Search

In recent years, deep learning models have dramatically influenced image inpainting. However, many existing studies still suffer from over‐smoothed or blurred textures when missing regions are large or contain rich visual details. To restore textures at a fine‐grained level, a multi‐stage inpainting approach is proposed, which applies a series of partial inpainting modules as well as a progressive inpainting module to inpaint missing areas from their boundaries to the centre successively. Some improvements are made on the partial convolutions to reduce artifacts like blurriness, which require a convolution kernel to contain known pixels more than a certain proportion. Towards photorealistic inpainting results, the intermediate outputs from each stage are used to compute the loss. Finally, to facilitate the training process, a multi‐step training is designed that progressively adds inpainting modules to optimize the model. Experiments show that this method outperforms the current excellent techniques on the publicly available datasets: CelebA, Places2 and Paris StreetView. [ABSTRACT FROM AUTHOR]

The article proposes a real‐time apple picking method based on an improved YOLOv5. This method accurately recognizes different apple targets on fruit trees for robots and helps them adjust their position to avoid obstructions during fruit picking. Firstly, the original BottleneckCSP module in the YOLOv5 backbone network is enhanced to extract deeper features from images while maintaining lightweight. Secondly, the ECA module is embedded into the improved backbone network to better extract features of different apple targets. Lastly, the initial anchor frame size of the network is adjusted to avoid recognizing apples in distant planting rows. The results demonstrate that this improved model achieves high accuracy rates and recall rates for recognizing various types of apple picking methods with an average recognition time of 0.025s per image. Compared with other models tested on six types of apple picking methods, including the original YOLOv5 model as well as YOLOv3 and EfficientDet‐D0 algorithms, our improved model shows significant improvements in mAP by 1.95%, 17.6%, and 12.7% respectively. This method provides technical support for robots' picking hands to actively avoid obstructions caused by branches during fruit harvesting, effectively reducing apple loss. [ABSTRACT FROM AUTHOR]

Missing pixel imputation presents a critical challenge in image processing and computer vision, particularly in applications such as image restoration and inpainting. The primary objective of this paper is to accurately estimate and reconstruct missing pixel values to restore complete visual information. This paper introduces a novel model called the Enhanced Connected Pixel Identity GAN with Neutrosophic (ECP-IGANN), which is designed to address two fundamental issues inherent in existing GAN architectures for missing pixel generation: (1) mode collapse, which leads to a lack of diversity in generated pixels, and (2) the preservation of pixel integrity within the reconstructed images. ECP-IGANN incorporates two key innovations to improve missing pixel imputation. First, an identity block is integrated into the generation process to facilitate the retention of existing pixel values and ensure consistency. Second, the model calculates the values of the 8-connected neighbouring pixels around each missing pixel, thereby enhancing the coherence and integrity of the imputed pixels. The efficacy of ECP-IGANN was rigorously evaluated through extensive experimentation across five diverse datasets: BigGAN-ImageNet, the 2024 Medical Imaging Challenge Dataset, the Autonomous Vehicles Dataset, the 2024 Satellite Imagery Dataset, and the Fashion and Apparel Dataset 2024. These experiments assessed the model's performance in terms of diversity, pixel imputation accuracy, and mode collapse mitigation, with results demonstrating significant improvements in the Inception Score (IS) and Fréchet Inception Distance (FID). ECP-IGANN markedly enhanced image segmentation performance in the validation phase across all datasets. Key metrics, such as Dice Score, Accuracy, Precision, and Recall, were improved substantially for various segmentation models, including Spatial Attention U-Net, Dense U-Net, and Residual Attention U-Net. For example, in the 2024 Medical Imaging Challenge Dataset, the Residual Attention U-Net's Dice Score increased from 0.84 to 0.90, while accuracy improved from 0.88 to 0.93 following the application of ECP-IGANN. Similar performance enhancements were observed with the other datasets, highlighting the model's robust generalizability across diverse imaging domains. [ABSTRACT FROM AUTHOR]

In the field of remote sensing image processing, pansharpening technology stands as a critical advancement. This technology aims to enhance multispectral images that possess low resolution by integrating them with high-spatial-resolution panchromatic images, ultimately producing multispectral images with high resolution that are abundant in both spatial and spectral details. Thus, there remains potential for improving the quality of both the spectral and spatial domains of the fused images based on deep-learning-based pansharpening methods. This work proposes a new method for the task of pansharpening: the Multi-Stage Progressive Pansharpening Network with Detail Injection with Redundancy Reduction Mechanism (MSPPN-DIRRM). This network is divided into three levels, each of which is optimized for the extraction of spectral and spatial data at different scales. Particular spectral feature and spatial detail extraction modules are used at each stage. Moreover, a new image reconstruction module named the DRRM is introduced in this work; it eliminates both spatial and channel redundancy and improves the fusion quality. The effectiveness of the proposed model is further supported by experimental results using both simulated data and real data from the QuickBird, GaoFen1, and WorldView2 satellites; these results show that the proposed model outperforms deep-learning-based methods in both visual and quantitative assessments. Among various evaluation metrics, performance improves by 0.92–18.7% compared to the latest methods. [ABSTRACT FROM AUTHOR]

Imaging through scattering media is an important and challenging problem, and the technology has been used in many fields, such as autonomous driving, industrial inspections, remote sensing imaging, and biomedical imaging. However, most of the previous experiments used numbers or letters for close-range imaging, while objects in life are colorful. In this study, a new deep learning network, DesUNet, was constructed to image realistic objects at medium and long distances under sunlight through scattering media, and to realize object recognition. In addition, this study also compares the imaging results of different neural networks, and the results show that the DesUNet network improves the feature information storage ability and enhances the image reconstruction. It not only clearly restores the original appearance of the object, but also extracts the physical information about the object. In order to further verify the power of the DesUNet network, this study also conducted indoor near distance and outdoor medium distance imaging experiments. For indoor reconstructed objects, the appearance of the objects could be clearly identified. For outdoor reconstructed objects, the confidence level could reach above 0.9 through YOLO. The experiments show that the DesUNet network has good robustness and generalization. [ABSTRACT FROM AUTHOR]

Abstract In recent years, deep learning models have dramatically influenced image inpainting. However, many existing studies still suffer from over‐smoothed or blurred textures when missing regions are large or contain rich visual details. To restore textures at a fine‐grained level, a multi‐stage inpainting approach is proposed, which applies a series of partial inpainting modules as well as a progressive inpainting module to inpaint missing areas from their boundaries to the centre successively. Some improvements are made on the partial convolutions to reduce artifacts like blurriness, which require a convolution kernel to contain known pixels more than a certain proportion. Towards photorealistic inpainting results, the intermediate outputs from each stage are used to compute the loss. Finally, to facilitate the training process, a multi‐step training is designed that progressively adds inpainting modules to optimize the model. Experiments show that this method outperforms the current excellent techniques on the publicly available datasets: CelebA, Places2 and Paris StreetView.

Abstract Medical image colouring techniques enable to colourize grey‐scale medical images for assisting doctors in diagnosis. Benefiting from the non‐linear fitting ability of deep neural network, deep medical image colouring techniques have achieved remarkable results. However, existing methods are still facing content and structure feature leakage, unrealistic colouring and poor scale invariability. Thus, this paper, proposes a Transformer‐based medical image colouring algorithm with long‐term dependency to avoid feature leakage of coloured images. To be specific, this method employs two different Transformer encoders to generate and encode feature sequences for grey‐scale medical images and real human colour slice images, respectively. Then, a novel multi‐layer Transformer decoder is used to stylize grey‐scale map image features based on the real physical colour feature sequences. For colouring images at different scales, we implement content‐ aware positional encoding with scale invariance and propose style‐aware positional encoding strategy to take realistic and physical colour prior into account. Extensive experimental results indicate our method has achieved better colourization effects than recent state‐of‐the‐art medical image colourization methods.

Abstract Synthetic images facilitate stereo matching. However, synthetic images may suffer from image distortion, domain bias, and stereo mismatch, which would significantly restrict the widespread use of stereo matching models in the real world. The first goal in this paper is to synthesize real‐looking images for minimizing the domain bias between the synthesized and real images. For this purpose, sharpened disparity maps are produced from a mono real image. Then, stereo image pairs are synthesized using these imperfect disparity maps and the single real image in the proposed pipeline. Although the synthesized images are as realistic as possible, the domain styles of the synthesized images are always very different from the real images. Thus, the second goal is to enhance the domain generalization ability of the stereo matching network. For that, the feature extraction layer is replaced with a teacher–student model. Then, a constraint of binocular contrast features is imposed on the output of the model. When tested on the KITTI, ETH3D, and Middlebury datasets, the accuracy of the method outperforms traditional methods by at least 30%. Experiments demonstrate that the approaches are general and can be conveniently embedded into existing stereo networks.

Abstract Super-resolution imaging has, for more than seventy years, gradually evolved to produce advanced methods for enhancing the resolution of images beyond the diffraction limits. Notwithstanding its foreseeable practical capabilities, we noted that this technology has received undeserved attention. The present work provides an extensive review of super-resolution imaging since its first conception in 1952, contextualizing studies into four research directions: reviews, advances, applications, and hardware implementations. We have discussed achievements, challenges, and potential opportunities of super-resolution imaging to equip researchers, especially those in their early careers, with knowledge to further advance the technology. Our work may open interesting research avenues that may accelerate realization of the technology into commercial imaging devices.

Abstract In recent years, many methods for image super‐resolution (SR) have relied on pairs of low‐resolution (LR) and high‐resolution (HR) images for training, where the degradation process is predefined by bicubic downsampling. While such approaches perform well in standard benchmark tests, they often fail to accurately replicate the complexity of real‐world image degradation. To address this challenge, researchers have proposed the use of unpaired image training to implicitly model the degradation process. However, there is a significant domain gap between the real‐world LR and the synthetic LR images from HR, which severely degrades the SR performance. A novel unsupervised image‐blind super‐resolution method that exploits degradation feature‐based learning for real‐image super‐resolution reconstruction (RDFL) is proposed. Their approach learns the degradation process from HR to LR using a generative adversarial network (GAN) and constrains the data distribution of the synthetic LR with real degraded images. The authors then encode the degraded features into a Transformer‐based SR network for image super‐resolution reconstruction through degradation representation learning. Extensive experiments on both synthetic and real datasets demonstrate the effectiveness and superiority of the RDFL method, which achieves visually pleasing reconstruction results.

Medical image colouring techniques enable to colourize grey‐scale medical images for assisting doctors in diagnosis. Benefiting from the non‐linear fitting ability of deep neural network, deep medical image colouring techniques have achieved remarkable results. However, existing methods are still facing content and structure feature leakage, unrealistic colouring and poor scale invariability. Thus, this paper, proposes a Transformer‐based medical image colouring algorithm with long‐term dependency to avoid feature leakage of coloured images. To be specific, this method employs two different Transformer encoders to generate and encode feature sequences for grey‐scale medical images and real human colour slice images, respectively. Then, a novel multi‐layer Transformer decoder is used to stylize grey‐scale map image features based on the real physical colour feature sequences. For colouring images at different scales, we implement content‐ aware positional encoding with scale invariance and propose style‐aware positional encoding strategy to take realistic and physical colour prior into account. Extensive experimental results indicate our method has achieved better colourization effects than recent state‐of‐the‐art medical image colourization methods. [ABSTRACT FROM AUTHOR]

Synthetic images facilitate stereo matching. However, synthetic images may suffer from image distortion, domain bias, and stereo mismatch, which would significantly restrict the widespread use of stereo matching models in the real world. The first goal in this paper is to synthesize real‐looking images for minimizing the domain bias between the synthesized and real images. For this purpose, sharpened disparity maps are produced from a mono real image. Then, stereo image pairs are synthesized using these imperfect disparity maps and the single real image in the proposed pipeline. Although the synthesized images are as realistic as possible, the domain styles of the synthesized images are always very different from the real images. Thus, the second goal is to enhance the domain generalization ability of the stereo matching network. For that, the feature extraction layer is replaced with a teacher–student model. Then, a constraint of binocular contrast features is imposed on the output of the model. When tested on the KITTI, ETH3D, and Middlebury datasets, the accuracy of the method outperforms traditional methods by at least 30%. Experiments demonstrate that the approaches are general and can be conveniently embedded into existing stereo networks. [ABSTRACT FROM AUTHOR]

Atmospheric turbulence is a key factor contributing to data distortion in mid-to-long-range target observation tasks. Neural networks have become a powerful tool for dealing with such problems due to their strong ability to fit nonlinearities in the spatial domain. However, the degradation in data is not confined solely to the spatial domain but is also present in the frequency domain. In recent years, the academic community has come to recognize the significance of frequency domain information within neural networks. There remains a gap in research on how to combine dual-domain information to reconstruct high-quality images in the field of blind turbulence image restoration. Drawing upon the close association between spatial and frequency domain degradation information, we introduce a novel neural network architecture, termed Dual-Domain Removal Turbulence Network (DDRTNet), designed to improve the quality of reconstructed images. DDRTNet incorporates multiscale spatial and frequency domain attention mechanisms, combined with a dual-domain collaborative learning strategy, effectively integrating global and local information to achieve efficient restoration of atmospheric turbulence-degraded images. Experimental findings demonstrate significant advantages in performance for DDRTNet compared to existing methods, validating its effectiveness in the task of blind turbulence image restoration. [ABSTRACT FROM AUTHOR]

In the field of endoscopic imaging, challenges such as low resolution, complex textures, and blurred edges often degrade the quality of 3D reconstructed models. To address these issues, this study introduces an innovative endoscopic image super-resolution and 3D reconstruction technique named Omni-Directional Focus and Scale Resolution (OmDF-SR). This method integrates an Omnidirectional Self-Attention (OSA) mechanism, an Omnidirectional Scale Aggregation Group (OSAG), a Dual-stream Adaptive Focus Mechanism (DAFM), and a Dynamic Edge Adjustment Framework (DEAF) to enhance the accuracy and efficiency of super-resolution processing. Additionally, it employs Structure from Motion (SfM) and Multi-View Stereo (MVS) technologies to achieve high-precision medical 3D models. Experimental results indicate significant improvements in image processing with a PSNR of 38.2902 dB and an SSIM of 0.9746 at a magnification factor of ×2, and a PSNR of 32.1723 dB and an SSIM of 0.9489 at ×4. Furthermore, the method excels in reconstructing detailed 3D models, enhancing point cloud density, mesh quality, and texture mapping richness, thus providing substantial support for clinical diagnosis and surgical planning. [ABSTRACT FROM AUTHOR]

The quality of images captured digitally or transmitted over networks is distorted by noise during the process. The current methods of image restoration can be ineffective in dealing with intricate noise patterns or may be slow or imprecise. This paper fills this gap by presenting a new second-order continuous-time dynamical system for denoising of images in image restoration. The approach used in this work poses the problem as a convex quadratic program that can, thus, be solved for optimality. The existence and uniqueness of a global solution are theoretically demonstrated, and the condition for the global strong convergence of the system's trajectory is provided. The method presented in this paper is shown to be useful in a number of experiments on image restoration. As for the performance, it is higher than that of other known algorithms, with an average SNR equal to 34.78 and a Structural Similarity Index Measure (SSIM) of 0.959 for the reconstructed images. Such improvements demonstrate the effectiveness of the second-order dynamical system approach in actual image restoration applications. [ABSTRACT FROM AUTHOR]

Super-resolution imaging has, for more than seventy years, gradually evolved to produce advanced methods for enhancing the resolution of images beyond the diffraction limits. Notwithstanding its foreseeable practical capabilities, we noted that this technology has received undeserved attention. The present work provides an extensive review of super-resolution imaging since its first conception in 1952, contextualizing studies into four research directions: reviews, advances, applications, and hardware implementations. We have discussed achievements, challenges, and potential opportunities of super-resolution imaging to equip researchers, especially those in their early careers, with knowledge to further advance the technology. Our work may open interesting research avenues that may accelerate realization of the technology into commercial imaging devices. [ABSTRACT FROM AUTHOR]

The recovery of mutilated fingerprints plays an important role in improving the accuracy of fingerprint recognition and the speed of identity retrieval, so it is crucial to recover mutilated fingerprints efficiently and accurately. In this paper, we propose a method for the restoration of mutilated fingerprints based on the ridge texture and orientation field. First, the part to be restored is identified via the local quality of the fingerprint, and a mask image is generated. Second, a novel dual-stream fingerprint restoration network named IFSR is designed, which contains two branches, namely an orientation prediction branch guided by the fingerprint orientation field and a detail restoration branch guided by the high-quality fingerprint texture image, through which the damaged region of the mutilated fingerprint is restored. Finally, the method proposed in this paper is validated on a real dataset and an artificially damaged fingerprint dataset. The equal error rate (EER) achieved on the DB1, DB2, and DB4 datasets of FVC2002 is 0.10%, 0.12%, and 0.20%, respectively, while on the DB1, DB2, and DB4 datasets of FVC2004, the EER reaches 1.13%, 2.00%, and 0.27%, respectively. On the artificially corrupted fingerprint dataset, the restoration method achieves a peak signal-to-noise ratio (PSNR) of 16.6735. [ABSTRACT FROM AUTHOR]

Image denoising and super-resolution play vital roles in imaging systems, greatly reducing the preprocessing cost of many AI techniques for object detection, segmentation, and tracking. Various advancements have been accomplished in this field, but progress is still needed. In this paper, we have proposed a novel technique named the Enhanced Learning Enriched Features (ELEF) mechanism using a deep convolutional neural network, which makes significant improvements to existing techniques. ELEF consists of two major processes: (1) Denoising, which removes the noise from images; and (2) Super-resolution, which improves the clarity and details of images. Features are learned through deep CNN and not through traditional algorithms so that we can better refine and enhance images. To effectively capture features, the network architecture adopted Dual Attention Units (DUs), which align with the Multi-Scale Residual Block (MSRB) for robust feature extraction, working sidewise with the feature-matching Selective Kernel Extraction (SKF). In addition, resolution mismatching cases are processed in detail to produce high-quality images. The effectiveness of the ELEF model is highlighted by the performance metrics, achieving a Peak Signal-to-Noise Ratio (PSNR) of 42.99 and a Structural Similarity Index (SSIM) of 0.9889, which indicates the ability to carry out the desired high-quality image restoration and enhancement. [ABSTRACT FROM AUTHOR]

The discrepancy in data distribution between training and testing scenarios, as well as the inductive bias of convolutional neural networks towards image styles, reduces the model's generalization ability. Many unsupervised domain generalization methods based on feature decoupling suffer from an initial neglect of explicit decoupling of content and style features, resulting in content features that still contain considerable redundant information, thereby restricting improvements in generalization capability. To tackle this problem, this paper optimizes the learning process of domain‐invariant (content) features into an information compression issue, minimizing redundancy in content features. Furthermore, to enhance decoupled learning, this paper introduces innovative cross‐domain loss functions and image reconstruction modules that explicitly decouple and merge content and style across different domains. Extensive experiments demonstrate the method's significant enhancements over recent cutting‐edge approaches. [ABSTRACT FROM AUTHOR]

In recent years, with the increasing demand for high-quality images in various fields, more and more attention has been focused on noise removal techniques for image processing. The effective elimination of unwanted noise plays a crucial role in improving image quality. To meet this challenge, many noise removal methods have been proposed, among which the diffusion model has become one of the focuses of many researchers. In order to make the restored image closer to the real image and retain more features of the image, this paper proposes a DIR-SDE method with reference to the diffusion models of IR-SDE and IDM, which improve the feature retention of the image in the de-raining process, and then improve the realism of the image for the image de-raining task. In this study, IR-SDE was used as the base structure of the diffusion model, IR-SDE was improved, and DINO-ViT was combined to enhance the image features. During the diffusion process, the image features were extracted using DINO-ViT, and these features were fused with the original images to enhance the learning effect of the model. The model was also trained and validated with the Rain100H dataset. Compared with the IR-SDE method, it improved 0.003 in the SSIM, 0.003 in the LPIPS, and 1.23 in the FID. The experimental results show that the diffusion model proposed in this study can effectively improve the image restoration performance. [ABSTRACT FROM AUTHOR]

With the development of deep learning, the Super-Resolution (SR) reconstruction of microscopic images has improved significantly. However, the scarcity of microscopic images for training, the underutilization of hierarchical features in original Low-Resolution (LR) images, and the high-frequency noise unrelated with the image structure generated during the reconstruction process are still challenges in the Single Image Super-Resolution (SISR) field. Faced with these issues, we first collected sufficient microscopic images through Motic, a company engaged in the design and production of optical and digital microscopes, to establish a dataset. Secondly, we proposed a Residual Dense Attention Generative Adversarial Network (RDAGAN). The network comprises a generator, an image discriminator, and a feature discriminator. The generator includes a Residual Dense Block (RDB) and a Convolutional Block Attention Module (CBAM), focusing on extracting the hierarchical features of the original LR image. Simultaneously, the added feature discriminator enables the network to generate high-frequency features pertinent to the image's structure. Finally, we conducted experimental analysis and compared our model with six classic models. Compared with the best model, our model improved PSNR and SSIM by about 1.5 dB and 0.2, respectively. [ABSTRACT FROM AUTHOR]

In recent years, many methods for image super‐resolution (SR) have relied on pairs of low‐resolution (LR) and high‐resolution (HR) images for training, where the degradation process is predefined by bicubic downsampling. While such approaches perform well in standard benchmark tests, they often fail to accurately replicate the complexity of real‐world image degradation. To address this challenge, researchers have proposed the use of unpaired image training to implicitly model the degradation process. However, there is a significant domain gap between the real‐world LR and the synthetic LR images from HR, which severely degrades the SR performance. A novel unsupervised image‐blind super‐resolution method that exploits degradation feature‐based learning for real‐image super‐resolution reconstruction (RDFL) is proposed. Their approach learns the degradation process from HR to LR using a generative adversarial network (GAN) and constrains the data distribution of the synthetic LR with real degraded images. The authors then encode the degraded features into a Transformer‐based SR network for image super‐resolution reconstruction through degradation representation learning. Extensive experiments on both synthetic and real datasets demonstrate the effectiveness and superiority of the RDFL method, which achieves visually pleasing reconstruction results. [ABSTRACT FROM AUTHOR]

This document is a guest editorial from the journal IET Computer Vision, discussing the topic of advanced image restoration and enhancement. The editorial highlights the challenges faced in this field, such as the complexity of degradation models for real-world low-quality images and the difficulty of acquiring paired data. It also introduces a special issue of the journal that includes five accepted papers, which focus on video reconstruction and image super-resolution. The editorial concludes by providing brief summaries of each accepted paper. The guest editors of the special issue are also mentioned, along with their research interests and affiliations. [Extracted from the article]

Recent advancements in computer vision, especially deep learning models, have shown considerable promise in tasks related to plant image object detection. However, the efficiency of these deep learning models heavily relies on input image quality, with low-resolution images significantly hindering model performance. Therefore, reconstructing high-quality images through specific techniques will help extract features from plant images, thus improving model performance. In this study, we explored the value of super-resolution technology for improving object detection model performance on plant images. Firstly, we built a comprehensive dataset comprising 1030 high-resolution plant images, named the PlantSR dataset. Subsequently, we developed a super-resolution model using the PlantSR dataset and benchmarked it against several state-of-the-art models designed for general image super-resolution tasks. Our proposed model demonstrated superior performance on the PlantSR dataset, indicating its efficacy in enhancing the super-resolution of plant images. Furthermore, we explored the effect of super-resolution on two specific object detection tasks: apple counting and soybean seed counting. By incorporating super-resolution as a pre-processing step, we observed a significant reduction in mean absolute error. Specifically, with the YOLOv7 model employed for apple counting, the mean absolute error decreased from 13.085 to 5.71. Similarly, with the P2PNet-Soy model utilized for soybean seed counting, the mean absolute error decreased from 19.159 to 15.085. These findings underscore the substantial potential of super-resolution technology in improving the performance of object detection models for accurately detecting and counting specific plants from images. The source codes and associated datasets related to this study are available at Github. [ABSTRACT FROM AUTHOR]

Show-through phenomena have always been a challenging issue in color-document image processing, which is widely used in various fields such as finance, education, and administration. Existing methods for processing color-document images face challenges, including dealing with double-sided documents with show-through effects, accurately distinguishing between foreground and show-through parts, and addressing the issue of insufficient real image data for supervised training. To overcome these challenges, this paper proposes a self-supervised-learning-based method for removing show-through effects in color-document images. The proposed method utilizes a two-stage-structured show-through-removal network that incorporates a double-cycle consistency loss and a pseudo-similarity loss to effectively constrain the process of show-through removal. Moreover, we constructed two datasets consisting of different show-through mixing ratios and conducted extensive experiments to verify the effectiveness of the proposed method. Experimental results demonstrate that the proposed method achieves competitive performance compared to state-of-the-art methods and can effectively perform show-through removal without the need for paired datasets. Specifically, the proposed method achieves an average PSNR of 33.85 dB on our datasets, outperforming comparable methods by a margin of 0.89 dB. [ABSTRACT FROM AUTHOR]

Underwater images are often with biased colours and reduced contrast because of the absorption and scattering effects when light propagates in water. Such images with degradation cannot meet the needs of underwater operations. The main problem in classic underwater image restoration or enhancement methods is that they consume long calculation time, and often, the colour or contrast of the result images is still unsatisfied. Instead of using the complicated physical model of underwater imaging degradation, we propose a new method to deal with underwater images by imitating the colour constancy mechanism of human vision using double‐opponency. Firstly, the original image is converted to the LMS space. Then the signals are linearly combined, and Gaussian convolutions are performed to imitate the function of receptive fields (RFs). Next, two RFs with different sizes work together to constitute the double‐opponency response. Finally, the underwater light is estimated to correct the colours in the image. Further contrast stretching on the luminance is optional. Experiments show that the proposed method can obtain clarified underwater images with higher quality than before, and it spends significantly less time cost compared to other previously published typical methods. [ABSTRACT FROM AUTHOR]

Abstract Person image generation aims to generate images that maintain the original human appearance in different target poses. Recent works have revealed that the critical element in achieving this task is the alignment of appearance domain and pose domain. Previous alignment methods, such as appearance flow warping, correspondence learning and cross attention, often encounter challenges when it comes to producing fine texture details. These approaches suffer from limitations in accurately estimating appearance flows due to the lack of global receptive field. Alternatively, they can only perform cross‐domain alignment on high‐level feature maps with small spatial dimensions since the computational complexity increases quadratically with larger feature sizes. In this article, the significance of multi‐scale alignment, in both low‐level and high‐level domains, for ensuring reliable cross‐domain alignment of appearance and pose is demonstrated. To this end, a novel and effective method, named Multi‐scale Cross‐domain Alignment (MCA) is proposed. Firstly, MCA adopts global context aggregation transformer to model multi‐scale interaction between pose and appearance inputs, which employs pair‐wise window‐based cross attention. Furthermore, leveraging the integrated global source information for each target position, MCA applies flexible flow prediction head and point correlation to effectively conduct warping and fusing for final transformed person image generation. Our proposed MCA achieves superior performance on two popular datasets than other methods, which verifies the effectiveness of our approach.

Conventional photography can only provide a two-dimensional image of the scene, whereas emerging imaging modalities such as light field enable the representation of higher dimensional visual information by capturing light rays from different directions. Light fields provide immersive experiences, a sense of presence in the scene, and can enhance different vision tasks. Hence, research into light field processing methods has become increasingly popular. It does, however, come at the cost of higher data volume and computational complexity. With the growing deployment of machine-learning and deep architectures in image processing applications, a paradigm shift toward learning-based approaches has also been observed in the design of light field processing methods. Various learning-based approaches are developed to process the high volume of light field data efficiently for different vision tasks while improving performance. Taking into account the diversity of light field vision tasks and the deployed learning-based frameworks, it is necessary to survey the scattered learning-based works in the domain to gain insight into the current trends and challenges. This paper aims to review the existing learning-based solutions for light field imaging and to summarize the most promising frameworks. Moreover, evaluation methods and available light field datasets are highlighted. Lastly, the review concludes with a brief outlook for future research directions. [ABSTRACT FROM AUTHOR]

Background: X-ray computed tomography (CT) is a powerful tool for measuring plant root growth in soil. However, a rapid scan with larger pots, which is required for throughput-prioritized crop breeding, results in high noise levels, low resolution, and blurred root segments in the CT volumes. Moreover, while plant root segmentation is essential for root quantification, detailed conditional studies on segmenting noisy root segments are scarce. The present study aimed to investigate the effects of scanning time and deep learning-based restoration of image quality on semantic segmentation of blurry rice (Oryza sativa) root segments in CT volumes. Results: VoxResNet, a convolutional neural network-based voxel-wise residual network, was used as the segmentation model. The training efficiency of the model was compared using CT volumes obtained at scan times of 33, 66, 150, 300, and 600 s. The learning efficiencies of the samples were similar, except for scan times of 33 and 66 s. In addition, The noise levels of predicted volumes differd among scanning conditions, indicating that the noise level of a scan time ≥ 150 s does not affect the model training efficiency. Conventional filtering methods, such as median filtering and edge detection, increased the training efficiency by approximately 10% under any conditions. However, the training efficiency of 33 and 66 s-scanned samples remained relatively low. We concluded that scan time must be at least 150 s to not affect segmentation. Finally, we constructed a semantic segmentation model for 150 s-scanned CT volumes, for which the Dice loss reached 0.093. This model could not predict the lateral roots, which were not included in the training data. This limitation will be addressed by preparing appropriate training data. Conclusions: A semantic segmentation model can be constructed even with rapidly scanned CT volumes with high noise levels. Given that scanning times ≥ 150 s did not affect the segmentation results, this technique holds promise for rapid and low-dose scanning. This study offers insights into images other than CT volumes with high noise levels that are challenging to determine when annotating. [ABSTRACT FROM AUTHOR]

The Intelligent Evaluation System for Calligraphy Characters (IESCC) is used for teaching calligraphy, and users can learn calligraphy through the modifications given by the system. Chinese character skeleton extraction is an important step in the intelligent evaluation algorithm of calligraphic characters. The skeletons of Chinese characters extracted by traditional refinement algorithms are prone to redundant branches and deformed skeletons, which can lead to skeleton extraction results that do not conform to the topology of the original character. In this study, the focus lies on hard‐pen regular script, and skeleton repair and extraction are performed for these characters. According to the writing characteristics of regular script, the redundant burs are removed and the deformation zone of the thinned skeleton is detected, and then the idea of first splitting is used, then restructuring, to propose a skeleton extraction algorithm based on stroke characterization and ambiguous zone detection for hard‐pen regular script, referred to as SCAD. First, a thinning algorithm is used to extract the skeleton of Chinese characters and remove redundant pixels. By analyzing the stroke characteristics of regular script, the burrs are classified and different conditions are set to detect and remove the burrs. Then the ambiguous zones are detected according to the different kinds of junction points. Then, curvature, stroke width and direction deviation are used to analyze the continuity of stroke segments, and the decision function is used to classify the stroke segments. Finally, the stroke segments with optimal pairings were compensated by interpolation according to the direction trend. This concludes the skeleton extraction. Skeleton extraction is performed on 1000 sample characters, and the SCAD algorithm can extract the skeleton of Chinese characters with an accuracy of up to 98.37%. It is proved that the SCAD method proposed here is a practical and effective method to extract the skeleton of hard‐pen regular script. [ABSTRACT FROM AUTHOR]

Despite the existence of various super‐resolution (SR) methods, most of them focus on designing models for specific upscaling factors rather than fully exploiting inter‐scale correlation to improve efficiency. In contrast, multi‐scale SR methods can effectively reduce the redundancy of network parameters by aggregating the feature extraction processes corresponding to multiple scales into a unified process. The aim of this study is to enhance the compactness and efficiency of the SR model. Thus, an efficient multi‐scale SR method called the diverse branch feature refinement network (DBFRN) is proposed. By decoupling the training process and inference process based on the idea of structural re‐parameterization, multi‐branch topology is adopted to enrich multi‐scale learning and merge branches to achieve efficient inference with equivalent effects. Specifically, two re‐parameterization strategies are designed and two corresponding feature refinement blocks for different feature levels in multi‐scale SR network. Extensive experiments demonstrate that the proposed multi‐scale SR method is effective and efficient, and it can outperform advanced single‐scale methods in terms of quantity and quality. [ABSTRACT FROM AUTHOR]

This article proposes a target image processing method based on super-resolution reconstruction and machine learning algorithms, which solves the low-resolution problem in medical images during imaging. This method uses nonlocal autoregressive learning based on a medical image super-resolution reconstruction method. The autoregressive model is introduced into the sparse representation-based medical image super-resolution reconstruction model by utilizing medical image data inherent nonlocal similarity characteristics. At the same time, a clustering algorithm is used to obtain a classification dictionary, improving experimental efficiency. The experimental results show that ten randomly selected CT/MR images are used as test images, and each image's peak signal-to-noise ratio and structural similarity values are calculated separately. Compared with other methods, the method proposed in this paper is significantly better and can achieve ideal results, with the highest value being 31.49. This method demonstrates the feasibility of using super-resolution reconstruction and machine learning algorithms in medical image resolution. [ABSTRACT FROM AUTHOR]

Remote sensing image dehazing is a well-known remote sensing image processing task focused on restoring clean images from hazy images. The Transformer network, based on the self-attention mechanism, has demonstrated remarkable advantages in various image restoration tasks, due to its capacity to capture long-range dependencies within images. However, it is weak at modeling local context. Conversely, convolutional neural networks (CNNs) are adept at capturing local contextual information. Local contextual information could provide more details, while long-range dependencies capture global structure information. The combination of long-range dependencies and local context modeling is beneficial for remote sensing image dehazing. Therefore, in this paper, we propose a CNN-based adaptive local context enrichment module (ALCEM) to extract contextual information within local regions. Subsequently, we integrate our proposed ALCEM into the multi-head self-attention and feed-forward network of the Transformer, constructing a novel locally enhanced attention (LEA) and a local continuous-enhancement feed-forward network (LCFN). The LEA utilizes the ALCEM to inject local context information that is complementary to the long-range relationship modeled by multi-head self-attention, which is beneficial to removing haze and restoring details. The LCFN extracts multi-scale spatial information and selectively fuses them by the the ALCEM, which supplements more informative information compared with existing regular feed-forward networks with only position-specific information flow. Powered by the LEA and LCFN, a novel Transformer-based dehazing network termed LCEFormer is proposed to restore clear images from hazy remote sensing images, which combines the advantages of CNN and Transformer. Experiments conducted on three distinct datasets, namely DHID, ERICE, and RSID, demonstrate that our proposed LCEFormer achieves the state-of-the-art performance in hazy scenes. Specifically, our LCEFormer outperforms DCIL by 0.78 dB and 0.018 for PSNR and SSIM on the DHID dataset. [ABSTRACT FROM AUTHOR]

Image processing is an important area to bring out the best in an image for human interpretation. This process has been widely used in various fields such as machine vision, remote sensing image analysis, medical diagnosis, image restoration, image pattern recognition, and video processing. During the acquisition and transmission of digital images, several random signals, known as noise, can affect image quality. To reduce or to remove efficiently the noise in the acquired or transmitted images, various image denoising techniques can be applied. The performance of denoising methods increases progressively when the noise parameters are taken into account as input parameters. Traditional denoising approaches adopt some assumptions to model noise, such noise is known as purely additive or multiplicative, pixel-independent, and channel-invariant. Usually, these assumptions limit the denoising effect due to inaccurate estimation of noise parameters in these algorithm models. However, the real noise model is signal-dependent and even device-dependent. In this paper, a new denoising method called signal-dependant noise-reducing anisotropic diffusion is developed, which is a version of the speckle reducing anisotropic diffusion (SRAD) filter. It differs from the standard SRAD filter approach by the insertion of a suitable noise parameters estimation framework. The new filter is designed to handle a variety of images corrupted by several common types of signal-dependent noises that are produced by charge-coupled device sensors. As well as it offers great potential for denoising with preserving textures and fine details. Extensive experiments demonstrate a significant increase in the image denoising performance in terms of SNR and RMSE. Qualitative (visual) results underline the efficacy of the proposed algorithm for filtering mixed signal-dependent noise. [ABSTRACT FROM AUTHOR]

Reconstruction of 2-dimensional transcerebellar images into 3-dimensional transcerebellar has an important role in diagnosis and treatment planning in neurology. In this study, we propose the 3D-FHNet method to produce an accurate three-dimensional representation of transcerebellar images. The process begins with the selection of images and the application of image augmentation and enhancement techniques to improve the quality of the initial image. Furthermore, segmentation was carried out using two different architectures, namely U-Net and LinkNet, to compare the performance of the two. After training, both architectures can process object segmentation properly. The best performance was produced by U-Net with a pixel accuracy of 99.83%, Mean IU of 89.71%, FPR of 0.91%, Precision of 85.78%, Recall of 85.31%, and F1 Score of 85.31%. With this accuracy, cross-validation was carried out using a 10-Fold. The experimental results show that U-Net gives better results in terms of transcerebellar image segmentation. After the segmentation process, an initial reconstruction is carried out using the PiFUHD architecture which produces a three-dimensional object. The results of this reconstruction are then taken from four sides to be introduced to the 3D-FHNet architecture, because the main concept of 3DFHNet is to use multiview input. To measure the accuracy of the model, the IoU (Intersection over Union) metric is used for the ground truth obtained from the PiFUHD method. This metric is used to compare the similarities between the reconstruction results and the ground truth, so that information can be obtained about the extent to which the model has succeeded in reconstructing three-dimensional objects accurately. Model performance can be calculated by using the input as ground truth or the IoU (Intersection over Union) method with an average of 76.76%. By using the 3D-FHNet method, three-dimensional image reconstruction of the ultrasound image of the fetal head can be performed accurately. The experimental results show that the 3D-FHNet method produces a more accurate three-dimensional representation of the transcerebellar image compared to the previous method. These results demonstrate great potential in improving diagnosis and treatment planning in neurology, as well as making important contributions to the development of medical image processing technologies. [ABSTRACT FROM AUTHOR]

Person image generation aims to generate images that maintain the original human appearance in different target poses. Recent works have revealed that the critical element in achieving this task is the alignment of appearance domain and pose domain. Previous alignment methods, such as appearance flow warping, correspondence learning and cross attention, often encounter challenges when it comes to producing fine texture details. These approaches suffer from limitations in accurately estimating appearance flows due to the lack of global receptive field. Alternatively, they can only perform cross‐domain alignment on high‐level feature maps with small spatial dimensions since the computational complexity increases quadratically with larger feature sizes. In this article, the significance of multi‐scale alignment, in both low‐level and high‐level domains, for ensuring reliable cross‐domain alignment of appearance and pose is demonstrated. To this end, a novel and effective method, named Multi‐scale Cross‐domain Alignment (MCA) is proposed. Firstly, MCA adopts global context aggregation transformer to model multi‐scale interaction between pose and appearance inputs, which employs pair‐wise window‐based cross attention. Furthermore, leveraging the integrated global source information for each target position, MCA applies flexible flow prediction head and point correlation to effectively conduct warping and fusing for final transformed person image generation. Our proposed MCA achieves superior performance on two popular datasets than other methods, which verifies the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Hazy images suffer from two problems. The low contrast can be enhanced by estimating a transmission layer, and the colour cast can be restored by estimating an airlight. These two variables, together with the albedo layer, are the constitutive elements of a hazy image. The resulting quality of dehazed images is inextricably linked to the accurate estimation of these components. However, it is ill‐posed to decompose these variables from a single image. As such, this paper presents an innovative algorithm intended to facilitate the optimal decomposition of a hazy image. Using the Markov random field model, an optimal framework is established that allows the simultaneous estimation of the three components across the three‐colour channels. To improve the visual quality, three improvements are proposed in the variational solution for the optimal components. The dehazed result is recomposed from the components with the transmission enhanced to circumvent any potential artefacts or information loss. Extensive experiments on natural images corroborate that the proposed algorithm outperforms state‐of‐the‐art dehazing methods, both qualitatively and quantitatively. [ABSTRACT FROM AUTHOR]

Inverse halftoning is a technology that converts a binary image into a continuous tone image. Due to the wide application of inverse halftoning, many scholars have proposed several deep convolutional neural networks (DCNN) to optimize their performance. According to the observation, there is still room for improvement in content generation and detail recovery of the inverse halftone images generated by using the existing methods. Therefore, an inverse halftoning method based on supervised DCNN is proposed in this paper. The method consists of two parts: the multi‐level feature extraction model uses the down‐sampling to extract the features from the halftone image and remove the halftone noise dots on flat areas, which is implemented by four convolutional layers; the image reconstruction model uses up‐sampling to reconstruct image information, which is realized by four convolutional layers and two dense residual blocks. At the same time, in order to further recover the details, the down‐sampling feature maps and up‐sampling feature maps of the same size are concatenated by addition layers. Experimental results show that compared with other methods, the inverse halftone images obtained by the proposed network have better results in both subjective and objective evaluations. [ABSTRACT FROM AUTHOR]

Single image dehazing in real scenarios is still a particularly challenging task due to the large domain discrepancy between synthetic and real hazy images. To address this problem, the dual‐route domain‐aware adaptation framework with the bridging domain is proposed. First, inspired by the effective multistage strategy in low‐level tasks, the bridging domain is constructed with the color‐preserved adaptive histogram equalization to facilitate the adaptation and improve the performance of real hazy images. Second, the totally shared structure is relaxed and the residual dual‐path domain‐aware modules (RDDM) for synthetic and the bridging domains are proposed, which facilitates extracting the domain‐specific haze information with the different parameters. Third, a half‐cyclic constraint is proposed for the unsupervised hazy images to avoid structure distortion during the unsupervised adversarial training process. Finally, for convenience in the inference stage, feature enhancement modules (FEM) are proposed for the original real hazy images to learn the pre‐process operation in the training stage. Extensive qualitative and quantitative experiments demonstrate that the proposed method significantly improves the dehazing performance on synthetic and real hazy images. [ABSTRACT FROM AUTHOR]

Coding algorithms are usually designed to faithfully reconstruct images, which limits the expected gains in compression. A new approach based on generative models allows for new compression algorithms that can reach drastically lower compression rates. Instead of pixel fidelity, these algorithms aim at faithfully generating images that have the same high-level interpretation as their inputs. In that context, the challenge becomes to set a good representation for the semantics of an image. While text or segmentation maps have been investigated and have shown their limitations, in this paper, we ask the following question: do powerful foundation models such as CLIP provide a semantic description suited for compression? By suited for compression, we mean that this description is robust to traditional compression tools and, in particular, quantization. We show that CLIP fulfills semantic robustness properties. This makes it an interesting support for generative compression. To make that intuition concrete, we propose a proof-of-concept for a generative codec based on CLIP. Results demonstrate that our CLIP-based coder beats state-of-the-art compression pipelines at extremely low bitrates (0.0012 BPP), both in terms of image quality (65.3 for MUSIQ) and semantic preservation (0.86 for the Clip score).

Abstract Rain removal is very important for many applications in computer vision, and it is a challenging problem due to its ill‐posed nature, especially for single‐image deraining. In order to remove rain streaks more thoroughly, as well as to retain more details, a progressive dilation dense residual fusion network is proposed. The entire network is designed in a cascade manner with multiple fusion blocks. The fusion block consists of a dilation dense residual block (DDRB) and a dense residual feature fusion block (DRFFB), where DDRB is created for feature extraction and DRFFB is mainly designed for feature fusion operation. Meanwhile, detail compensation memory mechanism (DCMM) is leveraged between each of two cascade modules to retain more background details. Compared with previous state‐of‐the‐art methods, extensive experiments show that the proposed method can achieve better results, in terms of rain streaks removal and background details preservation. Furthermore, the authors’ network also shows its superiority for image noise removal.

Image inpainting networks based on deep learning techniques have been widely used in many important fields. However, most inpainting networks fail to generate desirable repaired images. This may be due to their failure to extract effective features and accurately assign high weights to the undamaged regions. To alleviate these problems, an image inpainting network based on gated convolution and multi‐level attention mechanism (IIN‐GCMAM) is proposed in this paper. This network follows encoder–decoder architecture, consisting of the gated convolution encoder (GC‐encoder) and the multi‐level attention mechanism decoder (MAM‐decoder). The GC‐encoder weighs the extracted features with gated convolutions, which reduces the interference caused by the damaged regions. The multi‐level attention mechanism employed in the MAM‐decoder uses multi‐scale feature maps spatially and channel‐wise to improve the consistency in global structure and the fineness of repaired results. Extensive experiments are conducted on the common datasets, Paris StreetView and CelebA. Experimental results indicate that the proposed IIN‐GCMAM can achieve a good performance on the common evaluation metrics and visual effects. It can achieve 0.0408, 0.720, and 22.27 in MAE, SSIM, and PSNR at the mask ratio of 50%–60%, respectively. [ABSTRACT FROM AUTHOR]

To develop a fast and effective image deblurring method, the blind recovery of motion‐blurred images based on DeblurGAN(GAN, Generative Adversarial Networks) is researched.Firstly, the number of residual modules in the DeblurGAN network is changed, and an attempt is made to optimize the network structure to achieve better results in the blind recovery of motion‐blurred images. Secondly, an image deblurring method based on the improved DeblurGAN network is proposed.This paper makes corresponding adjustments to the generator and discriminator networks to change their input and output size to 512 × 512 while keeping the overall structure of the network unchanged, and the experimental results show that the quality of the restored images has been greatly improved. In addition, the images were divided into four classes of images,to achieve improved restoration of blurred images with specific content. Experimental results on the benchmark GoPro dataset validate that the improved DeblurGAN achieves more than 1.5 dB improvement than DeblurGAN in terms of peak signal‐to‐noise ratio (PSNR) as trained utilizing the same amount of data, and structural similarity assessment (SSIM) evaluation means and maximum values increased between 0.2 and 0.3. The improved DeblurGAN is more effective in terms of both blur removal and detail recovery. [ABSTRACT FROM AUTHOR]

The problem of monitoring space debris (SD) of anthropogenic origin is discussed to ensure the safety of near-Earth space environment exploration. For dynamic low-orbit SD fragments, acquisition and processing technologies of short-exposure image series, obtained in a non-redundant aperture synthesis array, are proposed and investigated. Specific features of the optical transfer function (OTF) of the non-redundant array, namely, its "insular" character, lead to difficulties in restoring the spatial spectrum of the control object, undistorted by the atmosphere, in the whole spatial-frequency domain of the array. It is shown that to restore the object's spatial spectrum module, undistorted by the atmosphere, in the whole spatial frequency domain of the array, it is necessary to use algorithm of the Labeyrie method, modified by the authors of the present work. To restore the phase of the spatial spectrum, it is necessary to use the algorithms of unfolding the closed phase equations of the triple correlations method. By physical modeling of the formation and registration processes of analog images in the atmosphere-array system and subsequent statistical processing of digitized images, experimental evidence is given of the effectiveness of the proposed restoring technologies to achieve high resolution of SD fragments in the array, aimed at their subsequent identification and removal. Recommendations are given for practical implementation of the considered promising construction and application technologies of optical systems to control SD fragments. • The problem of space debris (SD) control is investigated to ensure the safety of space flights. • Technologies for short-exposure images of control objects in a non-redundant aperture synthesis array for SD were proposed. • Algorithms of digital processing and restoration of the diffraction images of control objects in the array were developed. • Restoring technologies of SD image processing in non-redundant aperture synthesis array were studied by PC simulation methods. • High resolution of the observed SD fragments for their subsequent identification and removal were experimentally shown. [ABSTRACT FROM AUTHOR]

Seismic interrogation of the upper mantle from the base of the crust to the top of the mantle transition zone has revealed discontinuities that are variable in space, depth, lateral extent, amplitude and lack a unified explanation for their origin. Improved constraints on the detectability and properties of mantle discontinuities can be obtained with P-to-S receiver function (Ps-RF) where energy scatters from P to S as seismic waves propagate across discontinuities of interest. However, due to the interference of crustal multiples, uppermost mantle discontinuities are more commonly imaged with lower resolution S-to-P receiver function (Sp-RF). In this study, a new method called CRISP-RF (Clean Receiver-function Imaging using SParse Radon Filters) is proposed, which incorporates ideas from compressive sensing and model-based image reconstruction. The central idea involves applying a sparse Radon transform to effectively decompose the Ps-RF into its underlying wavefield contributions, that is direct conversions, multiples, and noise, based on the phase moveout and coherence. A masking filter is then designed and applied to create a multiple-free and denoised Ps-RF. We demonstrate, using synthetic experiment, that our implementation of the Radon transform using a sparsity-promoting regularization outperforms the conventional least-squares methods and can effectively isolate direct Ps conversions. We further apply the CRISP-RF workflow on real data, including single station data on cratons, common-conversion-point stack at continental margins and seismic data from ocean islands. The application of CRISP-RF to global data sets will advance our understanding of the enigmatic origins of the upper mantle discontinuities like the ubiquitous mid-lithospheric discontinuity and the elusive X-discontinuity. [ABSTRACT FROM AUTHOR]

Multi-exposure image fusion (MEF) is a computational approach that amalgamates multiple images, each captured at varying exposure levels, into a singular, high-quality image that faithfully encapsulates the visual information from all the contributing images. Deep learning-based MEF methodologies often confront obstacles due to the inherent inflexibilities of neural network structures, presenting difficulties in dynamically handling an unpredictable amount of exposure inputs. In response to this challenge, we introduce Ref-MEF, a method for color image multi-exposure fusion guided by a reference image designed to deal with an uncertain amount of inputs. We establish a reference-guided exposure correction (REC) module based on channel attention and spatial attention, which can correct input features and enhance pre-extraction features. The exposure-guided feature fusion (EGFF) module combines original image information and uses Gaussian filter weights for feature fusion while keeping the feature dimensions constant. The image reconstruction is completed through a gated context aggregation network (GCAN) and global residual learning GRL. Our refined loss function incorporates gradient fidelity, producing high dynamic range images that are rich in detail and demonstrate superior visual quality. In evaluation metrics focused on image features, our method exhibits significant superiority and leads in holistic assessments as well. It is worth emphasizing that as the number of input images increases, our algorithm exhibits notable computational efficiency. [ABSTRACT FROM AUTHOR]

How to represent a human face pattern? While it is presented in a continuous way in human visual system, computers often store and process it in a discrete manner with 2D arrays of pixels. The authors attempt to learn a continuous surface representation for face image with explicit function. First, an explicit model (EmFace) for human face representation is proposed in the form of a finite sum of mathematical terms, where each term is an analytic function element. Further, to estimate the unknown parameters of EmFace, a novel neural network, EmNet, is designed with an encoder‐decoder structure and trained from massive face images, where the encoder is defined by a deep convolutional neural network and the decoder is an explicit mathematical expression of EmFace. The authors demonstrate that our EmFace represents face image more accurate than the comparison method, with an average mean square error of 0.000888, 0.000936, 0.000953 on LFW, IARPA Janus Benchmark‐B, and IJB‐C datasets. Visualisation results show that, EmFace has a higher representation performance on faces with various expressions, postures, and other factors. Furthermore, EmFace achieves reasonable performance on several face image processing tasks, including face image restoration, denoising, and transformation. [ABSTRACT FROM AUTHOR]

Recording neuronal activity using multiple electrodes has been widely used to understand the functional mechanisms of the brain. Increasing the number of electrodes allows us to decode more variety of functionalities. However, handling massive amounts of multichannel electrophysiological data is still challenging due to the limited hardware resources and unavoidable thermal tissue damage. Here, we present machine learning (ML)-based reconstruction of high-frequency neuronal spikes from subsampled low-frequency band signals. Inspired by the equivalence between high-frequency restoration and super-resolution in image processing, we applied a transformer ML model to neuronal data recorded from both in vitro cultures and in vivo male mouse brains. Even with the x8 downsampled datasets, our trained model reasonably estimated high-frequency information of spiking activity, including spike timing, waveform, and network connectivity. With our ML-based data reduction applicable to existing multichannel recording hardware while achieving neuronal signals of broad bandwidths, we expect to enable more comprehensive analysis and control of brain functions. Multichannel neural recording enhances understanding of brain function, but handling large data is challenging. Here, the authors develop machine learning-based high frequency spike reconstruction from subsampled low-frequency neuronal signals. [ABSTRACT FROM AUTHOR]

In the domain of synthetic aperture radar (SAR) image processing, a prevalent issue persists wherein research predominantly focuses on single-task learning, often neglecting the concurrent impact of speckle noise and low resolution on SAR images. Currently, there are two main processing strategies. The first strategy involves conducting speckle reduction and super-resolution processing step by step. The second strategy involves performing speckle reduction as an auxiliary step, with a focus on enhancing the primary task of super-resolution processing. However, both of these strategies exhibit clear deficiencies. Nevertheless, both tasks jointly focus on two key aspects, enhancing SAR quality and restoring details. The fusion of these tasks can effectively leverage their task correlation, leading to a significant improvement in processing effectiveness. Additionally, multi-temporal SAR images covering imaging information from different time periods exhibit high correlation, providing deep learning models with a more diverse feature expression space, greatly enhancing the model's ability to address complex issues. Therefore, this study proposes a deep learning network for integrated speckle reduction and super-resolution in multi-temporal SAR (ISSMSAR). The network aims to reduce speckle in multi-temporal SAR while significantly improving the image resolution. Specifically, it consists of two subnetworks, each taking the SAR image at time 1 and the SAR image at time 2 as inputs. Each subnetwork includes a primary feature extraction block (PFE), a high-level feature extraction block (HFE), a multi-temporal feature fusion block (FFB), and an image reconstruction block (REC). Following experiments on diverse data sources, the results demonstrate that ISSMSAR surpasses speckle reduction and super-resolution methods based on a single task in terms of both subjective perception and objective evaluation metrics regarding the quality of image restoration. [ABSTRACT FROM AUTHOR]