SESF-Fuse: an unsupervised deep model for multi-focus image fusion
- PDF / 1,931,652 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 61 Downloads / 160 Views
(0123456789().,-volV)(0123456789(). ,- volV)
ORIGINAL ARTICLE
SESF-Fuse: an unsupervised deep model for multi-focus image fusion Boyuan Ma1,2,3 • Yu Zhu1,2,3 • Xiang Yin1,2,3 • Xiaojuan Ban1,2,3
•
Haiyou Huang1,4 • Michele Mukeshimana5
Received: 12 May 2020 / Accepted: 8 September 2020 Ó Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Muti-focus image fusion is the extraction of focused regions from different images to create one all-in-focus fused image. The key point is that only objects within the depth-of-field have a sharp appearance in the photograph, while other objects are likely to be blurred. We propose an unsupervised deep learning model for multi-focus image fusion. We train an encoder–decoder network in an unsupervised manner to acquire deep features of input images. Then, we utilize spatial frequency, a gradient-based method to measure sharp variation from these deep features, to reflect activity levels. We apply some consistency verification methods to adjust the decision map and draw out the fused result. Our method analyzes sharp appearances in deep features instead of original images, which can be seen as another success story of unsupervised learning in image processing. Experimental results demonstrate that the proposed method achieves state-of-the-art fusion performance compared to 16 fusion methods in objective and subjective assessments, especially in gradient-based fusion metrics. Keywords Multi-focus image fusion Unsupervised deep learning Spatial frequency
1 Introduction Multi-focus image fusion is an important issue in image processing. Optical lenses have the limitation that only objects within the depth-of-field (DOF) have a sharp appearance in a photograph, while other objects are likely to be blurred. Hence, it is difficult for objects at varying & Xiaojuan Ban [email protected] & Haiyou Huang [email protected] 1
Beijing Advanced Innovation Center for Materials Genome Engineering, University of Science and Technology Beijing, Beijing, China
2
Beijing Key Laboratory of Knowledge Engineering for Materials Science, University of Science and Technology Beijing, Beijing, China
3
Institute of Artificial Intelligence, School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing, China
4
Institute for Advanced Materials and Technology, University of Science and Technology Beijing, Beijing, China
5
Faculty of Engineering Sciences, University of Burundi, Bujumbura, Burundi
distances to all be in focus in one camera shot [20]. Many algorithms have been designed to create an all-in-focus image by fusing multiple images that capture the same scene with different focus points. The fused image can be used for human visualization or computer processing, such as feature extraction, segmentation, or object recognition. Deep learning has had great success in image processing, and some multi-fusion methods based on a convolutional neural network (CNN) have been proposed. A supervised CNN-based multi-focus image fus
Data Loading...
