Search Results (302)

Image super-resolution (SR) is essential for enhancing image quality in critical applications, such as medical imaging and satellite remote sensing. However, existing methods were often limited in their ability to effectively process and integrate multi-scales information from fine textures to global structures. To address these limitations, this paper proposes DSCNN, a dynamic snake convolution neural network for enhanced image super-resolution. DSCNN optimizes feature extraction and network architecture to enhance both performance and efficiency: To improve feature extraction, the core innovation is a feature extraction and enhancement module with dynamic snake convolution that dynamically adjusts the convolution kernel’s shape and position to better fit the image’s geometric structures, significantly improving feature extraction. To optimize the network’s structure, DSCNN employs an enhanced residual network framework. This framework utilizes parallel convolutional layers and a global feature fusion mechanism to further strengthen feature extraction capability and gradient flow efficiency. Additionally, the network incorporates a SwishReLU-based activation function and a multi-scale convolutional concatenation structure. This multi-scale design effectively captures both local details and global image structure, enhancing SR reconstruction. In summary, the proposed DSCNN outperforms existing methods in both objective metrics and visual perception (e.g., our method achieved optimal PSNR and SSIM results on the Set5 $\times 4$ dataset). Full article

Deep learning-based super-resolution (SR) frameworks are widely used in remote sensing applications. However, existing SR models still face limitations, particularly in recovering contours, fine features, and textures, as well as in effectively integrating channel information. To address these challenges, this study introduces a novel residual model named OARN (Optimized Attention Residual Network) specifically designed to enhance the visual quality of low-resolution images. The network operates on the Y channel of the YCbCr color space and integrates LKA (Large Kernel Attention) and OCM (Optimized Convolutional Module) blocks. These components can restore large-scale spatial relationships and refine textures and contours, improving feature reconstruction without significantly increasing computational complexity. The performance of OARN was evaluated using satellite images from WorldView-2, GaoFen-2, and Microsoft Virtual Earth. Evaluation was conducted using objective quality metrics, such as Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), Edge Preservation Index (EPI), and Perceptual Image Patch Similarity (LPIPS), demonstrating superior results compared to state-of-the-art methods in both objective measurements and subjective visual perception. Moreover, OARN achieves this performance while maintaining computational efficiency, offering a balanced trade-off between processing time and reconstruction quality. Full article

The goal of single-image super-resolution (SISR) tasks is to generate high-definition images from low-quality inputs, with practical uses spanning healthcare diagnostics, aerial imaging, and surveillance systems. Although cnns have considerably improved image reconstruction quality, existing methods still face limitations, including inadequate restoration of high-frequency details, high computational complexity, and insufficient adaptability to complex scenes. To address these challenges, we propose an Omni-dimensional Dynamic Convolutional Network (ODConvNet) tailored for SISR tasks. Specifically, ODConvNet comprises four key components: a Feature Extraction Block (FEB) that captures low-level spatial features; an Omni-dimensional Dynamic Convolution Block (DCB), which utilizes a multidimensional attention mechanism to dynamically reweight convolution kernels across spatial, channel, and kernel dimensions, thereby enhancing feature expressiveness and context modeling; a Deep Feature Extraction Block (DFEB) that stacks multiple convolutional layers with residual connections to progressively extract and fuse high-level features; and a Reconstruction Block (RB) that employs subpixel convolution to upscale features and refine the final HR output. This mechanism significantly enhances feature extraction and effectively captures rich contextual information. Additionally, we employ an improved residual network structure combined with a refined Charbonnier loss function to alleviate gradient vanishing and exploding to enhance the robustness of model training. Extensive experiments conducted on widely used benchmark datasets, including DIV2K, Set5, Set14, B100, and Urban100, demonstrate that, compared with existing deep learning-based SR methods, our ODConvNet method improves Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), and the visual quality of SR images is also improved. Ablation studies further validate the effectiveness and contribution of each component in our network. The proposed ODConvNet offers an effective, flexible, and efficient solution for the SISR task and provides promising directions for future research. Full article

Single-image super-resolution (SISR) plays a critical role in enhancing visual quality for real-world applications, including industrial inspection and embedded vision systems. While deep learning-based approaches have made significant progress in SR, existing lightweight SR models often fail to accurately reconstruct high-frequency textures, especially under complex degradation scenarios, resulting in blurry edges and structural artifacts. To address this challenge, we propose a Dense Residual Fused Attention Network (DRFAN), a novel lightweight hybrid architecture designed to enhance high-frequency texture recovery in challenging degradation conditions. Moreover, by coupling convolutional layers and attention mechanisms through gated interaction modules, the DRFAN enhances local details and global dependencies with linear computational complexity, enabling the efficient utilization of multi-level spatial information while effectively alleviating the loss of high-frequency texture details. To evaluate its effectiveness, we conducted ×4 super-resolution experiments on five public benchmarks. The DRFAN achieves the best performance among all compared lightweight models. Visual comparisons show that the DRFAN restores more accurate geometric structures, with up to +1.2 dB/+0.0281 SSIM gain over SwinIR-S on Urban100 samples. Additionally, on a domain-specific rice grain dataset, the DRFAN outperforms SwinIR-S by +0.19 dB in PSNR and +0.0015 in SSIM, restoring clearer textures and grain boundaries essential for industrial quality inspection. The proposed method provides a compelling balance between model complexity and image reconstruction fidelity, making it well-suited for deployment in resource-constrained visual systems and industrial applications. Full article

Deep learning has shown significant advantages over traditional spatial interpolation methods in single image super-resolution (SISR). Recently, many studies have applied super-resolution (SR) methods to generate high-resolution (HR) digital bathymetry models (DBMs), but substantial differences between DBM and natural images have been ignored, which leads to serious distortions and inaccuracies. Given the critical role of HR DBM in marine resource exploitation, economic development, and scientific innovation, we propose a frequency-aware texture matching transformer (FTT) for DBM SR, incorporating global terrain feature extraction (GTFE), high-frequency feature extraction (HFFE), and a terrain matching block (TMB). GTFE has the capability to perceive spatial heterogeneity and spatial locations, allowing it to accurately capture large-scale terrain features. HFFE can explicitly extract high-frequency priors beneficial for DBM SR and implicitly refine the representation of high-frequency information in the global terrain feature. TMB improves fidelity of generated HR DBM by generating position offsets to restore warped textures in deep features. Experimental results have demonstrated that the proposed FTT has superior performance in terms of elevation, slope, aspect, and fidelity of generated HR DBM. Notably, the root mean square error (RMSE) of elevation in steep terrain has been reduced by 4.89 m, which is a significant improvement in the accuracy and precision of the reconstruction. This research holds significant implications for improving the accuracy of DBM SR methods and the usefulness of HR bathymetry products for future marine research. Full article

In recent years, remote sensing imagery has become indispensable for applications such as environmental monitoring, land use classification, and urban planning. However, the physical constraints of satellite imaging systems frequently limit the spatial resolution of these images, impeding the extraction of fine-grained information critical to downstream tasks. Super-resolution (SR) techniques thus emerge as a pivotal solution to enhance the spatial fidelity of remote sensing images via computational approaches. While deep learning-based SR methods have advanced reconstruction accuracy, their high computational complexity and large parameter counts restrict practical deployment in real-world remote sensing scenarios—particularly on edge or low-power devices. To address this gap, we propose LSANet, a lightweight SR network customized for remote sensing imagery. The core of LSANet is the large separable kernel attention mechanism, which efficiently expands the receptive field while retaining low computational overhead. By integrating this mechanism into an enhanced residual feature distillation module, the network captures long-range dependencies more effectively than traditional shallow residual blocks. Additionally, a residual feature enhancement module, leveraging contrast-aware channel attention and hierarchical skip connections, strengthens the extraction and integration of multi-level discriminative features. This design preserves fine textures and ensures smooth information propagation across the network. Extensive experiments on public datasets such as UC Merced Land Use and NWPU-RESISC45 demonstrate LSANet’s competitive or superior performance compared to state-of-the-art methods. On the UC Merced Land Use dataset, LSANet achieves a PSNR of 34.33, outperforming the best-baseline HSENet with its PSNR of 34.23 by 0.1. For SSIM, LSANet reaches 0.9328, closely matching HSENet’s 0.9332 while demonstrating excellent metric-balancing performance. On the NWPU-RESISC45 dataset, LSANet attains a PSNR of 35.02, marking a significant improvement over prior methods, and an SSIM of 0.9305, maintaining strong competitiveness. These results, combined with the notable reduction in parameters and floating-point operations, highlight the superiority of LSANet in remote sensing image super-resolution tasks. Full article

Image super-resolution (SR) aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs. It plays a crucial role in applications such as medical imaging, surveillance, and remote sensing. However, due to the ill-posed nature of the task and the inherent limitations of imaging sensors, obtaining accurate HR images remains challenging. While numerous methods have been proposed, the traditional approaches suffer from oversmoothing and limited generalization; CNN-based models lack the ability to capture long-range dependencies; and Transformer-based solutions, although effective in modeling global context, are computationally intensive and prone to texture loss. To address these issues, we propose a hybrid CNN–Transformer architecture that cascades a pixel-wise self-attention non-local means module (PSNLM) and an adaptive dual-path multi-scale fusion block (ADMFB). The PSNLM is inspired by the non-local means (NLM) algorithm. We use weighted patches to estimate the similarity between pixels centered at each patch while limiting the search region and constructing a communication mechanism across ranges. The ADMFB enhances texture reconstruction by adaptively aggregating multi-scale features through dual attention paths. The experimental results demonstrate that our method achieves superior performance on multiple benchmarks. For instance, in challenging ×4 super-resolution, our method outperforms the second-best method by 0.0201 regarding the Structural Similarity Index (SSIM) on the BSD100 dataset. On the texture-rich Urban100 dataset, our method achieves a 26.56 dB Peak Signal-to-Noise Ratio (PSNR) and 0.8133 SSIM. Full article

Horse iris recognition is a non-invasive identification method with great potential for precise management in intelligent horse farms. However, horses’ natural vigilance often leads to stress and resistance when exposed to close-range infrared cameras. This behavior makes it challenging to capture clear iris images, thereby reducing recognition performance. This paper addresses the challenge of generating high-resolution iris images from existing low-resolution counterparts. To this end, we propose a novel hybrid-architecture image super-resolution (SR) network. Central to our approach is the design of Paired Asymmetric Transformer Block (PATB), which incorporates Contextual Query Generator (CQG) to efficiently capture contextual information and model global feature interactions. Furthermore, we introduce an Efficient Residual Dense Block (ERDB), specifically engineered to effectively extract finer-grained local features inherent in the image data. By integrating PATB and ERDB, our network achieves superior fusion of global contextual awareness and local detail information, thereby significantly enhancing the reconstruction quality of horse iris images. Experimental evaluations on our self-constructed dataset of horse irises demonstrate the effectiveness of the proposed method. In terms of standard image quality metrics, it achieves the PSNR of 30.5988 dB and SSIM of 0.8552. Moreover, in terms of identity-recognition performance, the method achieves Precision, Recall, and F1-Score of 81.48%, 74.38%, and 77.77%, respectively. This study provides a useful contribution to digital horse farm management and supports the ongoing development of smart animal husbandry. Full article

Reference-based super-resolution (RefSR) enhances the detail restoration capability of low-resolution images (LR) by utilizing the details and texture information of external reference images (Ref). This study proposes a RefSR method based on hash adaptive matching and progressive multi-scale dynamic aggregation to improve the super-resolution reconstruction capability. Firstly, to address the issue of feature matching, this chapter proposes a hash adaptive matching module. On the basis of similarity calculation between traditional LR images and Ref images, self-similarity information of LR images is added to assist in super-resolution reconstruction. By dividing the feature space into multiple hash buckets through spherical hashing, the matching range is narrowed down from global search to local neighborhoods, enabling efficient matching in more informative regions. This not only retains global modeling capabilities, but also significantly reduces computational costs. In addition, a learnable similarity scoring function has been designed to adaptively optimize the similarity score between LR images and Ref images, improving matching accuracy. Secondly, in the process of feature transfer, this chapter proposes a progressive multi-scale dynamic aggregation module. This module utilizes dynamic decoupling filters to simultaneously perceive texture information in both spatial and channel domains, extracting key information more accurately and effectively suppressing irrelevant texture interference. In addition, this module enhances the robustness of the model to large-scale biases by gradually adjusting features at different scales, ensuring the accuracy of texture transfer. The experimental results show that this method achieves superior super-resolution reconstruction performance on multiple benchmark datasets. Full article

In image processing, the acquisition step plays a fundamental role because it determines image quality. The present paper focuses on the issue of blur and suggests ways of assessing contrast. The logic of this work consists in evaluating the sharpness of an image by means of objective measures based on mathematical, physical, and optical justifications in connection with the human visual system. This is why the Logarithmic Image Processing (LIP) framework was chosen. The sharpness of an image is usually assessed near objects’ boundaries, which encourages the use of gradients, with some major drawbacks. Within the LIP framework, it is possible to overcome such problems using a “contour detector” tool based on the notion of Logarithmic Additive Contrast (LAC). Considering a sequence of images increasingly blurred, we show that the use of LAC enables images to be re-classified in accordance with their defocus level, demonstrating the relevance of the method. The proposed algorithm has been shown to outperform five conventional methods for assessing image sharpness. Moreover, it is the only method that is insensitive to brightness variations. Finally, various application examples are presented, like automatic autofocus control or the comparison of two blur removal algorithms applied to the same image, which particularly concerns the field of Super Resolution (SR) algorithms. Such algorithms multiply (×2, ×3, ×4) the resolution of an image using powerful tools (deep learning, neural networks) while correcting the potential defects (blur, noise) that could be generated by the resolution extension itself. We conclude with the prospects for this work, which should be part of a broader approach to estimating image quality, including sharpness and perceived contrast. Full article

A majority of hyperspectral super-resolution methods aim to enhance the spatial resolution of hyperspectral imaging data (HSI) by integrating high-resolution multispectral imaging data (MSI), leveraging rich spectral information for various geospatial applications. Key challenges include spectral distortions from high-frequency spatial data, high computational complexity, and limited training data, particularly for new-generation sensors with unique noise patterns. In this contribution, we propose a novel parallel patch-wise sparse residual learning (P²SR) algorithm for resolution enhancement based on fusion of HSI and MSI. The proposed method uses multi-decomposition techniques (i.e., Independent component analysis, Non-negative matrix factorization, and 3D wavelet transforms) to extract spatial and spectral features to form a sparse dictionary. The spectral and spatial characteristics of the scene encoded in the dictionary enable reconstruction through a first-order optimization algorithm to ensure an efficient sparse representation. The final spatially enhanced HSI is reconstructed by combining the learned features from low-resolution HSI and applying an MSI-regulated guided filter to enhance spatial fidelity while minimizing artifacts. P²SR is deployable on a high-performance computing (HPC) system with parallel processing, ensuring scalability and computational efficiency for large HSI datasets. Extensive evaluations on three diverse study sites demonstrate that P²SR consistently outperforms traditional and state-of-the-art (SOA) methods in both quantitative metrics and qualitative spatial assessments. Specifically, P²SR achieved the best average PSNR (25.2100) and SAM (12.4542) scores, indicating superior spatio-spectral reconstruction contributing to sharper spatial features, reduced mixed pixels, and enhanced geological features. P²SR also achieved the best average ERGAS (8.9295) and Q2n (0.5156), which suggests better overall fidelity across all bands and perceptual accuracy with the least spectral distortions. Importantly, we show that P²SR preserves critical spectral signatures, such as Fe²⁺ absorption, and improves the detection of fine-scale environmental and geological structures. P²SR’s ability to maintain spectral fidelity while enhancing spatial detail makes it a powerful tool for high-precision remote sensing applications, including mineral mapping, land-use analysis, and environmental monitoring. Full article

Facial image super-resolution seeks to reconstruct high-quality details from low-resolution inputs, yet traditional methods, such as interpolation, convolutional neural networks (CNNs), and generative adversarial networks (GANs), often fall short, suffering from insufficient realism, loss of high-frequency details, and training instability. Furthermore, many existing models inadequately incorporate facial structural attributes and semantic information, leading to semantically inconsistent generated images. To overcome these limitations, this study introduces an attribute-prior conditional diffusion implicit model that enhances the controllability of super-resolution generation and improves detail restoration capabilities. Methodologically, the framework consists of four components: a pre-super-resolution module, a facial attribute extraction module, a global feature encoder, and an enhanced conditional diffusion implicit model. Specifically, low-resolution images are subjected to preliminary super-resolution and attribute extraction, followed by adaptive group normalization to integrate feature vectors. Additionally, residual convolutional blocks are incorporated into the diffusion model to utilize attribute priors, complemented by self-attention mechanisms and skip connections to optimize feature transmission. Experiments conducted on the CelebA and FFHQ datasets demonstrate that the proposed model achieves an increase of 2.16 dB in PSNR and 0.08 in SSIM under an 8× magnification factor compared to SR3, with the generated images displaying more realistic textures. Moreover, manual adjustment of attribute vectors allows for directional control over generation outcomes (e.g., modifying facial features or lighting conditions), ensuring alignment with anthropometric characteristics. This research provides a flexible and robust solution for high-fidelity face super-resolution, offering significant advantages in detail preservation and user controllability. Full article

Image super-resolution (SR) models based on the generative adversarial network (GAN) face challenges such as unnatural facial detail restoration and local blurring. This paper proposes an improved GAN-based model to address these issues. First, a Multi-scale Hybrid Attention Residual Block (MHARB) is designed, which dynamically enhances feature representation in critical face regions through dual-branch convolution and channel-spatial attention. Second, an Edge-guided Enhancement Block (EEB) is introduced, generating adaptive detail residuals by combining edge masks and channel attention to accurately recover high-frequency textures. Furthermore, a multi-scale discriminator with a weighted sub-discriminator loss is developed to balance global structural and local detail generation quality. Additionally, a phase-wise training strategy with dynamic adjustment of learning rate (Lr) and loss function weights is implemented to improve the realism of super-resolved face images. Experiments on the CelebA-HQ dataset demonstrate that the proposed model achieves a PSNR of 23.35 dB, a SSIM of 0.7424, and a LPIPS of 24.86, outperforming classical models and delivering superior visual quality in high-frequency regions. Notably, this model also surpasses the SwinIR model (PSNR: 23.28 dB → 23.35 dB, SSIM: 0.7340 → 0.7424, and LPIPS: 30.48 → 24.86), validating the effectiveness of the improved model and the training strategy in preserving facial details. Full article

Background: T1 mapping has become a valuable technique in cardiac magnetic resonance imaging (CMR) for evaluating myocardial tissue properties. However, its quantitative accuracy remains limited by noise-related variability. Super-resolution deep learning-based reconstruction (SR-DLR) has shown potential in enhancing image quality across various MRI applications, yet its effectiveness in myocardial T1 mapping has not been thoroughly investigated. This study aimed to evaluate the impact of SR-DLR on noise reduction and measurement consistency in myocardial T1 mapping. Methods: This single-center retrospective observational study included 36 patients who underwent CMR between July and December 2023. T1 mapping was performed using a modified Look-Locker inversion recovery (MOLLI) sequence before and after contrast administration. Images were reconstructed with and without SR-DLR using identical scan data. Phantom studies using seven homemade phantoms with different Gd-DOTA dilution ratios were also conducted. Quantitative evaluation included mean T1 values, standard deviation (SD), and coefficient of variation (CV). Intraclass correlation coefficients (ICCs) were calculated to assess inter-observer agreement. Results: SR-DLR had no significant effect on mean native or post-contrast T1 values but significantly reduced SD and CV in both patient and phantom studies. SD decreased from 44.0 to 31.8 ms (native) and 20.0 to 14.1 ms (post-contrast), and CV also improved. ICCs indicated excellent inter-observer reproducibility (native: 0.822; post-contrast: 0.955). Conclusions: SR-DLR effectively reduces measurement variability while preserving T1 accuracy, enhancing the reliability of myocardial T1 mapping in both clinical and research settings. Full article

In recent years, great progress has been made in the field of super-resolution (SR) reconstruction based on deep learning techniques. Although image SR techniques show strong potential in image reconstruction, the effective application of these techniques to SR reconstruction of digital elevation models (DEMs) remains an important research challenge. The complexity and diversity of DEMs limits existing methods to capture subtle changes and features of the terrain, thus affecting the quality of reconstruction. To solve this problem, a DEM SR reconstruction network based on multi-path feature extraction and transformer feature enhancement is proposed in this paper. The network structure has three parts: feature extraction, image reconstruction, and feature enhancement. The feature extraction component consists of three feature extraction blocks, and each feature extraction block contains multiple multi-path feature residuals to enhance the interaction between spatial information and semantic information, so as to fully extract image features. In addition, the transformer feature enhancement module uses an encoder and decoder based design, leveraging the correlation between low- and high-dimensional features to further improve network performance. Through repeated testing and improvement, the model shows excellent performance in high-resolution DEM image reconstruction, and can generate more accurate DEMs. In terms of elevation and slope evaluation indexes, the model was 3.41% and 1.11% better compared with the existing reconstruction methods, which promotes the application of SR reconstruction technology in terrain data. Full article