Scale-Space Feature Recalibration Network for Single Image Deraining
Computer vision technology is increasingly being used in areas such as intelligent security and autonomous driving. Users need accurate and reliable visual information, but the images obtained under severe weather conditions are often disturbed by rainy weather, causing image scenes to look blurry....
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/18/6823 |
| _version_ | 1797482606034419712 |
|---|---|
| author | Pengpeng Li Jiyu Jin Guiyue Jin Lei Fan |
| author_facet | Pengpeng Li Jiyu Jin Guiyue Jin Lei Fan |
| author_sort | Pengpeng Li |
| collection | DOAJ |
| description | Computer vision technology is increasingly being used in areas such as intelligent security and autonomous driving. Users need accurate and reliable visual information, but the images obtained under severe weather conditions are often disturbed by rainy weather, causing image scenes to look blurry. Many current single image deraining algorithms achieve good performance but have limitations in retaining detailed image information. In this paper, we design a Scale-space Feature Recalibration Network (SFR-Net) for single image deraining. The proposed network improves the image feature extraction and characterization capability of a Multi-scale Extraction Recalibration Block (MERB) using dilated convolution with different convolution kernel sizes, which results in rich multi-scale rain streaks features. In addition, we develop a Subspace Coordinated Attention Mechanism (SCAM) and embed it into MERB, which combines coordinated attention recalibration and a subspace attention mechanism to recalibrate the rain streaks feature information learned from the feature extraction phase and eliminate redundant feature information to enhance the transfer of important feature information. Meanwhile, the overall SFR-Net structure uses dense connection and cross-layer feature fusion to repeatedly utilize the feature maps, thus enhancing the understanding of the network and avoiding gradient disappearance. Through extensive experiments on synthetic and real datasets, the proposed method outperforms the recent state-of-the-art deraining algorithms in terms of both the rain removal effect and the preservation of image detail information. |
| first_indexed | 2024-03-09T22:34:49Z |
| format | Article |
| id | doaj.art-6c81c49bef4e47f199163645157f9f39 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T22:34:49Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-6c81c49bef4e47f199163645157f9f392023-11-23T18:49:53ZengMDPI AGSensors1424-82202022-09-012218682310.3390/s22186823Scale-Space Feature Recalibration Network for Single Image DerainingPengpeng Li0Jiyu Jin1Guiyue Jin2Lei Fan3School of Information Science and Engineering, Dalian Polytechnic University, Dalian 116034, ChinaSchool of Information Science and Engineering, Dalian Polytechnic University, Dalian 116034, ChinaSchool of Information Science and Engineering, Dalian Polytechnic University, Dalian 116034, ChinaSchool of Information Science and Engineering, Dalian Polytechnic University, Dalian 116034, ChinaComputer vision technology is increasingly being used in areas such as intelligent security and autonomous driving. Users need accurate and reliable visual information, but the images obtained under severe weather conditions are often disturbed by rainy weather, causing image scenes to look blurry. Many current single image deraining algorithms achieve good performance but have limitations in retaining detailed image information. In this paper, we design a Scale-space Feature Recalibration Network (SFR-Net) for single image deraining. The proposed network improves the image feature extraction and characterization capability of a Multi-scale Extraction Recalibration Block (MERB) using dilated convolution with different convolution kernel sizes, which results in rich multi-scale rain streaks features. In addition, we develop a Subspace Coordinated Attention Mechanism (SCAM) and embed it into MERB, which combines coordinated attention recalibration and a subspace attention mechanism to recalibrate the rain streaks feature information learned from the feature extraction phase and eliminate redundant feature information to enhance the transfer of important feature information. Meanwhile, the overall SFR-Net structure uses dense connection and cross-layer feature fusion to repeatedly utilize the feature maps, thus enhancing the understanding of the network and avoiding gradient disappearance. Through extensive experiments on synthetic and real datasets, the proposed method outperforms the recent state-of-the-art deraining algorithms in terms of both the rain removal effect and the preservation of image detail information.https://www.mdpi.com/1424-8220/22/18/6823image derainingmulti-scaleattention recalibrationfeature fusion |
| spellingShingle | Pengpeng Li Jiyu Jin Guiyue Jin Lei Fan Scale-Space Feature Recalibration Network for Single Image Deraining Sensors image deraining multi-scale attention recalibration feature fusion |
| title | Scale-Space Feature Recalibration Network for Single Image Deraining |
| title_full | Scale-Space Feature Recalibration Network for Single Image Deraining |
| title_fullStr | Scale-Space Feature Recalibration Network for Single Image Deraining |
| title_full_unstemmed | Scale-Space Feature Recalibration Network for Single Image Deraining |
| title_short | Scale-Space Feature Recalibration Network for Single Image Deraining |
| title_sort | scale space feature recalibration network for single image deraining |
| topic | image deraining multi-scale attention recalibration feature fusion |
| url | https://www.mdpi.com/1424-8220/22/18/6823 |
| work_keys_str_mv | AT pengpengli scalespacefeaturerecalibrationnetworkforsingleimagederaining AT jiyujin scalespacefeaturerecalibrationnetworkforsingleimagederaining AT guiyuejin scalespacefeaturerecalibrationnetworkforsingleimagederaining AT leifan scalespacefeaturerecalibrationnetworkforsingleimagederaining |