Resolution Enhancement Method Research Articles

A new PDE-based resolution enhancement technique for the analysis of low SNR particle displacement images

Speckle images have found multiple applications in image-based measurement techniques, where the objective of measurement is the estimation of speckle displacements between two often consecutive images. Depending on the characteristics of a measurement system and its measurands, the signal-to-noise ratio (SNR) of recorded images may become low and even marginal. One solution to improve the accuracy of speckle displacement estimation in such cases is the use of super-resolution methods. In this article a robust and versatile partial differential equation based resolution enhancement method is presented for the analysis of low SNR speckle displacement signals. The application of the method to synthetic one-dimensional and two-dimensional signals as well as actual particle displacement images indicate that the method is capable of reducing noise effects and partially recovering lost correlation, which in turn makes the selection of smaller interrogation windows and achieving higher resolution and accuracy of displacement estimation possible. The method is applicable to both speckle and particle displacement images.

Image Spectral Resolution Enhancement for Mapping Native Plant Species in a Typical Area of the Three-River Headwaters Region, China

Large-scale multispectral remote sensing data are often unavailable for some practical applications. Spectral resolution enhancement for large-scale multispectral remote sensing images by incorporating small-scale hyperspectral remote sensing images is an alternative way to generate remote sensing images with both large spatial range and high spectral resolution. This paper proposes an improved spectral resolution enhancement method (ISREM) using spectral matrix and weighting the spectral angle of the transformation matrix. ISREM is tested in a typical area of the Three-River Headwaters region (TRHR) to produce a synthetic hyperspectral image (HSI). Two existing spectral resolution enhancement methods, the color resolution improvement software package (CRISP) and spectral resolution enhancement method (SREM), are adopted to compare with ISREM. To further test the practicality of the synthetic HSIs generated by the ISREM, CRISP and SREM, they are used to estimate the coverage of native plant species (NPS) using support vector machines (SVM) and random forest (RF) regressions. The experimental results are as follows. (1) For the Pearson correlation coefficient between the synthetic HSI and original image, ISREM yielded the largest value of 0.9582, followed by CRISP and SREM with values of 0.9480 and 0.9514. For spectral similarity, the HSI generated by the ISREM was the closest to the original reference HSI in the spectral curve. It also showed the best cumulative performance with the use of the three quality evaluation indexes. (2) The identification accuracies of native plant species were 93.51%, 90.91%, 89.61% and 89.61% using generated HSIs and original multispectral image (MSI) within a threshold of 20%, respectively. Compared with original MSI, the synthetic HSI showed better ability to identify NPS in the study area, which further illustrated the effectiveness of the ISREM. (3) The ISREM can reduce the strict requirement of pure pixels and maintain the quality of synthetic HSI by spectral angle weighting. Hence, the proposed ISREM outperforms the existing CRISP and SREM methods in image spectral resolution enhancement of multispectral remote sensing images.

A new resolution enhancement method for sandstone thin-section images using perceptual GAN

The thin section is a sliver cut of rock sample and photographed by a microscope to help researchers investigate the mineral, structure, and geology details of a rock sample. For the purpose of photographing using the microsctope, only a small portion of the thin section can be visualized. The results, however, can hardly satisfy the requirement of a high resolution and larger scale visualization. The thin section is then trapped into a dilemma of contradiction between resolution and field of view. To solve this problem, we propose the Generative Adversarial Networks（GAN） based architectures representing the perceptual SR in this paper. Four perceptual super-resolution methods and two pixel-wise super-resolution methods are trained and tested for comparison with the dataset of sandstone thin-section. The perceptual GAN based super-resolution method is demonstrated to compensate for this limitation by enhancing the resolution of a large field of view image, which will provide a higher resolution detail recovery and larger field of view simultaneously. Based on the digital experiment analysis and comparison, thin-section perceptual loss with Residual-in-Residual Dense Block (RRDB) and Relativistic average GANs (RaGAN) shows a more realistic texture and evident edge for minerals and pores than other methods. With the new perceptual GAN based image enhancement method, the petroleum geologist can significantly improve the accuracy and reliability of all their thin section based research quality.

Component Decomposition-Based Hyperspectral Resolution Enhancement for Mineral Mapping

Combining both spectral and spatial information with enhanced resolution provides not only elaborated qualitative information on surfacing mineralogy but also mineral interactions of abundance, mixture, and structure. This enhancement in the resolutions helps geomineralogic features such as small intrusions and mineralization become detectable. In this paper, we investigate the potential of the resolution enhancement of hyperspectral images (HSIs) with the guidance of RGB images for mineral mapping. In more detail, a novel resolution enhancement method is proposed based on component decomposition. Inspired by the principle of the intrinsic image decomposition (IID) model, the HSI is viewed as the combination of a reflectance component and an illumination component. Based on this idea, the proposed method is comprised of several steps. First, the RGB image is transformed into the luminance component, blue-difference and red-difference chroma components (YCbCr), and the luminance channel is considered as the illumination component of the HSI with an ideal high spatial resolution. Then, the reflectance component of the ideal HSI is estimated with the downsampled HSI image and the downsampled luminance channel. Finally, the HSI with high resolution can be reconstructed by utilizing the obtained illumination and the reflectance components. Experimental results verify that the fused results can successfully achieve mineral mapping, producing better results qualitatively and quantitatively over single sensor data.

Image resolution enhancement via sparse interpolation on wavelet domain

The image processing algorithms collectively known as super-resolution (SR) have proven effective in producing high-quality imagery from low-resolution (LR) images. This paper focuses on a novel image resolution enhancement method employing the wavelet domain techniques. In order to preserve more edge information, additional edge extraction step is proposed employing high-frequency (HF) sub-band images - low-high (LH), high-low (HL), and high-high (HH) - via the Discrete Wavelet Transform (DWT). In the designed procedure, the LR image is used in the sparse interpolation for the resolution-enhancement obtaining low-low (LL) sub- band. Additionally, all sub-bands (LL, LH, HL and HH) are performed via the Lanczos interpolation. Finally, the estimated sub-band images are used to form the new high-resolution (HR) image using the inverse DWT (IDWT). Experimental results on real data sets have confirmed the effectiveness of the proposed framework in terms of objective criteria as well as in subjective perception.

Monitoring early stage invasion of exotic Spartina alterniflora using deep-learning super-resolution techniques based on multisource high-resolution satellite imagery: A case study in the Yellow River Delta, China

Over the past decades, Spartina alterniflora, one of the top exotic invasive plants in China, has expanded throughout coastal China. In the Yellow River Delta (YRD), the rapid expansion of S. alterniflora has caused serious negative ecological effects. Current studies have concentrated primarily on mapping the distribution of S. alterniflora with medium-resolution satellite imagery at the regional or landscape scale, which have a limited capability in early detection and monitoring of the invasive process at the patch scale. In this study, we proposed a framework for monitoring the early stage invasion of S. alterniflora patches in the YRD using multiyear multisource high-spatial-resolution satellite imagery with various ground sampling distances (WorldView-2, SPOT-6, GaoFen-1, GaoFen-2, and GaoFen-6 from 2012 to 2019). First, we proposed to use deep-learning-based image super-resolution models to enhance all images to submeter (0.5 m) resolution. Then, we adopted stepwise evolution analysis-based image segmentation and object-based classification rules to detect and delineate S. alterniflora patches from the super-resolved imagery. By investigating Super-Resolution Convolutional Neural Networks (SRCNN) and Fast Super-Resolution Convolutional Neural Networks (FSRCNN) and comparing these methods with the conventional bicubic interpolation method for image resolution enhancement, we concluded that FSRCNN was superior in constructing spectral and structural details from the 1 m/1.5 m/2 m resolution images to 0.5 m resolution. FSRCNN, in particular, was more effective and efficient in discerning and estimating the size of small S. alterniflora patches (<50 m2). Using our method, 76 of 83 field-measured small patches were accurately detected and the delineated S. alterniflora patch perimeters agreed well with the field-measured patch perimeters (root mean square error [RMSE] = 8.29 m, mean absolute percentage error [MAPE] = 23.46 %). The invasion process showed fast expansion from 2012 to 2015 and slow growth from 2016 to 2019. We observed that the landward limits of S. alterniflora patches were influenced by elevation and vicinity to tidal creeks.

Structural Investigation of Single Specimens with a Femtosecond X-Ray Laser: Routes to Signal-to-Noise Ratio Enhancement

Interest in atomic scale structures of individual specimens has invigorated developments of high-resolution probes, which include single-particle imaging using x-ray free-electron lasers (XFELs). The demonstrated spatial resolution, however, remains at tens of nanometers with difficulty in collecting diffraction signals at high frequency distinguished from noises. As such, various resolution-enhancement methods have been introduced, but few experimental verifications are available. Here, by carrying out XFEL single-pulse diffraction experiments, we explicitly unveil the dependence of SNRs on incident x-ray flux, data averaging, or multiparticle interference. We further propose a data-accumulation method of resolution-shell averaging as a robust scheme to improve the SNR. This study establishes a roadmap with which high-resolution XFEL single-pulse experiments can be contrived.

Complexation-induced resolution enhancement of 3D-printed hydrogel constructs

Three-dimensional (3D) hydrogel printing enables production of volumetric architectures containing desired structures using programmed automation processes. Our study reports a unique method of resolution enhancement purely relying on post-printing treatment of hydrogel constructs. By immersing a 3D-printed patterned hydrogel consisting of a hydrophilic polyionic polymer network in a solution of polyions of the opposite net charge, shrinking can rapidly occur resulting in various degrees of reduced dimensions comparing to the original pattern. This phenomenon, caused by complex coacervation and water expulsion, enables us to reduce linear dimensions of printed constructs while maintaining cytocompatible conditions in a cell type-dependent manner. We anticipate our shrinking printing technology to find widespread applications in promoting the current 3D printing capacities for generating higher-resolution hydrogel-based structures without necessarily having to involve complex hardware upgrades or other printing parameter alterations.

Optimized Wavelet Decomposition and Gaussian interpolation based Satellite Image Enhancement

Satellite images (SI) play a vital role in various remote sensing applications like geoscience, geographical studies, observing the earth's atmosphere, monitoring natural disasters, etc. The SI are used in these applications require high-resolution. The performance of the wavelet transforms based resolution enhancement methods depends on the type of the mother wavelet used and it varies with image to image. The novel robust SI resolution enhancement technique including Optimized wavelet transform based image decomposition and Gaussian interpolation is proposed in this paper. Optimized wavelet decomposition is obtained using the Stochastic Diffusion Search algorithm and the Gaussian distribution function is used for interpolation. The proposed method is compared with the Discrete wavelet decomposition and Gaussian interpolation resolution enhancement method and proved that the proposed method gives the best results for any image.

An Adaptive Noise-Free Method of Seismic Resolution Enhancement Based on Extrapolated Multiresolution Singular Value Decomposition

The maximum vertical seismic resolution is a critical parameter in seismic data processing and interpretation. In general, conventional methods such as spectral whitening and zero-phase deconvolution assume randomness in the reflectivity spectrum that may not always be valid. The method proposed herein, extrapolated multiresolution singular value decomposition (EMRSVD), avoids some of these assumptions. We reasonably assume that original seismic data constitute a low-frequency component of high-resolution data, of which high-frequency component is attenuated. To recover this high-resolution component, EMRSVD adaptively extrapolates attenuated high-frequency information based on multiresolution singular value decomposition (MRSVD) and polynomial fitting to broaden the valid frequency bandwidth and improve the resolution of original seismic data. In our work, we first decompose a signal into a series of approximate and detailed subsignals exhibiting different resolutions using MRSVD. Then, we apply a polynomial fitting algorithm to the singular values corresponding to the detailed subsignals and obtain an expression of the singular values. Next, based on this expression, we extrapolate new singular values to construct new detailed subsignals. We repeat this process iteratively, where the number of iterations is determined by a modified variance model with an exponential transformation. Finally, we add these extrapolated detailed subsignals back into the original signal to obtain a high-resolution result. Examples with two synthetic signals and one field seismic data demonstrate that the proposed method significantly improves the resolution of seismic data without introducing noise and efficiently broadens the frequency bandwidth of reflection signals without sacrificing low-frequency information, which facilitates the resolution of thin layers.

Understanding and optimizing EBIC pn-junction characterization from modeling insights

In this paper, the physical mechanisms involved in electron-beam-induced current (EBIC) imaging of semiconductor pn-junctions are reviewed to propose a model and optimize the acquisition of experimental data. Insights are drawn on the dependence of the EBIC signal with electron accelerating voltage and surface conditions. It is concluded that improvements in the resolution of EBIC are possible when the surface conditions of the specimens are carefully considered and optimized. A lower accelerating voltage and an increase of the surface recombination velocities are quantitatively shown to maximize the EBIC lateral resolution in locating the pn-junction. The effect of surface band bending is included in the model, and it is seen to primarily affect the surface recombination. Introducing controlled surface damage is shown as a potential method for resolution enhancement via focused ion beam milling with Ga+ ions. These findings contribute to the understanding of this technique and can produce further improvements to its application in semiconductor device technology.

Resolution Enhancement in Directly Interfaced System for Inductive Sensors

In this paper, a resolution enhancement method for direct interfacing of inductive sensors, without employing analog-to-digital converter and/or external driver circuit, is presented. In order to achieve better resolution with minimum relative error and current consumption factors, different schemes are implemented which vary in terms of: 1) step-input voltage provision methodology, i.e., through microcontroller ( $\mu \text{C}$ ) pin or through external supply; 2) number of switches used; and 3) method of voltage detection, i.e., either using capture module or employing built-in comparator only. The maximum achievable resolution limits are explored by reducing the lower reference voltage ( $\text{V}_{\mathrm {Ref L}}$ ) in case of detection through built-in analog comparator and/or reducing the series resistance. The one-point calibration technique is used to determine the accuracy. The experiments are performed at 16-MHz clock frequency in the inductance range of: 1) 10–100 $\mu \text{H}$ and 2) 100–1000 $\mu \text{H}$ . The results show that in the limits of a maximum current sourcing/sinking capacity of a typical $\mu \text{C}$ , i.e., 40 mA, the proposed methodology enhances the resolution to $4~\mu \text{H}$ with a relative error of 1.2 in the percentage of full-scale span (%FSS). The maximum nonlinearity error in %FSS for inductance range 10–100 $\mu \text{H}$ and 100–1000 $\mu \text{H}$ is 1.1% and <0.4%, respectively.

Improved generative adversarial networks using the total gradient loss for the resolution enhancement of fluorescence images.

Because of the optical properties of medical fluorescence images (FIs) and hardware limitations, light scattering and diffraction constrain the image quality and resolution. In contrast to device-based approaches, we developed a post-processing method for FI resolution enhancement by employing improved generative adversarial networks. To overcome the drawback of fake texture generation, we proposed total gradient loss for network training. Fine-tuning training procedure was applied to further improve the network architecture. Finally, a more agreeable network for resolution enhancement was applied to actual FIs to produce sharper and clearer boundaries than in the original images.

Data extension near-field acoustic holography based on improved regularization method for resolution enhancement

In recent years, patch near-field acoustic holography (NAH) has gradually emerged for the reconstruction of sound field radiated from large-scale sound source under the condition of small holographic aperture in engineering practice. However, this technique has not solved the problem that when the projection of large-scale sound source is located on the edge of the expanded hologram, high-precision reconstruction results are unavailable. In addition, the reconstruction of sound field is a typical inverse problem, so the existence of measurement errors makes the solution of the inverse problem be ill-posed and the accuracy of sound field reconstruction is greatly reduced. To overcome these difficulties, a new technique combining an improved method of orthogonal spherical wave superposition, data extension and modified regularization method is proposed to enhance the spatial resolution of reconstructed images. First, the sound pressure is reconstructed on an extended hologram using the improved method orthogonal spherical wave superposition and data extension. At the same time, the method modified Tikhonov regularization (MTR) is introduced to solve the edge error and other ill-posed reconstruction problems. Through iteration the accuracy of the regenerated data is improved. Finally, the extended data are used to reconstruct the sound field. The results of numerical simulation and experiments show that this technique can effectively improve the spatial resolution of the reconstructed image and the reconstruction is more robust.

Spectral Super Resolution of Hyperspectral Images via Coupled Dictionary Learning

High-spectral resolution imaging systems play a critical role in the identification and characterization of objects in a scene of interest. Unfortunately, multiple factors impair spectral resolution, as in the case of modern snapshot spectral imagers that associate each hyperpixel with a specific spectral band. In this paper, we introduce a novel postacquisition computational technique aiming to enhance the spectral dimensionality of imaging systems by exploiting the mathematical frameworks of sparse representations and dictionary learning. We propose a coupled dictionary learning model which considers joint feature spaces, composed of low- and high-spectral resolution hypercubes, in order to achieve spectral superresolution performance. We formulate our spectral coupled dictionary learning optimization problem within the context of the alternating direction method of multipliers, and we manage to update the involved quantities via closed-form expressions. In addition, we consider a realistic spectral subsampling scenario, taking into account the spectral response functions of different satellites. Moreover, we apply our spectral superresolution algorithm on real satellite data acquired by Landsat-8 and Sentinel-2 sensors. Finally, we have investigated the problem of hyperspectral image unmixing using the recovered high-spectral resolution data cube, and we are able to demonstrate that the proposed scheme provides significant value in hyperspectral image understanding techniques. Experimental results demonstrate the ability of the proposed approach to synthesize high-spectral-resolution 3-D hypercubes, achieving better performance compared to state-of-the-art resolution enhancement methods.

Method for enhancement of spatial resolution of eddy current imaging

We propose a method for spatial resolution enhancement of eddy current images obtained by magnetic induction 2D scanning. A measurement technique based on a scanning differential source of an alternating magnetic field combined with a receiver is considered. A flat spiral coil is used as a receiver. The dimensions of the receiver and source coils are larger than or comparable with the dimensions of the objects under study. We represent an electrically conductive object as a set of small diameter conductive coils with a specific current induced by an external source. The method proposed solves the forward and inverse problems for linear coefficients describing the distribution of small coils representing a flat conductive object of arbitrary shape. This method was verified numerically and experimentally. The visualization of the object at a range of 15 mm with appropriate spatial resolution is shown experimentally at a frequency of 30 kHz. This method could be applied for the non-destructive testing and visualization of metallic objects.The novelty of the proposed method lies in the fact that it provides the ability to use probe coils with sizes much larger than inhomogeneities. We also proposed a method for calculating the hardware function of the system and solving the inverse problem. Detailed object visualization with sizes smaller than the probe system of the coils is provided. The possibility of increasing the resolution by seven times at a noise level of 0.1% is shown numerically.

Deep Learning-Based Super-Resolution Applied to Dental Computed Tomography

The resolution of dental computed tomography (CT) images is limited by detector geometry, sensitivity, patient movement, the reconstruction technique and the need to minimize radiation dose. Recently, the use of convolutional neural network (CNN) architectures has shown promise as a resolution enhancement method. In the current work, two CNN architectures—a subpixel network and the so called U-net—have been considered for the resolution enhancement of 2-D cone-beam CT image slices of ex vivo teeth. To do so, a training set of 5680 cross-sectional slices of 13 teeth and a test set of 1824 slices of 4 structurally different teeth were used. Two existing reconstruction-based super-resolution methods using $\boldsymbol {\ell _{2}}$ -norm and total variation regularization were used for comparison. The results were evaluated with different metrics (peak signal-to-noise ratio, structure similarity index, and other objective measures estimating human perception) and subsequent image-segmentation-based analysis. In the evaluation, micro-CT images were used as ground truth. The results suggest the superiority of the proposed CNN-based approaches over reconstruction-based methods in the case of dental CT images, allowing better detection of medically salient features, such as the size, shape, or curvature of the root canal.

Resolution-Enhancement for an Integral Imaging Microscopy Using Deep Learning

A novel resolution-enhancement method for an integral imaging microscopy that applies interpolation and deep learning is proposed, and the complete system with both hardware and software components is implemented. The resolution of the captured elemental image array is increased by generating intermediate-view elemental images between each neighboring elemental image, and an orthographic-view visualization of the specimen is reconstructed. Then, a deep learning algorithm is used to generate maximum possible resolution for each reconstructed directional-view image with improved quality. Since a pretrained model is applied, the proposed system processes the images directly without data training. The experimental results indicate that the proposed system produces resolution-enhanced directional-view images, and quantitative evaluation methods for reconstructed images such as the peak signal-to-noise ratio and the power spectral density confirm that the proposed system provides improvements in image quality.

High order derivative to investigate the complexity of the near infrared spectra of aqueous solutions

Derivative calculation is a powerful method for resolution enhancement in spectral analysis. A high order derivative method based on continuous wavelet transform (CWT) is discussed in the analysis of near infrared (NIR) spectra. The results for a simulated spectrum obtained from conventional numerical differentiation (NM), Fourier transform (FT), Savitzky-Golay (SG) and CWT method were compared. CWT method was found to be as efficient as FT and SG, but easier for high order derivative computation, and the fourth order derivative was proved to be a good choice for resolution enhancement as well as reduction of noise and sidelobe effects. For the NIR spectra of water-ethanol mixtures, the complexity of the spectra can be observed from the fourth derivative, including the spectral features of OH and CH with various intermolecular interactions. Fitting the derivative spectra of the mixtures by those of pure water and ethanol, the obtained coefficients for ethanol show a linear relation with the content but that for water exhibit a non-linear relation, which reveals the influence of ethanol on water structure in the mixture. Furthermore, the information of the water-ethanol clusters was found in the residual spectra after the fitting. Therefore, high order derivative can be an efficient way to improve the resolution of NIR spectra for understanding the interactions in aqueous solutions.

大视场高分辨率数字全息成像技术综述

As an interference imaging method, digital holography (DH) can accurately record the phase information of objects, and has the advantages of fast, non-destructive and three-dimensional imaging. It is widely used in the field of biological imaging and materials science. Like other optical imaging methods, DH also faces the problem that the resolution and the field of view(FOV) are mutually constrained, resulting in limited spatial bandwidth product(SBP). To solve this problem, researchers proposed methods such as computational illumination, computational modulation, and computational probing to extend SBP by sacrificing other degrees of freedom(such as time and polarization) of the imaging system. This paper firstly reviews the theoretical analysis of information capacity of an optical system. On this basis, we systematically summarize the high-resolution and large-FOV digital holographic imaging technology in recent years, introduce the principle and implementation of oblique illumination, structured illumination, random modulation illumination, multi-position synthetic aperture and pixel super-resolution method for resolution enhancement, and angle multiplexing method for FOV extension, and make a comparative study. The potential ways to improve resolution and expand FOV are also prospected.