Guochao Wang2Junpo Yang3Jin Lu4Weichuan Zhang5School of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaFeature map reconstruction networks (FRN) have demonstrated significant potential by leveraging feature reconstruction. However, the typical process of FRN gives rise to two notable issues. First, FRN exhibits high sensitivity to noise, particularly ambient noise, which can lead to substantial reconstruction errors and hinder the network’s ability to extract meaningful features. Second, FRN is particularly vulnerable to changes in data distribution. Owing to the fine-grained nature of the training data, the model is highly susceptible to overfitting, which may compromise its ability to extract effective feature representations when confronted with new classes. To address these challenges, this paper proposes a novel main feature selection module (MFSM), which suppresses feature noise interference and enhances the discriminative capacity of feature representations through principal component analysis (PCA). Extensive experiments validate the effectiveness of MFSM, revealing substantial improvements in classification accuracy for few-shot fine-grained image classification (FSFGIC) tasks.https://www.mdpi.com/2227-7390/13/7/1098few-shot fine-grained image classificationprincipal component analysisprincipal component analysis-based feature optimization network |
| spellingShingle | Meijia Wang Boyuan Zheng Guochao Wang Junpo Yang Jin Lu Weichuan Zhang A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification Mathematics few-shot fine-grained image classification principal component analysis principal component analysis-based feature optimization network |
| title | A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification |
| title_full | A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification |
| title_fullStr | A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification |
| title_full_unstemmed | A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification |
| title_short | A Principal Component Analysis-Based Feature Optimization Network for Few-Shot Fine-Grained Image Classification |
| title_sort | principal component analysis based feature optimization network for few shot fine grained image classification |
| topic | few-shot fine-grained image classification principal component analysis principal component analysis-based feature optimization network |
| url | https://www.mdpi.com/2227-7390/13/7/1098 |
| work_keys_str_mv | AT meijiawang aprincipalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT boyuanzheng aprincipalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT guochaowang aprincipalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT junpoyang aprincipalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT jinlu aprincipalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT weichuanzhang aprincipalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT meijiawang principalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT boyuanzheng principalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT guochaowang principalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT junpoyang principalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT jinlu principalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification AT weichuanzhang principalcomponentanalysisbasedfeatureoptimizationnetworkforfewshotfinegrainedimageclassification |