Search Results (23)

Sparsity-driven regularization has undergone significant development in single-image restoration, particularly with the transition from handcrafted priors to trainable deep architectures. In this work, a geometric prior-enhanced deep image prior (DIP) framework, termed DIP-MC, is proposed that integrates mean curvature (MC) regularization to promote natural smoothness and structural coherence in reconstructed images. To strengthen the representational capacity of DIP, a self-attention module is incorporated between the encoder and decoder, enabling the network to capture long-range dependencies and preserve fine-scale textures. In contrast to total variation (TV), which frequently produces piecewise-constant artifacts and staircasing, MC regularization leverages curvature information, resulting in smoother transitions while maintaining sharp structural boundaries. DIP-MC is evaluated on standard grayscale and color image denoising and deblurring tasks using benchmark datasets including BSD68, Classic5, LIVE1, Set5, Set12, Set14, and the Levin dataset. Quantitative performance is assessed using peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) metrics. Experimental results demonstrate that DIP-MC consistently outperformed the DIP-TV baseline with 26.49 PSNR and 0.9 SSIM. It achieved competitive performance relative to BM3D and EPLL models with 28.6 PSNR and 0.87 SSIM while producing visually more natural reconstructions with improved detail fidelity. Furthermore, the learning dynamics of DIP-MC are analyzed by examining update-cost behavior during optimization, visualizing the best-performing network weights, and monitoring PSNR and SSIM progression across training epochs. These evaluations indicate that DIP-MC exhibits superior stability and convergence characteristics. Overall, DIP-MC establishes itself as a robust, scalable, and geometrically informed framework for high-quality single-image restoration. Full article

Phase images from gradient echo MRI sequences reflect underlying magnetic field inhomogeneities but are inherently wrapped within the range of −π to π, requiring phase unwrapping to recover the true phase. In this study, we present DIP-UP (Deep Image Prior for Unwrapping Phase), a framework designed to refine two pre-trained deep learning models for phase unwrapping: PHUnet3D and PhaseNet3D. We compared the DIP-refined models to their original versions, as well as to the conventional PRELUDE algorithm from FSL, using both simulated and in vivo brain data. Results demonstrate that DIP refinement improves unwrapping accuracy (achieving ~99%) and robustness to noise, surpassing the original networks and offering comparable performance to PRELUDE while being over three times faster. This framework shows strong potential for enhancing downstream MRI phase-based analyses. Full article

This study focuses on the problem of identifying and extracting the magnetopause boundary of the Earth’s magnetosphere using the Soft X-ray Imager (SXI) onboard the Solar Wind Magnetosphere Ionosphere Link Explorer (SMILE) mission. The SXI employs lobster-eye optics to perform panoramic imaging of the magnetosphere based on the Solar Wind Charge Exchange (SWCX) mechanism. However, several factors are expected to hinder future in-orbit observations, including the intrinsically low signal-to-noise ratio (SNR) of soft-X-ray emission, pronounced vignetting, and the non-uniform effective-area distribution of lobster-eye optics. These limitations could severely constrain the accurate interpretation of magnetospheric structures—especially the magnetopause boundary. To address these challenges, a boundary detection approach is developed that combines image calibration with denoising based on deep image prior (DIP). The method begins with calibration procedures to correct for vignetting and effective area variations in the SXI images, thereby restoring the accurate brightness distribution and improving spatial uniformity. Subsequently, a DIP-based denoising technique is introduced, which leverages the structural prior inherent in convolutional neural networks to suppress high-frequency noise without pretraining. This enhances the continuity and recognizability of boundary structures within the image. Experiments use ideal magnetospheric images generated from magnetohydrodynamic (MHD) simulations as reference data. The results demonstrate that the proposed method significantly improves the accuracy of magnetopause boundary identification under medium and high solar wind number density conditions (N = 10–20 cm⁻³). The extracted boundary curves consistently achieve a normalized mean squared error (NMSE) below 0.05 compared to the reference models. Additionally, the DIP-processed images show notable improvements in peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), indicating enhanced image quality and structural fidelity. This method provides adequate technical support for the precise extraction of magnetopause boundary structures in soft X-ray observations and holds substantial scientific and practical value. Full article

Three-dimensional (3D) cellular models, such as spheroids, serve as pivotal systems for understanding complex biological phenomena in histology, oncology, and tissue engineering. In response to the growing need for advanced imaging capabilities, we present a novel multi-modal Raman light sheet microscope designed to capture elastic (Rayleigh) and inelastic (Raman) scattering, along with fluorescence signals, in a single platform. By leveraging a shorter excitation wavelength (532 nm) to boost Raman scattering efficiency and incorporating robust fluorescence suppression, the system achieves label-free, high-resolution tomographic imaging without the drawbacks commonly associated with near-infrared modalities. An accompanying Deep Image Prior (DIP) seamlessly integrates with the microscope to provide unsupervised denoising and resolution enhancement, preserving critical molecular details and minimizing extraneous artifacts. Altogether, this synergy of optical and computational strategies underscores the potential for in-depth, 3D imaging of biomolecular and structural features in complex specimens and sets the stage for future advancements in biomedical research, diagnostics, and therapeutics. Full article

Advancements in Raman light sheet microscopy have provided a powerful, non-invasive, marker-free method for imaging complex 3D biological structures, such as cell cultures and spheroids. By combining 3D tomograms made by Rayleigh scattering, Raman scattering, and fluorescence detection, this modality captures complementary spatial and molecular data, critical for biomedical research, histology, and drug discovery. Despite its capabilities, Raman light sheet microscopy faces inherent limitations, including low signal intensity, high noise levels, and restricted spatial resolution, which impede the visualization of fine subcellular structures. Traditional enhancement techniques like Fourier transform filtering and spectral unmixing require extensive preprocessing and often introduce artifacts. More recently, deep learning techniques, which have shown great promise in enhancing image quality, face their own limitations. Specifically, conventional deep learning models require large quantities of high-quality, manually labeled training data for effective denoising and super-resolution tasks, which is challenging to obtain in multi-modal microscopy. In this study, we address these limitations by exploring advanced zero-shot and self-supervised learning approaches, such as ZS-DeconvNet, Noise2Noise, Noise2Void, Deep Image Prior (DIP), and Self2Self, which enhance image quality without the need for labeled and large datasets. This study offers a comparative evaluation of zero-shot and self-supervised learning methods, evaluating their effectiveness in denoising, resolution enhancement, and preserving biological structures in multi-modal Raman light sheet microscopic images. Our results demonstrate significant improvements in image clarity, offering a reliable solution for visualizing complex biological systems. These methods establish the way for future advancements in high-resolution imaging, with broad potential for enhancing biomedical research and discovery. Full article

High-resolution remote sensing of turbidity in the coastal environment with unmanned aerial vehicles (UAVs) can be adversely affected by the presence of obstructions of vessels and marine objects in images, which can introduce significant errors in turbidity modeling and predictions. This study evaluates the use of two deep-learning-based inpainting methods, namely, Decoupled Spatial–Temporal Transformer (DSTT) and Deep Image Prior (DIP), to recover the obstructed information. Aerial images of turbidity plumes in the coastal environment were first acquired using a UAV system with a multispectral sensor that included obstructions on the water surface at various obstruction percentages. The performance of the two inpainting models was then assessed through both qualitative and quantitative analyses of the inpainted data, focusing on the accuracy of turbidity retrieval. The results show that the DIP model performs well across a wide range of obstruction percentages from 10 to 70%. In comparison, the DSTT model produces good accuracy only with low percentages of less than 20% and performs poorly when the obstruction percentage increases. Full article

Model-based hyperspectral image (HSI) denoising methods have attracted continuous attention in the past decades, due to their effectiveness and interpretability. In this work, we aim at advancing model-based HSI denoising, through sophisticated investigation for both the fidelity and regularization terms, or correspondingly noise and prior, by virtue of several recently developed techniques. Specifically, we formulate a novel unified probabilistic model for the HSI denoising task, within which the noise is assumed as pixel-wise non-independent and identically distributed (non-i.i.d) Gaussian predicted by a pre-trained neural network, and the prior for the HSI image is designed by incorporating the deep image prior (DIP) with total variation (TV) and spatio-spectral TV. To solve the resulted maximum a posteriori (MAP) estimation problem, we design a Monte Carlo Expectation–Maximization (MCEM) algorithm, in which the stochastic gradient Langevin dynamics (SGLD) method is used for computing the E-step, and the alternative direction method of multipliers (ADMM) is adopted for solving the optimization in the M-step. Experiments on both synthetic and real noisy HSI datasets have been conducted to verify the effectiveness of the proposed method. Full article

Compared to supervised denoising models based on deep learning, the unsupervised Deep Image Prior (DIP) denoising approach offers greater flexibility and practicality by operating solely with the given noisy image. However, the random initialization of network input and network parameters in the DIP leads to a slow convergence during iterative training, affecting the execution efficiency heavily. To address this issue, we propose the Masked-Pre-training-Based Fast DIP (MPFDIP) Denoising Model in this paper. We enhance the classical Restormer framework by improving its Transformer core module and incorporating sampling, residual learning, and refinement techniques. This results in a fast network called FRformer (Fast Restormer). The FRformer model undergoes offline supervised training using the masked processing technique for pre-training. For a specific noisy image, the pre-trained FRformer network, with its learned parameters, replaces the UNet network used in the original DIP model. The online iterative training of the replaced model follows the DIP unsupervised training approach, utilizing multi-target images and an adaptive loss function. This strategy further improves the denoising effectiveness of the pre-trained model. Extensive experiments demonstrate that the MPFDIP model outperforms existing mainstream deep-learning-based denoising models in reducing Gaussian noise, mixed Gaussian–Poisson noise, and low-dose CT noise. It also significantly enhances the execution efficiency compared to the original DIP model. This improvement is mainly attributed to the FRformer network’s initialization parameters obtained through masked pre-training, which exhibit strong generalization capabilities for various types and intensities of noise and already provide some denoising effect. Using them as initialization parameters greatly improves the convergence speed of unsupervised iterative training in the DIP. Additionally, the techniques of multi-target images and the adaptive loss function further enhance the denoising process. Full article

This paper presents an extensive evaluation of the Deep Image Prior (DIP) technique for image inpainting on Synthetic Aperture Radar (SAR) images. SAR images are gaining popularity in various applications, but there may be a need to conceal certain regions of them. Image inpainting provides a solution for this. However, not all inpainting techniques are designed to work on SAR images. Some are intended for use on photographs, while others have to be specifically trained on top of a huge set of images. In this work, we evaluate the performance of the DIP technique that is capable of addressing these challenges: it can adapt to the image under analysis including SAR imagery; it does not require any training. Our results demonstrate that the DIP method achieves great performance in terms of objective and semantic metrics. This indicates that the DIP method is a promising approach for inpainting SAR images, and can provide high-quality results that meet the requirements of various applications. Full article

Image denoising poses a significant challenge in computer vision due to the high-level visual task’s dependency on image quality. Several advanced denoising models have been proposed in recent decades. Recently, deep image prior (DIP), using a particular network structure and a noisy image to achieve denoising, has provided a novel image denoising method. However, the denoising performance of the DIP model still lags behind that of mainstream denoising models. To improve the performance of the DIP denoising model, we propose a TripleDIP model with internal and external mixed images priors for image denoising. The TripleDIP comprises of three branches: one for content learning and two for independent noise learning. We firstly use a Transformer-based supervised model (i.e., Restormer) to obtain a pre-denoised image (used as external prior) from a given noisy image, and then take the noisy image and the pre-denoised image as the first and second target image, respectively, to perform the denoising process under the designed loss function. We add constraints between two-branch noise learning and content learning, allowing the TripleDIP to employ external prior while enhancing independent noise learning stability. Moreover, the automatic stop criterion we proposed prevents the model from overfitting the noisy image and improves the execution efficiency. The experimental results demonstrate that TripleDIP outperforms the original DIP by an average of 2.79 dB and outperforms classical unsupervised methods such as N2V by an average of 2.68 dB and the latest supervised models such as SwinIR and Restormer by an average of 0.63 dB and 0.59 dB on the Set12 dataset. This can mainly be attributed to the fact that two-branch noise learning can obtain more stable noise while constraining the content learning branch’s optimization process. Our proposed TripleDIP significantly enhances DIP denoising performance and has broad application potential in scenarios with insufficient training datasets. Full article

Deep image prior (DIP) is a powerful technique for image restoration that leverages an untrained network as a handcrafted prior. DIP can also be used for hyperspectral image (HSI) denoising tasks and has achieved impressive performance. Recent works further incorporate different regularization terms to enhance the performance of DIP and successfully show notable improvements. However, most DIP-based methods for HSI denoising rarely consider the distribution of complicated HSI mixed noise. In this paper, we propose the asymmetric Laplace noise modeling deep image prior (ALDIP) for HSI mixed noise removal. Based on the observation that real-world HSI noise exhibits heavy-tailed and asymmetric properties, we model the HSI noise of each band using an asymmetric Laplace distribution. Furthermore, in order to fully exploit the spatial–spectral correlation, we propose ALDIP-SSTV, which combines ALDIP with a spatial–spectral total variation (SSTV) term to preserve more spatial–spectral information. Experiments on both synthetic data and real-world data demonstrate that ALDIP and ALDIP-SSTV outperform state-of-the-art HSI denoising methods. Full article

A Multiple-Input Multiple-Output (MIMO) Frequency-Modulated Continuous Wave (FMCW) radar can provide a range-angle map that expresses the signal power against each range and angle. It is possible to estimate object locations by detecting the signal power that exceeds a threshold using an algorithm, such as Constant False Alarm Rate (CFAR). However, noise and multipath components often exist over the range-angle map, which could produce false alarms for an undesired location depending on the threshold setting. In other words, the threshold setting is sensitive in noisy range-angle maps. Therefore, if the noise is reduced, the threshold can be easily set to reduce the number of false alarms. In this paper, we propose a method that improves the CFAR threshold tolerance by denoising a range-angle map using Deep Image Prior (DIP). DIP is an unsupervised deep-learning technique that enables image denoising. In the proposed method, DIP is applied to the range-angle map calculated by the Curve-Length (CL) method, and then the object location is detected over the denoised range-angle map based on Cell-Averaging CFAR (CA-CFAR), which is a typical threshold setting algorithm. Through the experiments to estimate human locations in indoor environments, we confirmed that the proposed method with DIP reduced the number of false alarms and estimated the human location accurately while improving the tolerance of the threshold setting, compared to the method without DIP. Full article

Supervised image denoising methods based on deep neural networks require a large amount of noisy-clean or noisy image pairs for network training. Thus, their performance drops drastically when the given noisy image is significantly different from the training data. Recently, several unsupervised learning models have been proposed to reduce the dependence on training data. Although unsupervised methods only require noisy images for learning, their denoising effect is relatively weak compared with supervised methods. This paper proposes a two-stage unsupervised deep learning framework based on deep image prior (DIP) to enhance the image denoising performance. First, a two-target DIP learning strategy is proposed to impose a learning restriction on the DIP optimization process. A cleaner preliminary image, together with the given noisy image, was used as the learning target of the two-target DIP learning process. We then demonstrate that adding an extra learning target with better image quality in the DIP learning process is capable of constraining the search space of the optimization process and improving the denoising performance. Furthermore, we observe that given the same network input and the same learning target, the DIP optimization process cannot generate the same denoised images. This indicates that the denoised results are uncertain, although they are similar in image quality and are complemented by local details. To utilize the uncertainty of the DIP, we employ a supervised denoising method to preprocess the given noisy image and propose an up- and down-sampling strategy to produce multiple sampled instances of the preprocessed image. These sampled instances were then fed into multiple two-target DIP learning processes to generate multiple denoised instances with different image details. Finally, we propose an unsupervised fusion network that fuses multiple denoised instances into one denoised image to further improve the denoising effect. We evaluated the proposed method through extensive experiments, including grayscale image denoising, color image denoising, and real-world image denoising. The experimental results demonstrate that the proposed framework outperforms unsupervised methods in all cases, and the denoising performance of the framework is close to or superior to that of supervised denoising methods for synthetic noisy image denoising and significantly outperforms supervised denoising methods for real-world image denoising. In summary, the proposed method is essentially a hybrid method that combines both supervised and unsupervised learning to improve denoising performance. Adopting a supervised method to generate preprocessed denoised images can utilize the external prior and help constrict the search space of the DIP, whereas using an unsupervised method to produce intermediate denoised instances can utilize the internal prior and provide adaptability to various noisy images of a real scene. Full article

Single-pixel imaging (SPI) is a promising imaging scheme based on compressive sensing. However, its application in high-resolution and real-time scenarios is a great challenge due to the long sampling and reconstruction required. The Deep Learning Compressed Network (DLCNet) can avoid the long-time iterative operation required by traditional reconstruction algorithms, and can achieve fast and high-quality reconstruction; hence, Deep-Learning-based SPI has attracted much attention. DLCNets learn prior distributions of real pictures from massive datasets, while the Deep Image Prior (DIP) uses a neural network′s own structural prior to solve inverse problems without requiring a lot of training data. This paper proposes a compressed reconstruction network (DPAP) based on DIP for Single-pixel imaging. DPAP is designed as two learning stages, which enables DPAP to focus on statistical information of the image structure at different scales. In order to obtain prior information from the dataset, the measurement matrix is jointly optimized by a network and multiple autoencoders are trained as regularization terms to be added to the loss function. Extensive simulations and practical experiments demonstrate that the proposed network outperforms existing algorithms. Full article

In this paper, we propose a deep-image-prior-based demosaicing method for a random RGBW color filter array (CFA). The color reconstruction from the random RGBW CFA is performed by the deep image prior network, which uses only the RGBW CFA image as the training data. To our knowledge, this work is a first attempt to reconstruct the color image with a neural network using only a single RGBW CFA in the training. Due to the White pixels in the RGBW CFA, more light is transmitted through the CFA than in the case with the conventional RGB CFA. As the image sensor can detect more light, the signal-to-noise-ratio (SNR) increases and the proposed demosaicing method can reconstruct the color image with a higher visual quality than other existing demosaicking methods, especially in the presence of noise. We propose a loss function that can train the deep image prior (DIP) network to reconstruct the colors from the White pixels as well as from the red, green, and blue pixels in the RGBW CFA. Apart from using the DIP network, no additional complex reconstruction algorithms are required for the demosaicing. The proposed demosaicing method becomes useful in situations when the noise becomes a major problem, for example, in low light conditions. Experimental results show the validity of the proposed method for joint demosaicing and denoising. Full article