Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain
In order to obtain a panoramic image which is clearer, and has more layers and texture features, we propose an innovative multi-focus image fusion algorithm by combining with non-subsampled shearlet transform (NSST) and residual network (ResNet). First, NSST decomposes a pair of input images to prod...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8880627/ |
| _version_ | 1818620867215622144 |
|---|---|
| author | Shuaiqi Liu Jie Wang Yucong Lu Shaohai Hu Xiaole Ma Yifei Wu |
| author_facet | Shuaiqi Liu Jie Wang Yucong Lu Shaohai Hu Xiaole Ma Yifei Wu |
| author_sort | Shuaiqi Liu |
| collection | DOAJ |
| description | In order to obtain a panoramic image which is clearer, and has more layers and texture features, we propose an innovative multi-focus image fusion algorithm by combining with non-subsampled shearlet transform (NSST) and residual network (ResNet). First, NSST decomposes a pair of input images to produce subband coefficients of different frequencies for subsequent feature processing. Then, ResNet is applied to fuse the low frequency subband coefficients, and improved gradient sum of Laplace energy (IGSML) perform high frequency feature information processing. Finally, the inverse NSST is performed on the fused coefficients of different frequencies to obtain the final fused image. In our method, we fully consider the low frequency global features and high frequency detail information in image by using NSST. For low-frequency coefficients fusion, we can also obtain the spatial information features of low-frequency coefficient images by using ResNet, which has a deep network structure. IGSML can use different directional gradients to process high-frequency subband coefficients of different levels and directions, which is more conducive to the fusion of the coefficients. The experiment results show that the proposed method has been improved in the structural features and edge texture in the fusion images. |
| first_indexed | 2024-12-16T18:00:12Z |
| format | Article |
| id | doaj.art-47f53318d7fd4462aa5ba37b623de1af |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-16T18:00:12Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-47f53318d7fd4462aa5ba37b623de1af2022-12-21T22:22:04ZengIEEEIEEE Access2169-35362019-01-01715204315206310.1109/ACCESS.2019.29473788880627Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet DomainShuaiqi Liu0https://orcid.org/0000-0001-7520-8226Jie Wang1Yucong Lu2Shaohai Hu3Xiaole Ma4https://orcid.org/0000-0001-7578-7969Yifei Wu5College of Electronic and Information Engineering, Hebei University, Baoding, ChinaCollege of Electronic and Information Engineering, Hebei University, Baoding, ChinaCollege of Electronic and Information Engineering, Hebei University, Baoding, ChinaInstitute of Information Science, Beijing Jiaotong University, Beijing, ChinaInstitute of Information Science, Beijing Jiaotong University, Beijing, ChinaDepartment of Electrical and Computer Engineering, University of California at San Diego, San Diego, CA, USAIn order to obtain a panoramic image which is clearer, and has more layers and texture features, we propose an innovative multi-focus image fusion algorithm by combining with non-subsampled shearlet transform (NSST) and residual network (ResNet). First, NSST decomposes a pair of input images to produce subband coefficients of different frequencies for subsequent feature processing. Then, ResNet is applied to fuse the low frequency subband coefficients, and improved gradient sum of Laplace energy (IGSML) perform high frequency feature information processing. Finally, the inverse NSST is performed on the fused coefficients of different frequencies to obtain the final fused image. In our method, we fully consider the low frequency global features and high frequency detail information in image by using NSST. For low-frequency coefficients fusion, we can also obtain the spatial information features of low-frequency coefficient images by using ResNet, which has a deep network structure. IGSML can use different directional gradients to process high-frequency subband coefficients of different levels and directions, which is more conducive to the fusion of the coefficients. The experiment results show that the proposed method has been improved in the structural features and edge texture in the fusion images.https://ieeexplore.ieee.org/document/8880627/Image fusionmulti-focus image fusionNSSTResNet |
| spellingShingle | Shuaiqi Liu Jie Wang Yucong Lu Shaohai Hu Xiaole Ma Yifei Wu Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain IEEE Access Image fusion multi-focus image fusion NSST ResNet |
| title | Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain |
| title_full | Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain |
| title_fullStr | Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain |
| title_full_unstemmed | Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain |
| title_short | Multi-Focus Image Fusion Based on Residual Network in Non-Subsampled Shearlet Domain |
| title_sort | multi focus image fusion based on residual network in non subsampled shearlet domain |
| topic | Image fusion multi-focus image fusion NSST ResNet |
| url | https://ieeexplore.ieee.org/document/8880627/ |
| work_keys_str_mv | AT shuaiqiliu multifocusimagefusionbasedonresidualnetworkinnonsubsampledshearletdomain AT jiewang multifocusimagefusionbasedonresidualnetworkinnonsubsampledshearletdomain AT yuconglu multifocusimagefusionbasedonresidualnetworkinnonsubsampledshearletdomain AT shaohaihu multifocusimagefusionbasedonresidualnetworkinnonsubsampledshearletdomain AT xiaolema multifocusimagefusionbasedonresidualnetworkinnonsubsampledshearletdomain AT yifeiwu multifocusimagefusionbasedonresidualnetworkinnonsubsampledshearletdomain |