MMPCANet: An Improved PCANet for Occluded Face Recognition
Principal Component Analysis Network (PCANet) is a lightweight deep learning network, which is fast and effective in face recognition. However, the accuracy of faces with occlusion does not meet the optimal requirement for two reasons: 1. PCANet needs to stretch the two-dimensional images into colum...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/6/3144 |
| _version_ | 1797472920352587776 |
|---|---|
| author | Zewei Wang Yongjun Zhang Chengchang Pan Zhongwei Cui |
| author_facet | Zewei Wang Yongjun Zhang Chengchang Pan Zhongwei Cui |
| author_sort | Zewei Wang |
| collection | DOAJ |
| description | Principal Component Analysis Network (PCANet) is a lightweight deep learning network, which is fast and effective in face recognition. However, the accuracy of faces with occlusion does not meet the optimal requirement for two reasons: 1. PCANet needs to stretch the two-dimensional images into column vectors, which causes the loss of essential image spatial information; 2. When the training samples are few, the recognition accuracy of PCANet is low. To solve the above problems, this paper proposes a multi-scale and multi-layer feature fusion-based PCANet (MMPCANet) for occluded face recognition. Firstly, a channel-wise concatenation of the original image features and the output features of the first layer is conducted, and then the concatenated result is used as the input of the second layer; therefore, more image feature information is used. In addition, to avoid the loss of image spatial information, a spatial pyramid is used as the feature pooling layer of the network. Finally, the feature vector is sent to the random forest classifier for classification. The proposed algorithm is tested on several widely used facial image databases and compared with other similar algorithms. Our experimental results show that the proposed algorithm effectively improves the efficiency of the network training and the recognition accuracy of occluded faces under the same training and testing datasets. The average accuracies are 98.78% on CelebA, 97.58% on AR, and 97.15% on FERET. |
| first_indexed | 2024-03-09T20:08:53Z |
| format | Article |
| id | doaj.art-9972edef74f045e6a0bdda1295cdcaba |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T20:08:53Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-9972edef74f045e6a0bdda1295cdcaba2023-11-24T00:24:25ZengMDPI AGApplied Sciences2076-34172022-03-01126314410.3390/app12063144MMPCANet: An Improved PCANet for Occluded Face RecognitionZewei Wang0Yongjun Zhang1Chengchang Pan2Zhongwei Cui3Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province, College of Computer Science and Technology, Guizhou University, Guiyang 550025, ChinaKey Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province, College of Computer Science and Technology, Guizhou University, Guiyang 550025, ChinaKey Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province, College of Computer Science and Technology, Guizhou University, Guiyang 550025, ChinaSchool of Mathematics and Big Data, Guizhou Education University, Guiyang 550018, ChinaPrincipal Component Analysis Network (PCANet) is a lightweight deep learning network, which is fast and effective in face recognition. However, the accuracy of faces with occlusion does not meet the optimal requirement for two reasons: 1. PCANet needs to stretch the two-dimensional images into column vectors, which causes the loss of essential image spatial information; 2. When the training samples are few, the recognition accuracy of PCANet is low. To solve the above problems, this paper proposes a multi-scale and multi-layer feature fusion-based PCANet (MMPCANet) for occluded face recognition. Firstly, a channel-wise concatenation of the original image features and the output features of the first layer is conducted, and then the concatenated result is used as the input of the second layer; therefore, more image feature information is used. In addition, to avoid the loss of image spatial information, a spatial pyramid is used as the feature pooling layer of the network. Finally, the feature vector is sent to the random forest classifier for classification. The proposed algorithm is tested on several widely used facial image databases and compared with other similar algorithms. Our experimental results show that the proposed algorithm effectively improves the efficiency of the network training and the recognition accuracy of occluded faces under the same training and testing datasets. The average accuracies are 98.78% on CelebA, 97.58% on AR, and 97.15% on FERET.https://www.mdpi.com/2076-3417/12/6/3144PCANetface recognitionoccluded faceMMPCANetdeep learning |
| spellingShingle | Zewei Wang Yongjun Zhang Chengchang Pan Zhongwei Cui MMPCANet: An Improved PCANet for Occluded Face Recognition Applied Sciences PCANet face recognition occluded face MMPCANet deep learning |
| title | MMPCANet: An Improved PCANet for Occluded Face Recognition |
| title_full | MMPCANet: An Improved PCANet for Occluded Face Recognition |
| title_fullStr | MMPCANet: An Improved PCANet for Occluded Face Recognition |
| title_full_unstemmed | MMPCANet: An Improved PCANet for Occluded Face Recognition |
| title_short | MMPCANet: An Improved PCANet for Occluded Face Recognition |
| title_sort | mmpcanet an improved pcanet for occluded face recognition |
| topic | PCANet face recognition occluded face MMPCANet deep learning |
| url | https://www.mdpi.com/2076-3417/12/6/3144 |
| work_keys_str_mv | AT zeweiwang mmpcanetanimprovedpcanetforoccludedfacerecognition AT yongjunzhang mmpcanetanimprovedpcanetforoccludedfacerecognition AT chengchangpan mmpcanetanimprovedpcanetforoccludedfacerecognition AT zhongweicui mmpcanetanimprovedpcanetforoccludedfacerecognition |