Infrared and Visible Image Fusion Based on Res2Net-Transformer Automatic Encoding and Decoding

A novel image fusion network framework with an autonomous encoder and decoder is suggested to increase thevisual impression of fused images by improving the quality of infrared and visible light picture fusion. The networkcomprises an encoder module, fusion layer, decoder module, and edge improvementmodule. The encoder moduleutilizes an enhanced Inception module for shallow feature extraction, then combines Res2Net and Transformerto achieve deep-level co-extraction of local and global features from the original picture. An edge enhancementmodule (EEM) is created to extract significant edge features. A modal maximum difference fusion strategy isintroduced to enhance the adaptive representation of information in various regions of the source image, therebyenhancing the contrast of the fused image. The encoder and the EEM module extract features, which are thencombined in the fusion layer to create a fused picture using the decoder. Three datasets were chosen to test thealgorithmproposed in this paper. The results of the experiments demonstrate that the network effectively preservesbackground and detail information in both infrared and visible images, yielding superior outcomes in subjectiveand objective evaluations.

CAEFusion: A New Convolutional Autoencoder-Based Infrared and Visible Light Image Fusion Algorithm

To address the issues of incomplete information,blurred details,loss of details,and insufficient contrast in infrared and visible image fusion,an image fusion algorithm based on a convolutional autoencoder is proposed.The region attention module is meant to extract the background feature map based on the distinct properties of the background feature map and the detail feature map.A multi-scale convolution attention module is suggested to enhance the communication of feature information.At the same time,the feature transformation module is introduced to learn more robust feature representations,aiming to preserve the integrity of image information.This study uses three available datasets from TNO,FLIR,and NIR to perform thorough quantitative and qualitative trials with five additional algorithms.The methods are assessed based on four indicators:information entropy(EN),standard deviation(SD),spatial frequency(SF),and average gradient(AG).Object detection experiments were done on the M3FD dataset to further verify the algorithm’s performance in comparison with five other algorithms.The algorithm’s accuracy was evaluated using the mean average precision at a threshold of 0.5(mAP@0.5)index.Comprehensive experimental findings show that CAEFusion performs well in subjective visual and objective evaluation criteria and has promising potential in downstream object detection tasks.

Mangrove monitoring and extraction based on multi-source remote sensing data:a deep learning method based on SAR and optical image fusion

Mangroves are indispensable to coastlines,maintaining biodiversity,and mitigating climate change.Therefore,improving the accuracy of mangrove information identification is crucial for their ecological protection.Aiming at the limited morphological information of synthetic aperture radar(SAR)images,which is greatly interfered by noise,and the susceptibility of optical images to weather and lighting conditions,this paper proposes a pixel-level weighted fusion method for SAR and optical images.Image fusion enhanced the target features and made mangrove monitoring more comprehensive and accurate.To address the problem of high similarity between mangrove forests and other forests,this paper is based on the U-Net convolutional neural network,and an attention mechanism is added in the feature extraction stage to make the model pay more attention to the mangrove vegetation area in the image.In order to accelerate the convergence and normalize the input,batch normalization(BN)layer and Dropout layer are added after each convolutional layer.Since mangroves are a minority class in the image,an improved cross-entropy loss function is introduced in this paper to improve the model’s ability to recognize mangroves.The AttU-Net model for mangrove recognition in high similarity environments is thus constructed based on the fused images.Through comparison experiments,the overall accuracy of the improved U-Net model trained from the fused images to recognize the predicted regions is significantly improved.Based on the fused images,the recognition results of the AttU-Net model proposed in this paper are compared with its benchmark model,U-Net,and the Dense-Net,Res-Net,and Seg-Net methods.The AttU-Net model captured mangroves’complex structures and textural features in images more effectively.The average OA,F1-score,and Kappa coefficient in the four tested regions were 94.406%,90.006%,and 84.045%,which were significantly higher than several other methods.This method can provide some technical support for the monitoring and protection of mangrove ecosystems.

A Novel Multi-Stream Fusion Network for Underwater Image Enhancement

Due to the selective absorption of light and the existence of a large number of floating media in sea water, underwater images often suffer from color casts and detail blurs. It is therefore necessary to perform color correction and detail restoration. However,the existing enhancement algorithms cannot achieve the desired results. In order to solve the above problems, this paper proposes a multi-stream feature fusion network. First, an underwater image is preprocessed to obtain potential information from the illumination stream, color stream and structure stream by histogram equalization with contrast limitation, gamma correction and white balance, respectively. Next, these three streams and the original raw stream are sent to the residual blocks to extract the features. The features will be subsequently fused. It can enhance feature representation in underwater images. In the meantime, a composite loss function including three terms is used to ensure the quality of the enhanced image from the three aspects of color balance, structure preservation and image smoothness. Therefore, the enhanced image is more in line with human visual perception.Finally, the effectiveness of the proposed method is verified by comparison experiments with many stateof-the-art underwater image enhancement algorithms. Experimental results show that the proposed method provides superior results over them in terms of MSE,PSNR, SSIM, UIQM and UCIQE, and the enhanced images are more similar to their ground truth images.

Advancements in Remote Sensing Image Dehazing: Introducing URA-Net with Multi-Scale Dense Feature Fusion Clusters and Gated Jump Connection

The degradation of optical remote sensing images due to atmospheric haze poses a significant obstacle,profoundly impeding their effective utilization across various domains.Dehazing methodologies have emerged as pivotal components of image preprocessing,fostering an improvement in the quality of remote sensing imagery.This enhancement renders remote sensing data more indispensable,thereby enhancing the accuracy of target iden-tification.Conventional defogging techniques based on simplistic atmospheric degradation models have proven inadequate for mitigating non-uniform haze within remotely sensed images.In response to this challenge,a novel UNet Residual Attention Network(URA-Net)is proposed.This paradigmatic approach materializes as an end-to-end convolutional neural network distinguished by its utilization of multi-scale dense feature fusion clusters and gated jump connections.The essence of our methodology lies in local feature fusion within dense residual clusters,enabling the extraction of pertinent features from both preceding and current local data,depending on contextual demands.The intelligently orchestrated gated structures facilitate the propagation of these features to the decoder,resulting in superior outcomes in haze removal.Empirical validation through a plethora of experiments substantiates the efficacy of URA-Net,demonstrating its superior performance compared to existing methods when applied to established datasets for remote sensing image defogging.On the RICE-1 dataset,URA-Net achieves a Peak Signal-to-Noise Ratio(PSNR)of 29.07 dB,surpassing the Dark Channel Prior(DCP)by 11.17 dB,the All-in-One Network for Dehazing(AOD)by 7.82 dB,the Optimal Transmission Map and Adaptive Atmospheric Light For Dehazing(OTM-AAL)by 5.37 dB,the Unsupervised Single Image Dehazing(USID)by 8.0 dB,and the Superpixel-based Remote Sensing Image Dehazing(SRD)by 8.5 dB.Particularly noteworthy,on the SateHaze1k dataset,URA-Net attains preeminence in overall performance,yielding defogged images characterized by consistent visual quality.This underscores the contribution of the research to the advancement of remote sensing technology,providing a robust and efficient solution for alleviating the adverse effects of haze on image quality.

Research on Multi-Scale Feature Fusion Network Algorithm Based on Brain Tumor Medical Image Classification

Gliomas have the highest mortality rate of all brain tumors.Correctly classifying the glioma risk period can help doctors make reasonable treatment plans and improve patients’survival rates.This paper proposes a hierarchical multi-scale attention feature fusion medical image classification network(HMAC-Net),which effectively combines global features and local features.The network framework consists of three parallel layers:The global feature extraction layer,the local feature extraction layer,and the multi-scale feature fusion layer.A linear sparse attention mechanism is designed in the global feature extraction layer to reduce information redundancy.In the local feature extraction layer,a bilateral local attention mechanism is introduced to improve the extraction of relevant information between adjacent slices.In the multi-scale feature fusion layer,a channel fusion block combining convolutional attention mechanism and residual inverse multi-layer perceptron is proposed to prevent gradient disappearance and network degradation and improve feature representation capability.The double-branch iterative multi-scale classification block is used to improve the classification performance.On the brain glioma risk grading dataset,the results of the ablation experiment and comparison experiment show that the proposed HMAC-Net has the best performance in both qualitative analysis of heat maps and quantitative analysis of evaluation indicators.On the dataset of skin cancer classification,the generalization experiment results show that the proposed HMAC-Net has a good generalization effect.

Multimodality Medical Image Fusion Based on Pixel Significance with Edge-Preserving Processing for Clinical Applications

Multimodal medical image fusion has attained immense popularity in recent years due to its robust technology for clinical diagnosis.It fuses multiple images into a single image to improve the quality of images by retaining significant information and aiding diagnostic practitioners in diagnosing and treating many diseases.However,recent image fusion techniques have encountered several challenges,including fusion artifacts,algorithm complexity,and high computing costs.To solve these problems,this study presents a novel medical image fusion strategy by combining the benefits of pixel significance with edge-preserving processing to achieve the best fusion performance.First,the method employs a cross-bilateral filter(CBF)that utilizes one image to determine the kernel and the other for filtering,and vice versa,by considering both geometric closeness and the gray-level similarities of neighboring pixels of the images without smoothing edges.The outputs of CBF are then subtracted from the original images to obtain detailed images.It further proposes to use edge-preserving processing that combines linear lowpass filtering with a non-linear technique that enables the selection of relevant regions in detailed images while maintaining structural properties.These regions are selected using morphologically processed linear filter residuals to identify the significant regions with high-amplitude edges and adequate size.The outputs of low-pass filtering are fused with meaningfully restored regions to reconstruct the original shape of the edges.In addition,weight computations are performed using these reconstructed images,and these weights are then fused with the original input images to produce a final fusion result by estimating the strength of horizontal and vertical details.Numerous standard quality evaluation metrics with complementary properties are used for comparison with existing,well-known algorithms objectively to validate the fusion results.Experimental results from the proposed research article exhibit superior performance compared to other competing techniques in the case of both qualitative and quantitative evaluation.In addition,the proposed method advocates less computational complexity and execution time while improving diagnostic computing accuracy.Nevertheless,due to the lower complexity of the fusion algorithm,the efficiency of fusion methods is high in practical applications.The results reveal that the proposed method exceeds the latest state-of-the-art methods in terms of providing detailed information,edge contour,and overall contrast.

PSMFNet:Lightweight Partial Separation and Multiscale Fusion Network for Image Super-Resolution

The employment of deep convolutional neural networks has recently contributed to significant progress in single image super-resolution(SISR)research.However,the high computational demands of most SR techniques hinder their applicability to edge devices,despite their satisfactory reconstruction performance.These methods commonly use standard convolutions,which increase the convolutional operation cost of the model.In this paper,a lightweight Partial Separation and Multiscale Fusion Network(PSMFNet)is proposed to alleviate this problem.Specifically,this paper introduces partial convolution(PConv),which reduces the redundant convolution operations throughout the model by separating some of the features of an image while retaining features useful for image reconstruction.Additionally,it is worth noting that the existing methods have not fully utilized the rich feature information,leading to information loss,which reduces the ability to learn feature representations.Inspired by self-attention,this paper develops a multiscale feature fusion block(MFFB),which can better utilize the non-local features of an image.MFFB can learn long-range dependencies from the spatial dimension and extract features from the channel dimension,thereby obtaining more comprehensive and rich feature information.As the role of the MFFB is to capture rich global features,this paper further introduces an efficient inverted residual block(EIRB)to supplement the local feature extraction ability of PSMFNet.A comprehensive analysis of the experimental results shows that PSMFNet maintains a better performance with fewer parameters than the state-of-the-art models.

A deep learning fusion model for accurate classification of brain tumours in Magnetic Resonance images

Detecting brain tumours is complex due to the natural variation in their location, shape, and intensity in images. While having accurate detection and segmentation of brain tumours would be beneficial, current methods still need to solve this problem despite the numerous available approaches. Precise analysis of Magnetic Resonance Imaging (MRI) is crucial for detecting, segmenting, and classifying brain tumours in medical diagnostics. Magnetic Resonance Imaging is a vital component in medical diagnosis, and it requires precise, efficient, careful, efficient, and reliable image analysis techniques. The authors developed a Deep Learning (DL) fusion model to classify brain tumours reliably. Deep Learning models require large amounts of training data to achieve good results, so the researchers utilised data augmentation techniques to increase the dataset size for training models. VGG16, ResNet50, and convolutional deep belief networks networks extracted deep features from MRI images. Softmax was used as the classifier, and the training set was supplemented with intentionally created MRI images of brain tumours in addition to the genuine ones. The features of two DL models were combined in the proposed model to generate a fusion model, which significantly increased classification accuracy. An openly accessible dataset from the internet was used to test the model's performance, and the experimental results showed that the proposed fusion model achieved a classification accuracy of 98.98%. Finally, the results were compared with existing methods, and the proposed model outperformed them significantly.

Image Fusion Using Wavelet Transformation and XGboost Algorithm

Recently,there have been several uses for digital image processing.Image fusion has become a prominent application in the domain of imaging processing.To create one final image that provesmore informative and helpful compared to the original input images,image fusion merges two or more initial images of the same item.Image fusion aims to produce,enhance,and transform significant elements of the source images into combined images for the sake of human visual perception.Image fusion is commonly employed for feature extraction in smart robots,clinical imaging,audiovisual camera integration,manufacturing process monitoring,electronic circuit design,advanced device diagnostics,and intelligent assembly line robots,with image quality varying depending on application.The research paper presents various methods for merging images in spatial and frequency domains,including a blend of stable and curvelet transformations,everageMax-Min,weighted principal component analysis(PCA),HIS(Hue,Intensity,Saturation),wavelet transform,discrete cosine transform(DCT),dual-tree Complex Wavelet Transform(CWT),and multiple wavelet transform.Image fusion methods integrate data from several source images of an identical target,thereby enhancing information in an extremely efficient manner.More precisely,in imaging techniques,the depth of field constraint precludes images from focusing on every object,leading to the exclusion of certain characteristics.To tackle thess challanges,a very efficient multi-focus wavelet decomposition and recompositionmethod is proposed.The use of these wavelet decomposition and recomposition techniques enables this method to make use of existing optimized wavelet code and filter choice.The simulated outcomes provide evidence that the suggested approach initially extracts particular characteristics from images in order to accurately reflect the level of clarity portrayed in the original images.This study enhances the performance of the eXtreme Gradient Boosting(XGBoost)algorithm in detecting brain malignancies with greater precision through the integration of computational image analysis and feature selection.The performance of images is improved by segmenting them employing the K-Means algorithm.The segmentation method aids in identifying specific regions of interest,using Particle Swarm Optimization(PCA)for trait selection and XGBoost for data classification.Extensive trials confirm the model’s exceptional visual performance,achieving an accuracy of up to 97.067%and providing good objective indicators.

DCFNet:An Effective Dual-Branch Cross-Attention Fusion Network for Medical Image Segmentation

Automatic segmentation of medical images provides a reliable scientific basis for disease diagnosis and analysis.Notably,most existing methods that combine the strengths of convolutional neural networks(CNNs)and Transformers have made significant progress.However,there are some limitations in the current integration of CNN and Transformer technology in two key aspects.Firstly,most methods either overlook or fail to fully incorporate the complementary nature between local and global features.Secondly,the significance of integrating the multiscale encoder features from the dual-branch network to enhance the decoding features is often disregarded in methods that combine CNN and Transformer.To address this issue,we present a groundbreaking dual-branch cross-attention fusion network(DCFNet),which efficiently combines the power of Swin Transformer and CNN to generate complementary global and local features.We then designed the Feature Cross-Fusion(FCF)module to efficiently fuse local and global features.In the FCF,the utilization of the Channel-wise Cross-fusion Transformer(CCT)serves the purpose of aggregatingmulti-scale features,and the Feature FusionModule(FFM)is employed to effectively aggregate dual-branch prominent feature regions from the spatial perspective.Furthermore,within the decoding phase of the dual-branch network,our proposed Channel Attention Block(CAB)aims to emphasize the significance of the channel features between the up-sampled features and the features generated by the FCFmodule to enhance the details of the decoding.Experimental results demonstrate that DCFNet exhibits enhanced accuracy in segmentation performance.Compared to other state-of-the-art(SOTA)methods,our segmentation framework exhibits a superior level of competitiveness.DCFNet’s accurate segmentation of medical images can greatly assist medical professionals in making crucial diagnoses of lesion areas in advance.

A Lightweight Convolutional Neural Network with Hierarchical Multi-Scale Feature Fusion for Image Classification

Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.

Multi-Modal Medical Image Fusion Based on Improved Parameter Adaptive PCNN and Latent Low-Rank Representation

Multimodal medical image fusion can help physicians provide more accurate treatment plans for patients, as unimodal images provide limited valid information. To address the insufficient ability of traditional medical image fusion solutions to protect image details and significant information, a new multimodality medical image fusion method(NSST-PAPCNNLatLRR) is proposed in this paper. Firstly, the high and low-frequency sub-band coefficients are obtained by decomposing the source image using NSST. Then, the latent low-rank representation algorithm is used to process the low-frequency sub-band coefficients;An improved PAPCNN algorithm is also proposed for the fusion of high-frequency sub-band coefficients. The improved PAPCNN model was based on the automatic setting of the parameters, and the optimal method was configured for the time decay factor αe. The experimental results show that, in comparison with the five mainstream fusion algorithms, the new algorithm has significantly improved the visual effect over the comparison algorithm,enhanced the ability to characterize important information in images, and further improved the ability to protect the detailed information;the new algorithm has achieved at least four firsts in six objective indexes.

Infrared polarization image fusion based on combination of NSST and improved PCA

In view of the problem that current mainstream fusion method of infrared polarization image—Multiscale Geometry Analysis method only focuses on a certain characteristic to image representation.And spatial domain fusion method,Principal Component Analysis(PCA)method has the shortcoming of losing small target,this paper presents a new fusion method of infrared polarization images based on combination of Nonsubsampled Shearlet Transformation(NSST)and improved PCA.This method can make full use of the effectiveness to image details expressed by NSST and the characteristics that PCA can highlight the main features of images.The combination of the two methods can integrate the complementary features of themselves to retain features of targets and image details fully.Firstly,intensity and polarization images are decomposed into low frequency and high frequency components with different directions by NSST.Secondly,the low frequency components are fused with improved PCA,while the high frequency components are fused by joint decision making rule with local energy and local variance.Finally,the fused image is reconstructed with the inverse NSST to obtain the final fused image of infrared polarization.The experiment results show that the method proposed has higher advantages than other methods in terms of detail preservation and visual effect.

Application of Image Fusion Methods to Cell Imaging Processing

Aim To fuse the fluorescence image and transmission image of a cell into a single image containing more information than any of the individual image. Methods Image fusion technology was applied to biological cell imaging processing. It could match the images and improve the confidence and spatial resolution of the images. Using two algorithms, double thresholds algorithm and denoising algorithm based on wavelet transform,the fluorescence image and transmission image of a Cell were merged into a composite image. Results and Conclusion The position of fluorescence and the structure of cell can be displyed in the composite image. The signal-to-noise ratio of the exultant image is improved to a large extent. The algorithms are not only useful to investigate the fluorescence and transmission images, but also suitable to observing two or more fluoascent label proes in a single cell.

Characterization of neurovascular compression in facial neuralgia patients by 3D high-resolution MRI and image fusion technique

Objective:To describe the anatomical characteristics and patterns of neurovascular compression (NVC) in patients suffering trigeminal neuralgia(TN) by 3D high-resolution magnetic resonance imaging(MRI) method and image fusion technique.Methods:The anatomic structure of trigeminal nerve,brain stem and blood vessel was observed in 100 consecutive TN patients by 3D high resolution MRI(3D SPGR,contrast-enhanced T1 3D MP-RAGE and T2/T1 3D FIESTA). The 3D image sources were fused and visualized using 3D DOCTOR software.Results:One or several NVC sites,which usually appeared 0-9.8 mm away from brain stem,were found on the symptomatic side in 93%of the TN cases.Superior cerebellar artery was involved in 76%(71/93) of these cases.The other vessels including antero-inferior cerebellar artery,vertebral artery, basilar artery and veins also contributed to the occurrence of NVC.The NVC sites were found to be located in the proximal segment in 42%of these cases(39/93) and in the distal segment in 45% (42/93).Nerve dislocation or distortion was observed in 32%(30/93).Conclusions:Various 3D high resolution MRI methods combined with the image fusion technique could provide pathologic anatomic information for the diagnosis and treatment of TN.

Assessment of radiofrequency ablation margin by MRI-MRI image fusion in hepatocellular carcinoma

AIM:To investigate the feasibility and clinical value of magnetic resonance imaging(MRI)-MRI image fusion in assessing the ablative margin(AM) for hepatocellular carcinoma(HCC).METHODS:A newly developed ultrasound workstation for MRI-MRI image fusion was used to evaluate the AM of 62 tumors in 52 HCC patients after radiofrequency ablation(RFA).The lesions were divided into two groups:group A,in which the tumor was completely ablated and 5 mm AM was achieved(n = 32);and group B,in which the tumor was completely ablated but 5 mm AM was not achieved(n = 29).To detect local tumor progression(LTP),all patients were followed every two months by contrast-enhanced ultrasound,contrast-enhanced MRI or computed tomography(CT) in the first year after RFA.Then,the follow-up interval was prolonged to every three months after the first year.RESULTS:Of the 62 tumors,MRI-MRI image fusion was successful in 61(98.4%);the remaining case had significant deformation of the liver and massive ascites after RFA.The time required for creating image fusion and AM evaluation was 15.5 ± 5.5 min(range:8-22 min) and 9.6 ± 3.2 min(range:6-14 min),respectively.The follow-up period ranged from 1-23 mo(14.2 ± 5.4 mo).In group A,no LTP was detected in 32 lesions,whereas in group B,LTP was detected in 4 of 29 tumors,which occurred at 2,7,9,and 15 mo after RFA.The frequency of LTP in group B(13.8%;4/29) was significantly higher than that in group A(0/32,P = 0.046).All of the LTPs occurred in the area in which the 5 mm AM was not achieved.CONCLUSION:The MRI-MRI image fusion using an ultrasound workstation is feasible and useful for evaluating the AM after RFA for HCC.

Contourlet transform for image fusion using cycle spinning

A new method for image fusion based on Contourlet transform and cycle spinning is proposed. Contourlet transform is a flexible multiresolution, local and directional image expansion, also provids a sparse representation for two-dimensional piece-wise smooth signals resembling images. Due to lack of translation invariance property in Contourlet transform, the conventional image fusion algorithm based on Contourlet transform introduces many artifacts. According to the theory of cycle spinning applied to image denoising, an invariance transform can reduce the artifacts through a series of processing efficiently. So the technology of cycle spinning is introduced to develop the translation invariant Contourlet fusion algorithm. This method can effectively eliminate the Gibbs-like phenomenon, extract the characteristics of original images, and preserve more important information. Experimental results show the simplicity and effectiveness of the method and its advantages over the conventional approaches.

Multi-source image fusion algorithm based on fast weighted guided filter

In last few years,guided image fusion algorithms become more and more popular.However,the current algorithms cannot solve the halo artifacts.We propose an image fusion algorithm based on fast weighted guided filter.Firstly,the source images are separated into a series of high and low frequency components.Secondly,three visual features of the source image are extracted to construct a decision graph model.Thirdly,a fast weighted guided filter is raised to optimize the result obtained in the previous step and reduce the time complexity by considering the correlation among neighboring pixels.Finally,the image obtained in the previous step is combined with the weight map to realize the image fusion.The proposed algorithm is applied to multi-focus,visible-infrared and multi-modal image respectively and the final results show that the algorithm effectively solves the halo artifacts of the merged images with higher efficiency,and is better than the traditional method considering subjective visual consequent and objective evaluation.

SwinFusion: Cross-domain Long-range Learning for General Image Fusion via Swin Transformer

This study proposes a novel general image fusion framework based on cross-domain long-range learning and Swin Transformer,termed as SwinFusion.On the one hand,an attention-guided cross-domain module is devised to achieve sufficient integration of complementary information and global interaction.More specifically,the proposed method involves an intra-domain fusion unit based on self-attention and an interdomain fusion unit based on cross-attention,which mine and integrate long dependencies within the same domain and across domains.Through long-range dependency modeling,the network is able to fully implement domain-specific information extraction and cross-domain complementary information integration as well as maintaining the appropriate apparent intensity from a global perspective.In particular,we introduce the shifted windows mechanism into the self-attention and cross-attention,which allows our model to receive images with arbitrary sizes.On the other hand,the multi-scene image fusion problems are generalized to a unified framework with structure maintenance,detail preservation,and proper intensity control.Moreover,an elaborate loss function,consisting of SSIM loss,texture loss,and intensity loss,drives the network to preserve abundant texture details and structural information,as well as presenting optimal apparent intensity.Extensive experiments on both multi-modal image fusion and digital photography image fusion demonstrate the superiority of our SwinFusion compared to the state-of-theart unified image fusion algorithms and task-specific alternatives.Implementation code and pre-trained weights can be accessed at https://github.com/Linfeng-Tang/SwinFusion.