Image Super-resolution Method Research Articles

ConvSRGAN: super-resolution inpainting of traditional Chinese paintings

Existing image super-resolution methods have made remarkable advancements in enhancing the visual quality of real-world images. However, when it comes to restoring Chinese paintings, these methods encounter unique challenges. This is primarily due to the difficulty in preserving intricate non-realistic details and capturing comple semantic information with high dimensionality. Moreover, the preservation of the original artwork’s distinct style and subtle artistic nuances further amplifies this complexity. To address these challenges and effectively restore traditional Chinese paintings, we propose a Convolutional Super-Resolution Generative Adversarial Network for Chinese landscape painting super-resolution, termed ConvSRGAN. We employ Enhanced Adaptive Residual Module to delve deeply into multi-scale feature extraction in images, incorporating an Enhanced High-Frequency Retention Module that leverages an Adaptive Deep Convolution Block to capture fine-grained high-frequency details across multiple levels. By combining the Multi-Scale Structural Similarity loss with conventional losses, our ConvSRGAN ensures that the model produces outputs with improved fidelity to the original image’s texture and structure. Experimental validation demonstrates significant qualitative and quantitative results when processing traditional paintings and murals datasets, particularly excelling in high-definition reconstruction tasks for landscape paintings. The reconstruction effect showcases enhanced visual fidelity and liveliness, thus affirming the effectiveness and applicability of our approach in cultural heritage preservation and restoration.

An efficient blur kernel estimation method for blind image Super-Resolution

Many existing space-variant blind image Super-Resolution (SR) methods require the generation of blur kernels for all input pixels. However, the existing methods overlook the possibility of similar blur kernel patterns among adjacent pixels, leading to spatial incoherence and wasteful computational resources. In this study, we introduce an efficient two-stage method for estimating blur kernels. Instead of generating pixel-wise kernels, a limited set of kernels at pixels with distinct information are generated first and the remaining kernels are reconstructed based on this initial set. This novel method, referred to as Anchor Detection and Kernel Reconstruction, comprises two main components: an Anchor Detection Module (ADM) and a Kernel Reconstruction Module (KRM). The objective of ADM is to identify pixels in a given Low-Resolution image that contain rich information, which are called anchors. The corresponding blur kernels, denoted as anchor kernels, are then generated for these identified pixels using a complete backbone network. The remaining blur kernels are reconstructed using KRM with a lightweight interpolation method based on the anchor kernels to enhance spatial consistency among the reconstructed pixels. Extensive experiments demonstrate that the proposed ADKR method maintains comparable performance while estimating only 50% of blur kernels with a full backbone network, reaching approximately 20% reduction in FLOPs. The code has been made available at https://github.com/xuyimin0926/ADKR.

Deep multi-scale feature mixture model for image super-resolution with multiple-focal-length degradation

Single image super-resolution is a classic problem in computer vision. In recent years, deep learning-based models have achieved unprecedented success with this problem. However, most existing deep super-resolution models unavoidably produce degraded results when applied to real-world images captured by cameras with different focal lengths. The degradation in these images is called multiple-focal-length degradation, which is spatially variant and more complicated than the bicubic downsampling degradation. To address such a challenging issue, we propose a multi-scale feature mixture model in this paper. The proposed model can intensively exploit local patterns from different scales for image super-resolution. To improve the performance, we further propose a novel loss function based on the Laplacian pyramid, which guides the model to recover the information separately of different frequency subbands. Comprehensive experiments show that our proposed model has a better ability to preserve the structure of objects and generate high-quality images, leading to the best performance compared with other state-of-the-art deep single image super-resolution methods.

Multi-level feature interaction image super-resolution network based on convolutional nonlinear spiking neural model

Image super-resolution (ISR) is designed to recover lost detail information from low-resolution images, resulting in high-quality and high-definition high-resolution images. In the existing single ISR (SISR) methods based on convolutional neural networks (CNN), however, most of the models cannot effectively combine global and local information and are also easy to ignore the correlation between different hierarchical feature information. To address these problems, this study proposes a multi-level feature interactive image super-resolution network, which is constructed by the convolutional units inspired by nonlinear spiking mechanism in nonlinear spiking neural P systems, including shallow feature processing, deep feature extraction and fusion, and reconstruction modules. The different omni domain self-attention blocks are introduced to extract global information in the deep feature extraction and fusion stage and formed a feature enhancement module having a Transformer structure using a novel convolutional unit for extracting local information. Furthermore, to adaptively fuse features between different hierarchies, we design a multi-level feature fusion module, which not only can adaptively fuse features between different hierarchies, but also can better interact with contextual information. The proposed model is compared with 16 state-of-the-art or baseline models on five benchmark datasets. The experimental results show that the proposed model not only achieves good reconstruction performance, but also strikes a good balance between model parameters and performance.

Image Super-resolution Networks based on Structure Reparameterized Convolution

The traditional image super resolution method based on convolutional neural network generally brings two problems: First, the model has only a single scale of receptive field, which cannot use the image information in a smaller range; Second, the size of the feature map will become smaller and smaller in the process of continuous convolution, and it is necessary to continuously carry out edge zeroing operation to maintain the original size, which leads to the loss of part of the edge information. To solve the above two problems, we propose a network structure based on structure re-parameterized convolution, which sets a small convolution kernel in parallel beside a large convolution kernel in the same layer, trains the two cores simultaneously, and finally merges the two cores. The experimental results show that in this way, we make the large convolution kernel could capture smaller information, which not only improves the high-frequency details of the reconstructed image but also avoids frequent edge filling to reduce the information density, and effectively speeds up the reasoning speed after the model is re-parameterized. Compared with the advanced methods, our method achieves good results.

An Omnidirectional Image Super-Resolution Method Based on Enhanced SwinIR

For the significant distortion problem caused by the special projection method of equi-rectangular projection (ERP) images, this paper proposes an omnidirectional image super-resolution algorithm model based on position information transformation, taking SwinIR as the base. By introducing a space position transformation module that supports deformable convolution, the image preprocessing process is optimized to reduce the distortion effects in the polar regions of the ERP image. Meanwhile, by introducing deformable convolution in the deep feature extraction process, the model’s adaptability to local deformations of images is enhanced. Experimental results on publicly available datasets have shown that our method outperforms SwinIR, with an average improvement of over 0.2 dB in WS-PSNR and over 0.030 in WS-SSIM for ×4 pixel upscaling.

A boosted degradation representation learning for blind image super-resolution

The significant gap between the assumed and actual degradation model will undoubtedly lead to a serious performance decrease for deep learning based image super-resolution (SR) methods. To solve this shortcoming, this paper presents an enhanced blind image super-resolution (EBSR) network based on unsupervised degradation representation learning. The proposed EBSR mainly is made up of two branching subnetworks. The first subnetwork is a residual-based degradation encoder, which is responsible to learn a high-dimensional abstract degradation representation vector from the input low-resolution (LR) image patches using contrast learning. Another branch is the degradation-aware fusion SR (DAFSR) network that completes the image SR task with the aid of the degradation representation vector learned by the encoder. To obtain an improved SR performance under various degradation settings, the degradation-aware fusion block (DAFB), the core block of DAFSR, has been elaborated, which embeds the degradation model information learned by the encoder into every layer of the network. Additionally, a shallow feature extraction block (SFEB) is constructed for DAFSR to extract global information from the LR image. The proposed EBSR can flexibly adapt to various degradation models. Extensive experimental results on synthetic datasets and real images show that the proposed EBSR is able to achieve a leading reconstruction performance over other blind SR algorithms.

Target Image Processing based on Super-resolution Reconstruction and Machine Learning Algorithm

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.

Non-local degradation modeling for spatially adaptive single image super-resolution

Existing methods for single image super-resolution (SISR) model the blur kernel as spatially invariant across the entire image, and are susceptible to the adverse effects of textureless patches. To achieve improved results, adaptive estimation of the degradation kernel is necessary. We explore the synergy of joint global and local degradation modeling for spatially adaptive blind SISR. Our model, named spatially adaptive network for blind super-resolution (SASR), employs a simple encoder to estimate global degradation representations and a decoder to extract local degradation. These two representations are fused with a cross-attention mechanism and applied using spatially adaptive filtering to enhance the local image detail. Specifically, SASR contains two novel features: (1) a non-local degradation modeling with contrastive learning to learn global and local degradation representations, and (2) a non-local spatially adaptive filtering module (SAFM) that incorporates the global degradation and spatial-detail factors to preserve and enhance local details. We demonstrate that SASR can efficiently estimate degradation representations and handle multiple types of degradation. The local representations avoid the detrimental effect of estimating the entire super-resolved image with only one kernel through locally adaptive adjustments. Extensive experiments are performed to quantitatively and qualitatively demonstrate that SASR not only performs favorably for degradation estimation but also leads to state-of-the-art blind SISR performance when compared to alternative approaches.

Large-field and fast super-resolution microscopic imaging method based on laser interferometry

In recent years, structured illumination microscopy (SIM) has been drawing great attention for both technique development and application. However, conventional SIM, which uses a spatial light modulator (SLM) for fringe projection, often has a limited field of view. To meet the demand for high-throughput microscopic imaging in biomedicine research, a large-field super-resolution (SR) fluorescence microscopic imaging method based on laser interferometry was proposed. The method that combines a two-dimensional (2D) grating for fringe pattern projection and an SLM for selecting fringe orientation can break the limitation of fringe number limited by the digital projection devices. A spatial-domain reconstruction algorithm was developed to improve the computational speed of super-resolution imaging. Finally, an experimental platform for SIM microscopy was established. A large-field view of 1380 μm × 1035 μm under a 20×/NA0.75 objective is experimentally demonstrated, and an enhancement of 1.8-fold resolution is realized. The spatial-domain reconstruction algorithm can significantly improve the computational speed by approximately 10 times faster compared to the traditional frequency-domain algorithm.

Dual-space high-frequency learning for transformer-based MRI super-resolution

Background and ObjectiveMagnetic resonance imaging (MRI) can provide rich and detailed high-contrast information of soft tissues, while the scanning of MRI is time-consuming. To accelerate MR imaging, a variety of Transformer-based single image super-resolution methods are proposed in recent years, achieving promising results thanks to their superior capability of capturing long-range dependencies. Nevertheless, most existing works prioritize the design of transformer attention blocks to capture global information. The local high-frequency details, which are pivotal to faithful MRI restoration, are unfortunately neglected. MethodsIn this work, we propose a high-frequency enhanced learning scheme to effectively improve the awareness of high frequency information in current Transformer-based MRI single image super-resolution methods. Specifically, we present two entirely plug-and-play modules designed to equip Transformer-based networks with the ability to recover high-frequency details from dual spaces: 1) in the feature space, we design a high-frequency block (Hi-Fe block) paralleled with Transformer-based attention layers to extract rich high-frequency features; while 2) in the image intensity space, we tailor a high-frequency amplification module (HFA) to further refine the high-frequency details. By fully exploiting the merits of the two modules, our framework can recover abundant and diverse high-frequency information, rendering faithful MRI super-resolved results with fine details. ResultsWe integrated our modules with six Transformer-based models and conducted experiments across three datasets. The results indicate that our plug-and-play modules can enhance the super-resolution performance of all foundational models to varying degrees, surpassing the capabilities of existing state-of-the-art single image super-resolution networks. ConclusionComprehensive comparison of super-resolution images and high-frequency maps from various methods, clearly demonstrating that our module possesses the capability to restore high-frequency information, showing huge potential in clinical practice for accelerated MRI reconstruction.

Super-resolution reconstruction method of the optical synthetic aperture image using generative adversarial network

Abstract In order to solve the contradiction between large aperture elements and high-resolution images, in this study, we propose an improved image-resolution method based on generative adversarial network (GAN). First, we analyze the imaging principle of the optical synthetic aperture. Further, we improve a super-resolution GAN; especially, this network uses a multi-scale convolutional cascade to obtain global features of the image, and a multi-scale receptive field block and residual in residual dense block are built to obtain image details. In addition, this study uses the Mish function as the activation function of the discriminator to solve the problems of neuron extreme, gradient explosion, and poor generalization ability of the model. Through simulation, the results show that the proposed method can achieve a peak signal-to-noise ratio (PSNR) of 30 dB compared with traditional image super-resolution reconstruction methods for synthetic aperture image. The method proposed has an improvement of 2 dB in the PSNR and 0.016 in structure similarity index measure compared with the original super-resolution GAN. Therefore, this method can effectively reduce the image distortion and improve the quality of image reconstruction.

An airborne radar forward-looking imaging algorithm based on multi-scale U-shaped networks

ABSTRACT Radar forward-looking imaging has gained increasingly significance in diverse applications, including battlefield reconnaissance, target surveillance, and precision guidance. While Synthetic Aperture Radar techniques are widely employed to achieve high azimuth resolution, the domain of forward-looking imaging has its challenges. These challenges arise due to the minimal Doppler frequency changes in forward-looking areas and the symmetrical Doppler history about flight path. These constraints negatively impact the radar’s ability to accurately detect and image targets in forward-looking areas. Consequently, there has been a growing interest in the development of novel approaches to these problems. Due to the application of deep learning, image super-resolution methods based on convolutional neural networks (CNNs) have received extensive attention and demonstrated good performance. This avenue of research provides fresh insights into improving the azimuth resolution of radar imaging. In this letter, we propose a novel convolutional neural network, called MSUS-Net (Multi-scale U-shaped Network), for radar forward-looking imaging. It combines asymmetric feature fusion with a unique connection structure to replace traditional layers. Moreover, different from conventional image restoration networks that use content loss, we use the L1 distance in the frequency domain, restoring missing high-frequency details. Simulated and real radar data validate that this approach offers robust recovery and better imaging performance.

SwinIBSR: Towards real-world infrared image super-resolution

Traditional infrared image super-resolution (SR) methods often fall short in real-world scenarios, particularly when dealing with images degraded by various forms of noise and artifacts. Addressing this gap, this paper presents SwinIBSR, a blind SR approach optimized for real-world infrared imaging conditions. SwinIBSR introduces a practical degradation model, specifically designed to simulate the complex degradation processes encountered in real-world infrared imaging. Additionally, a mixed training strategy using both infrared and visible light datasets greatly enhances the network’s generalization capabilities, leading to the production of high-quality SR images. Building upon the Transformer-based architecture of SwinIR, SwinIBSR is adept at learning intricate mappings from low-resolution to high-resolution images. Our comprehensive experiments validate the effectiveness of SwinIBSR, showcasing its outstanding performance in processing real-world infrared images and its notable superiority over existing methods regarding visual quality. This work contributes to the field of infrared image SR, offering insights in enhancing SR techniques for practical applications.

OmniSR-M: A Rock Sheet with a Multi-Branch Structure Image Super-Resolution Lightweight Method

With the rapid development of digital core technology, the acquisition of high-resolution rock thin section images has become crucial. Due to the limitation of optical principles, thin section imaging involves a contradiction between resolution and field of view. In order to solve this problem, this paper proposes a lightweight, fully aggregated network with multi-branch structure for super resolution of rock thin section images. The experimental results on the rock thin section dataset demonstrate that the improved method, called OmniSR-M, achieves significant enhancement compared to the original OmniSR method and also surpasses other state-of-the-art methods. OmniSR-M effectively recovers image details while maintaining its lightweight nature. Specifically, OmniSR-M reduces the number of parameters by 26.56% and the computation by 27.66% compared to OmniSR. Moreover, this paper quantitatively analyzes both the facies porosity rate and grain size features in the application scenario. The results show that the images generated by OmniSR-M successfully recover key information about the rock thin section.

Spherical Pseudo-Cylindrical Representation for Omnidirectional Image Super-resolution

Omnidirectional images have attracted significant attention in recent years due to the rapid development of virtual reality technologies. Equirectangular projection (ERP), a naive form to store and transfer omnidirectional images, however, is challenging for existing two-dimensional (2D) image super-resolution (SR) methods due to its inhomogeneous distributed sampling density and distortion across latitude. In this paper, we make one of the first attempts to design a spherical pseudo-cylindrical representation, which not only allows pixels at different latitudes to adaptively adopt the best distinct sampling density but also is model-agnostic to most off-the-shelf SR methods, enhancing their performances. Specifically, we start by upsampling each latitude of the input ERP image and design a computationally tractable optimization algorithm to adaptively obtain a (sub)-optimal sampling density for each latitude of the ERP image. Addressing the distortion of ERP, we introduce a new viewport-based training loss based on the original 3D sphere format of the omnidirectional image, which inherently lacks distortion. Finally, we present a simple yet effective recursive progressive omnidirectional SR network to showcase the feasibility of our idea. The experimental results on public datasets demonstrate the effectiveness of the proposed method as well as the consistently superior performance of our method over most state-of-the-art methods both quantitatively and qualitatively.

Noise-Free Optimization in Early Training Steps for Image Super-resolution

Recent deep-learning-based single image super-resolution (SISR) methods have shown impressive performance whereas typical methods train their networks by minimizing the pixel-wise distance with respect to a given high-resolution (HR) image. However, despite the basic training scheme being the predominant choice, its use in the context of ill-posed inverse problems has not been thoroughly investigated. In this work, we aim to provide a better comprehension of the underlying constituent by decomposing target HR images into two subcomponents: (1) the optimal centroid which is the expectation over multiple potential HR images, and (2) the inherent noise defined as the residual between the HR image and the centroid. Our findings show that the current training scheme cannot capture the ill-posed nature of SISR and becomes vulnerable to the inherent noise term, especially during early training steps. To tackle this issue, we propose a novel optimization method that can effectively remove the inherent noise term in the early steps of vanilla training by estimating the optimal centroid and directly optimizing toward the estimation. Experimental results show that the proposed method can effectively enhance the stability of vanilla training, leading to overall performance gain. Codes are available at github.com/2minkyulee/ECO.

MFTN: Multi-Level Feature Transfer Network Based on MRI-Transformer for MR Image Super-resolution

Due to the unique environment and inherent properties of magnetic resonance imaging (MRI) instruments, MR images typically have lower resolution. Therefore, improving the resolution of MR images is beneficial for assisting doctors in diagnosing the condition. Currently, the existing MR image super-resolution (SR) methods still have the problem of insufficient detail reconstruction. To overcome this issue, this paper proposes a multi-level feature transfer network (MFTN) based on MRI-Transformer to realize SR of low-resolution MRI data. MFTN consists of a multi-scale feature reconstruction network (MFRN) and a multi-level feature extraction branch (MFEB). MFRN is constructed as a pyramid structure to gradually reconstruct image features at different scales by integrating the features obtained from MFEB, and MFEB is constructed to provide detail information at different scales for low resolution MR image SR reconstruction by constructing multiple MRI-Transformer modules. Each MRI-Transformer module is designed to learn the transfer features from the reference image by establishing feature correlations between the reference image and low-resolution MR image. In addition, a contrast learning constraint item is added to the loss function to enhance the texture details of the SR image. A large number of experiments show that our network can effectively reconstruct high-quality MR Images and achieves better performance compared to some state-of-the-art methods. The source code of this work will be released on GitHub.

Enhancing Hyperspectral Images via Diffusion Model and Group-Autoencoder Super-resolution Network

Existing hyperspectral image (HSI) super-resolution (SR) methods struggle to effectively capture the complex spectral-spatial relationships and low-level details, while diffusion models represent a promising generative model known for their exceptional performance in modeling complex relations and learning high and low-level visual features. The direct application of diffusion models to HSI SR is hampered by challenges such as difficulties in model convergence and protracted inference time. In this work, we introduce a novel Group-Autoencoder (GAE) framework that synergistically combines with the diffusion model to construct a highly effective HSI SR model (DMGASR). Our proposed GAE framework encodes high-dimensional HSI data into low-dimensional latent space where the diffusion model works, thereby alleviating the difficulty of training the diffusion model while maintaining band correlation and considerably reducing inference time. Experimental results on both natural and remote sensing hyperspectral datasets demonstrate that the proposed method is superior to other state-of-the-art methods both visually and metrically.

SkipDiff: Adaptive Skip Diffusion Model for High-Fidelity Perceptual Image Super-resolution

It is well-known that image quality assessment usually meets with the problem of perception-distortion (p-d) tradeoff. The existing deep image super-resolution (SR) methods either focus on high fidelity with pixel-level objectives or high perception with generative models. The emergence of diffusion model paves a fresh way for image restoration, which has the potential to offer a brand-new solution for p-d trade-off. We experimentally observed that the perceptual quality and distortion change in an opposite direction with the increase of sampling steps. In light of this property, we propose an adaptive skip diffusion model (SkipDiff), which aims to achieve high-fidelity perceptual image SR with fewer sampling steps. Specifically, it decouples the sampling procedure into coarse skip approximation and fine skip refinement stages. A coarse-grained skip diffusion is first performed as a high-fidelity prior to obtaining a latent approximation of the full diffusion. Then, a fine-grained skip diffusion is followed to further refine the latent sample for promoting perception, where the fine time steps are adaptively learned by deep reinforcement learning. Meanwhile, this approach also enables faster sampling of diffusion model through skipping the intermediate denoising process to shorten the effective steps of the computation. Extensive experimental results show that our SkipDiff achieves superior perceptual quality with plausible reconstruction accuracy and a faster sampling speed.