相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Han Xu et al.
Summary: This study proposes a novel unified and unsupervised end-to-end image fusion network, U2Fusion, which can solve different fusion problems. By training the model with adaptive information preservation, the network avoids the requirements of previous image fusion methods and has broad applicability. Moreover, a new dataset RoadScene is released for evaluation and comparison.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Engineering, Electrical & Electronic
Zixiang Zhao et al.
Summary: In this paper, a model-based convolutional neural network model called Algorithm Unrolling Image Fusion (AUIF) is proposed for infrared and visible image fusion. The AUIF model combines traditional optimization models and CNN layers, allowing for decomposition of low-frequency base and high-frequency detail information from source images. Through training and testing phases, the model can robustly generate fusion images containing highlight targets and legible details, surpassing existing methods in terms of speed and efficiency.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Linfeng Tang et al.
Summary: This paper proposes a semantic-aware real-time image fusion network (SeAFusion), which effectively boosts the performance of high-level vision tasks on fused images and outperforms existing alternatives in terms of maintaining pixel intensity distribution and preserving texture detail.
INFORMATION FUSION
(2022)
Article
Automation & Control Systems
Jiayi Ma et al.
Summary: This study proposes a novel image fusion framework called SwinFusion, which combines cross-domain long-range learning and Swin Transformer. The framework integrates complementary information and achieves global interaction through attention-guided cross-domain modules. It also addresses multi-scene image fusion problems by preserving structure, detail, and intensity. Extensive experiments prove the superiority of SwinFusion compared to other state-of-the-art fusion algorithms. The implementation code and pre-trained weights are available at https://github.com/Linfeng-Tang/SwinFusion.
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Jinyuan Liu et al.
Summary: This study introduces a bilevel optimization formulation for fusing infrared and visible images, and utilizes the TarDAL network for fusion and detection. The results show that the proposed method not only produces visually appealing fused images but also achieves higher detection mAP compared to state-of-the-art approaches.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Article
Computer Science, Information Systems
Rencan Nie et al.
Summary: This paper proposes a total variation-based fusion method for infrared and visible images, which uses weighted fidelity to merge infrared objects and salient scenes. It also utilizes joint norms and constraints for better results.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Artificial Intelligence
Hao Zhang et al.
Summary: Image fusion aims to extract and combine meaningful information from different source images to generate a single informative image. The development of deep learning, with techniques like generative adversarial networks and autoencoders, has significantly advanced image fusion. However, a comprehensive review of the latest deep-learning methods in different fusion scenarios is lacking, which is addressed in this survey.
INFORMATION FUSION
(2021)
Article
Computer Science, Artificial Intelligence
Guofa Li et al.
Summary: This paper proposes a new infrared and visible image fusion method based on multi-scale transformation and norm optimization, which uses a new loss function and the split Bregman method for image fusion. Experimental results show that the method outperforms others in highlighting targets and retaining effective detail information.
INFORMATION FUSION
(2021)
Article
Computer Science, Artificial Intelligence
Hui Li et al.
Summary: In the field of image fusion, designing deep learning-based fusion methods is challenging due to the need to choose an appropriate fusion strategy for specific tasks. The study introduces a novel end-to-end fusion network architecture and utilizes a residual architecture to replace traditional fusion methods, with proposed loss functions for training and a two-stage training strategy, achieving superior performance to existing methods.
INFORMATION FUSION
(2021)
Article
Engineering, Electrical & Electronic
Han Xu et al.
Summary: In this article, a novel decomposition method for visible and infrared image fusion (DRF) is proposed, which disentangles images into scene- and sensor modality-related representations and applies different fusion strategies, leading to comparable performance in terms of visual effect and quantitative metrics compared to the state of the art.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Jiayi Ma et al.
Summary: The proposed STDFusionNet is a fusion network for infrared and visible images that preserves thermal targets and texture structures. By utilizing salient target detection and a specific loss function, it successfully extracts and reconstructs features to achieve high-quality fusion results. Through qualitative and quantitative experiments, the algorithm has shown superiority in terms of speed and quality improvement over existing methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Han Xu et al.
Summary: This study achieves interpretable importance evaluation of feature maps in a deep learning manner and proposes a pixel-wise classification saliency-based fusion rule, which reduces distortion without human intervention.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2021)
Article
Computer Science, Information Systems
Jing Li et al.
Summary: In this paper, a method named AttentionFGAN is proposed to fuse infrared and visible images by integrating multi-scale attention mechanism into Generative Adversarial Networks (GAN). The generator and discriminator both apply attention mechanism to emphasize the focus on typical regions of source images during fusion. Ablation experiments demonstrate the effectiveness of the method, and extensive qualitative and quantitative experiments on three public datasets show the advantages of AttentionFGAN compared to other state-of-the-art methods.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Computer Science, Artificial Intelligence
Huafeng Li et al.
Summary: This paper proposes a meta learning-based deep framework for the fusion of infrared and visible images, which is flexible and capable of accepting input source images of different resolutions to generate the fused image of arbitrary resolution. The framework enhances the capability of detail extraction in the fused image through modules such as feature extraction, dual attention mechanism-based feature fusion, and residual compensation.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Engineering, Electrical & Electronic
Lihua Jian et al.
Summary: The proposed image fusion method SEDRFuse utilizes a neural network to extract features and fuse infrared and visible images, achieving a high-quality fusion result. Experimental results demonstrate that SEDRFuse outperforms existing methods in terms of fusion effects.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Jiayi Ma et al.
Summary: Combining visible and infrared images into a single image requires balancing contrast and texture details, the proposed GANMcC framework achieves more reasonable fusion by introducing multiclassification constraints and specific content loss to constrain the generator for better information extraction from source images. Extensive experiments demonstrate the advantages of GANMcC over state-of-the-art methods in terms of qualitative effect and quantitative metric, even for overexposed visible images.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Artificial Intelligence
Yu Zhang et al.
INFORMATION FUSION
(2020)
Article
Computer Science, Information Systems
Jun Chen et al.
INFORMATION SCIENCES
(2020)
Article
Computer Science, Artificial Intelligence
Jiayi Ma et al.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2020)
Article
Engineering, Electrical & Electronic
Hui Li et al.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2020)
Article
Engineering, Electrical & Electronic
Ruichao Hou et al.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2020)
Article
Computer Science, Artificial Intelligence
Jiayi Ma et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2020)
Article
Computer Science, Artificial Intelligence
Hui Li et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2019)
Article
Computer Science, Artificial Intelligence
Jiayi Ma et al.
INFORMATION FUSION
(2019)
Article
Multidisciplinary Sciences
Alexander Toet
Proceedings Paper
Computer Science, Artificial Intelligence
Gao Huang et al.
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
(2017)
Article
Computer Science, Artificial Intelligence
Jiayi Ma et al.
INFORMATION FUSION
(2016)
Proceedings Paper
Computer Science, Artificial Intelligence
Ali Borji et al.
COMPUTER VISION - ECCV 2012, PT II
(2012)
Article
Computer Science, Artificial Intelligence
HR Sheikh et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2006)
Article
Engineering, Electrical & Electronic
GH Qu et al.
ELECTRONICS LETTERS
(2002)