A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections
In the field of biometric recognition, finger vein recognition has received widespread attention by virtue of its advantages, such as biopsy, which is not easy to be stolen. However, due to the limitation of acquisition conditions such as noise and illumination, as well as the limitation of computat...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/10/3691 |
| _version_ | 1797495732690747392 |
|---|---|
| author | Dingzhong Feng Shanyu He Zihao Zhou Ye Zhang |
| author_facet | Dingzhong Feng Shanyu He Zihao Zhou Ye Zhang |
| author_sort | Dingzhong Feng |
| collection | DOAJ |
| description | In the field of biometric recognition, finger vein recognition has received widespread attention by virtue of its advantages, such as biopsy, which is not easy to be stolen. However, due to the limitation of acquisition conditions such as noise and illumination, as well as the limitation of computational resources, the discriminative features are not comprehensive enough when performing finger vein image feature extraction. It will lead to such a result that the accuracy of image recognition cannot meet the needs of large numbers of users and high security. Therefore, this paper proposes a novel feature extraction method called principal component local preservation projections (PCLPP). It organically combines principal component analysis (PCA) and locality preserving projections (LPP) and constructs a projection matrix that preserves both the global and local features of the image, which will meet the urgent needs of large numbers of users and high security. In this paper, we apply the Shandong University homologous multi-modal traits (SDUMLA-HMT) finger vein database to evaluate PCLPP and add “Salt and pepper” noise to the dataset to verify the robustness of PCLPP. The experimental results show that the image recognition rate after applying PCLPP is much better than the other two methods, PCA and LPP, for feature extraction. |
| first_indexed | 2024-03-10T01:53:50Z |
| format | Article |
| id | doaj.art-3c8a034cfc4d4ab7a790a0e9deb6ec8b |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T01:53:50Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-3c8a034cfc4d4ab7a790a0e9deb6ec8b2023-11-23T12:59:31ZengMDPI AGSensors1424-82202022-05-012210369110.3390/s22103691A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving ProjectionsDingzhong Feng0Shanyu He1Zihao Zhou2Ye Zhang3College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou 310014, ChinaCollege of Mechanical Engineering, Zhejiang University of Technology, Hangzhou 310014, ChinaCollege of Mechanical Engineering, Zhejiang University of Technology, Hangzhou 310014, ChinaCollege of Mechanical Engineering, Zhejiang University of Technology, Hangzhou 310014, ChinaIn the field of biometric recognition, finger vein recognition has received widespread attention by virtue of its advantages, such as biopsy, which is not easy to be stolen. However, due to the limitation of acquisition conditions such as noise and illumination, as well as the limitation of computational resources, the discriminative features are not comprehensive enough when performing finger vein image feature extraction. It will lead to such a result that the accuracy of image recognition cannot meet the needs of large numbers of users and high security. Therefore, this paper proposes a novel feature extraction method called principal component local preservation projections (PCLPP). It organically combines principal component analysis (PCA) and locality preserving projections (LPP) and constructs a projection matrix that preserves both the global and local features of the image, which will meet the urgent needs of large numbers of users and high security. In this paper, we apply the Shandong University homologous multi-modal traits (SDUMLA-HMT) finger vein database to evaluate PCLPP and add “Salt and pepper” noise to the dataset to verify the robustness of PCLPP. The experimental results show that the image recognition rate after applying PCLPP is much better than the other two methods, PCA and LPP, for feature extraction.https://www.mdpi.com/1424-8220/22/10/3691finger vein recognitionbiometric recognitionfeature extraction methodalgorithm |
| spellingShingle | Dingzhong Feng Shanyu He Zihao Zhou Ye Zhang A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections Sensors finger vein recognition biometric recognition feature extraction method algorithm |
| title | A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections |
| title_full | A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections |
| title_fullStr | A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections |
| title_full_unstemmed | A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections |
| title_short | A Finger Vein Feature Extraction Method Incorporating Principal Component Analysis and Locality Preserving Projections |
| title_sort | finger vein feature extraction method incorporating principal component analysis and locality preserving projections |
| topic | finger vein recognition biometric recognition feature extraction method algorithm |
| url | https://www.mdpi.com/1424-8220/22/10/3691 |
| work_keys_str_mv | AT dingzhongfeng afingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT shanyuhe afingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT zihaozhou afingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT yezhang afingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT dingzhongfeng fingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT shanyuhe fingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT zihaozhou fingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections AT yezhang fingerveinfeatureextractionmethodincorporatingprincipalcomponentanalysisandlocalitypreservingprojections |