Multi-focus image fusion with a deep convolutional neural network

By Yu Liu, Xun Chen, Hu Peng, Zengfu Wang. Information Fusion, 2017

1,234 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/fcff24b706cd985359c8117d4ef661e54be6ab89

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Jiayi Ma
2. Hui Li
3. Xiaojun Wu
4. Linfeng Tang
5. Josef Kittler
6. Han Xu
7. Hao Zhang
8. Junjun Jiang

In Media Technology

1. Jiayi Ma
2. Hui Li
3. Xiaojun Wu
4. Linfeng Tang
5. Han Xu
6. Josef Kittler
7. Hao Zhang
8. Yiannis Demiris

Citations in the corpus

• Infrared and Visible Image Fusion using a Deep Learning Framework (2018, methodology): “In addition, Yu Liu et al.[14] also proposed a convolutional neural network(CNN)-based fusion method.”
• A review: Deep learning for medical image segmentation using multi-modality fusion (2019, background): “Conventional image fusion strategy learns a direct mapping between source images and target images, the fusion strategy consists of two basic stages: activity level measurement and fusion rule [45].”
• Artificial intelligence in the creative industries: a review (2020, methodology): “cal image fusion. (b) 2D to 3D face conversion generated using the algorithm proposed in [201] all-in-focus image is created using multiple images of the same scene taken with dierent focal settings [189] (Fig. 10 (a)). Multi-exposure deep fusion is used to create high-dynamic range images in [190].”
• Rethinking the Image Fusion: A Fast Unified Image Fusion Network based on Proportional Maintenance of Gradient and Intensity (2020, methodology): “Five multi-focus image fusion methods are selected to compare with our PMGI: CNN (Liu et al. 2017), DctVar (Haghighat, Aghagolzadeh, and Seyedarabi 2011), DWTDE (Liu and Wang 2013), GD (Paul, Sevcenco, and Agathoklis 2016) and QBMF (Bai et al.”
• Medical Image Fusion With Parameter-Adaptive Pulse Coupled Neural Network in Nonsubsampled Shearlet Transform Domain (2019, background): “[30] proposed a convolutional neural network (CNN)-based fusion strategy [32] under the LP-based framework for medical image fusion.”
• FusionGAN: A generative adversarial network for infrared and visible image fusion (2019, methodology): “[26] trained a deep convolutional neural network (CNN) to jointly activity level measurement and fusion rule, and they also applied their model infrared and visible images [27].”
• Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “Liu et al. (2017) used the CNN to jointly generate the activity level measurement and fusion rules for multi-focus image fusion.”
• FusionDN: A Unified Densely Connected Network for Image Fusion (2020, methodology): “In (Liu et al. 2017b), for fusing multi-focus images, a deep CNN trained by high-quality image patches and corresponding artificially made blurred versions is adopted to encode a direct mapping between source images and the focus map.”
• A medical image fusion method based on convolutional neural networks (2017, methodology): “2) The two inputs of the CNN model in [24] are assumed to have similar local structures.”
• NestFuse: An Infrared and Visible Image Fusion Architecture Based on Nest Connection and Spatial/Channel Attention Models (2020, methodology): “In 2017, Liu et al. [24] proposed a convolutional neural network (CNN) [30] used a fusion framework for the multi-focus image fusion task.”
• GANMcC: A Generative Adversarial Network With Multiclassification Constraints for Infrared and Visible Image Fusion (2021, methodology): “In multifocus image fusion, convolutional neural network (CNN) is used for the first time in [28] to learn the mapping from source images to decision maps, in which the network is trained to learn the detection of clear or fuzzy areas by artificially designing false labels.”
• U2Fusion: A Unified Unsupervised Image Fusion Network (2020, methodology): “We compare our method with DSIFT [50], GBM [48], CNN [25], GFDF [8], and SESF-Fuse [18] with qualitative results shown in Fig.”
• SwinFusion: Cross-domain Long-range Learning for General Image Fusion via Swin Transformer (2022, methodology): “Another type of CNN-based image fusion framework is utilizing CNN to learn a direct mapping between source images and fused images (or focus map) in an end-to-end manner [2], [44].”
• LRRNet: A Novel Representation Learning Guided Fusion Network for Infrared and Visible Images (2023, methodology): “The pre-trained CNN is employed to extract deep features from source images and to generate an initial decision map.”