In this paper, we present a novel deep learning architecture for infrared and visible images fusion problems. In contrast to conventional convolutional networks, our encoding network is combined with ...convolutional layers, a fusion layer, and dense block in which the output of each layer is connected to every other layer. We attempt to use this architecture to get more useful features from source images in the encoding process, and two fusion layers (fusion strategies) are designed to fuse these features. Finally, the fused image is reconstructed by a decoder. Compared with existing fusion methods, the proposed fusion method achieves the state-of-the-art performance in objective and subjective assessment.
Image decomposition is crucial for many image processing tasks, as it allows to extract salient features from source images. A good image decomposition method could lead to a better performance, ...especially in image fusion tasks. We propose a multi-level image decomposition method based on latent low-rank representation(LatLRR), which is called MDLatLRR. This decomposition method is applicable to many image processing fields. In this paper, we focus on the image fusion task. We build a novel image fusion framework based on MDLatLRR which is used to decompose source images into detail parts(salient features) and base parts. A nuclear-norm based fusion strategy is used to fuse the detail parts and the base parts are fused by an averaging strategy. Compared with other state-of-the-art fusion methods, the proposed algorithm exhibits better fusion performance in both subjective and objective evaluation.
In this article, we propose a novel method for infrared and visible image fusion where we develop nest connection-based network and spatial/channel attention models. The nest connection-based network ...can preserve significant amounts of information from input data in a multiscale perspective. The approach comprises three key elements: encoder, fusion strategy, and decoder, respectively. In our proposed fusion strategy, spatial attention models and channel attention models are developed that describe the importance of each spatial position and of each channel with deep features. First, the source images are fed into the encoder to extract multiscale deep features. The novel fusion strategy is then developed to fuse these features for each scale. Finally, the fused image is reconstructed by the nest connection-based decoder. Experiments are performed on publicly available data sets. These exhibit that our proposed approach has better fusion performance than other state-of-the-art methods. This claim is justified through both subjective and objective evaluations. The code of our fusion method is available at https://github.com/hli1221/imagefusion-nestfuse .
In the image fusion field, the design of deep learning-based fusion methods is far from routine. It is invariably fusion-task specific and requires a careful consideration. The most difficult part of ...the design is to choose an appropriate strategy to generate the fused image for a specific task in hand. Thus, devising learnable fusion strategy is a very challenging problem in the community of image fusion. To address this problem, a novel end-to-end fusion network architecture (RFN-Nest) is developed for infrared and visible image fusion. We propose a residual fusion network (RFN) which is based on a residual architecture to replace the traditional fusion approach. A novel detail-preserving loss function, and a feature enhancing loss function are proposed to train RFN. The fusion model learning is accomplished by a novel two-stage training strategy. In the first stage, we train an auto-encoder based on an innovative nest connection (Nest) concept. Next, the RFN is trained using the proposed loss functions. The experimental results on public domain data sets show that, compared with the existing methods, our end-to-end fusion network delivers a better performance than the state-of-the-art methods in both subjective and objective evaluation. The code of our fusion method is available at https://github.com/hli1221/imagefusion-rfn-nest.
•Residual fusion network (RFN) is proposed to supersede handcrafted fusion strategies.•Two-stage training strategy is developed to train RFN.•Detail preservation and feature enhancement loss functions are designed.•The proposed method achieves better performance compare with existing methods.
With efficient appearance learning models, discriminative correlation filter (DCF) has been proven to be very successful in recent video object tracking benchmarks and competitions. However, the ...existing DCF paradigm suffers from two major issues, i.e., spatial boundary effect and temporal filter degradation. To mitigate these challenges, we propose a new DCF-based tracking method. The key innovations of the proposed method include adaptive spatial feature selection and temporal consistent constraints, with which the new tracker enables joint spatial-temporal filter learning in a lower dimensional discriminative manifold. More specifically, we apply structured spatial sparsity constraints to multi-channel filters. Consequently, the process of learning spatial filters can be approximated by the lasso regularization. To encourage temporal consistency, the filter model is restricted to lie around its historical value and updated locally to preserve the global structure in the manifold. Last, a unified optimization framework is proposed to jointly select temporal consistency preserving spatial features and learn discriminative filters with the augmented Lagrangian method. Qualitative and quantitative evaluations have been conducted on a number of well-known benchmarking datasets such as OTB2013, OTB50, OTB100, Temple-Colour, UAV123, and VOT2018. The experimental results demonstrate the superiority of the proposed method over the state-of-the-art approaches.
The clinical features and immune responses of asymptomatic individuals infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have not been well described. We studied 37 ...asymptomatic individuals in the Wanzhou District who were diagnosed with RT-PCR-confirmed SARS-CoV-2 infections but without any relevant clinical symptoms in the preceding 14 d and during hospitalization. Asymptomatic individuals were admitted to the government-designated Wanzhou People's Hospital for centralized isolation in accordance with policy
. The median duration of viral shedding in the asymptomatic group was 19 d (interquartile range (IQR), 15-26 d). The asymptomatic group had a significantly longer duration of viral shedding than the symptomatic group (log-rank P = 0.028). The virus-specific IgG levels in the asymptomatic group (median S/CO, 3.4; IQR, 1.6-10.7) were significantly lower (P = 0.005) relative to the symptomatic group (median S/CO, 20.5; IQR, 5.8-38.2) in the acute phase. Of asymptomatic individuals, 93.3% (28/30) and 81.1% (30/37) had reduction in IgG and neutralizing antibody levels, respectively, during the early convalescent phase, as compared to 96.8% (30/31) and 62.2% (23/37) of symptomatic patients. Forty percent of asymptomatic individuals became seronegative and 12.9% of the symptomatic group became negative for IgG in the early convalescent phase. In addition, asymptomatic individuals exhibited lower levels of 18 pro- and anti-inflammatory cytokines. These data suggest that asymptomatic individuals had a weaker immune response to SARS-CoV-2 infection. The reduction in IgG and neutralizing antibody levels in the early convalescent phase might have implications for immunity strategy and serological surveys.
CuO as a catalyst has shown promising application prospects in photocatalytic splitting of water into hydrogen (H2). However, the instability of CuO in amine aqueous solution limits the applications ...of CuO‐based photocatalysts in the photocatalytic H2 evolution. In this work, a novel dodecahedral nitrogen (N)‐doped carbon (C) coated CuO‐In2O3 p–n heterojunction (DNCPH) is designed and synthesized by directly pyrolyzing benzimidazole‐modified dodecahedral Cu/In‐based metal‐organic frameworks, showing long‐term stability in triethanolamine (TEOA) aqueous solution and excellent photocatalytic H2 production efficiency. The improved stability of DNCPH in TEOA solution is ascribed to the alleviation of electron deficiency in CuO by forming the p–n heterojunction and the protection with coated N‐doped C layer. Based on detailed theoretical calculations and experimental studies, it is found that the improved separation efficiency of photogenerated electron/hole pairs and the mediated adsorption behavior (|∆GH*|→0) by coupling N‐doped C layer with CuO‐In2O3 p–n heterojunction lead to the excellent photocatalytic H2 production efficiency of DNCPH. This work provides a feasible strategy for effectively applying CuO‐based photocatalysts in photocatalytic H2 production.
A novel nitrogen‐doped carbon‐coated CuO/In2O3 p–n heterojunction with long‐term stability in triethanolamine aqueous solution and excellent photocatalytic hydrogen production efficiency is fabricated by directly pyrolyzing benzimidazole‐modified dodecahedral Cu/In‐based metal‐organic frameworks.
In this paper, a new integral transform operator, which is similar to Fourier transform, is proposed for the first time. As a testing example, an application to the one-dimensional heat-diffusion ...problem is discussed. The result demonstrates accuracy and efficiency of the present technology to find the analytical solution for the heat-transfer problem.