Spatial Domain Model Research Articles

Image restoration in frequency space using complex-valued CNNs.

Real-valued convolutional neural networks (RV-CNNs) in the spatial domain have outperformed classical approaches in many image restoration tasks such as image denoising and super-resolution. Fourier analysis of the results produced by these spatial domain models reveals the limitations of these models in properly processing the full frequency spectrum. This lack of complete spectral information can result in missing textural and structural elements. To address this limitation, we explore the potential of complex-valued convolutional neural networks (CV-CNNs) for image restoration tasks. CV-CNNs have shown remarkable performance in tasks such as image classification and segmentation. However, CV-CNNs for image restoration problems in the frequency domain have not been fully investigated to address the aforementioned issues. Here, we propose several novel CV-CNN-based models equipped with complex-valued attention gates for image denoising and super-resolution in the frequency domains. We also show that our CV-CNN-based models outperform their real-valued counterparts for denoising super-resolution structured illumination microscopy (SR-SIM) and conventional image datasets. Furthermore, the experimental results show that our proposed CV-CNN-based models preserve the frequency spectrum better than their real-valued counterparts in the denoising task. Based on these findings, we conclude that CV-CNN-based methods provide a plausible and beneficial deep learning approach for image restoration in the frequency domain.

Comparison of Different Impact Factors and Spatial Scales in PM2.5 Variation

PM2.5 particles with an aerodynamic diameter of less than 2.5 μm are receiving increasing attention in China. Understanding how complex factors affect PM2.5 particles is crucial for the prevention of air pollution. This study investigated the influence of meteorological factors and land use on the dynamics of PM2.5 concentrations in four urban agglomerations of China at different scales from 2010 to 2020, using the Durbin spatial domain model (SDM) at five different grid scales. The results showed that the average annual PM2.5 concentration in four core urban agglomerations in China generally had a downward trend, and the meteorological factors and land use types were closely related to the PM2.5 concentration. The impact of temperature on PM2.5 changed significantly with an increase in grid scale, while other factors did not lead to obvious changes. The direct and spillover effects of different factors on PM2.5 in inland and coastal urban agglomerations were not entirely consistent. The influence of wind speed on coastal urban clusters (the Pearl River urban agglomeration (PRD) and Yangtze River urban agglomeration (YRD)) was not significant among the meteorological factors, but it had a significant impact on inland urban clusters (the Beijing–Tianjin–Hebei urban agglomeration (BTH) and Chengdu–Chongqing urban agglomeration (CC)). The direct effect of land use type factors showed an obvious U-shaped change with an increase in the research scale in the YRD, and the direct effect of land use type factors was almost twice as large as the spillover effect. Among land use type factors, human factors (impermeable surfaces) were found to have a greater impact in inland urban agglomerations, while natural factors (forests) had a greater impact in coastal urban agglomerations. Therefore, targeted policies to alleviate PM2.5 should be formulated in inland and coastal urban agglomerations, combined with local climate measures such as artificial precipitation, and urban land planning should be carried out under the consideration of known impacts.

MSBR-GNet: A High-Resolution Imagery Generative Optimization Model for Building Rooftop Boundary Guided by Interpretable Statistical Model in Spatial and Spectral Domain

MSBR-GNet: A High-Resolution Imagery Generative Optimization Model for Building Rooftop Boundary Guided by Interpretable Statistical Model in Spatial and Spectral Domain

Objective Image Quality Optimization in Augmented Reality Using Spatial Frequency Domain Models.

Augmented reality (AR) blends the digital and physical worlds by overlapping a virtual image onto the see-through physical environment. However, contrast reduction and noise superposition in an AR head-mounted display (HMD) can substantially limit image quality and human perceptual performance in both the digital and physical spaces. To assess image quality in AR, we performed human and model observer studies for various imaging tasks with targets placed in the digital and physical worlds. A target detection model was developed for the complete AR system including the optical see-through. Target detection performance using different observer models developed in the spatial frequency domain was compared with the human observer results. The non-prewhitening model with eye filter and internal noise results closely track human perception performance as measured by the area under the receiver operating characteristic curve (AUC), especially for tasks with high image noise. The AR HMD non-uniformity limits the low-contrast target (less than 0.02) observer performance for low image noise. In augmented reality conditions, the detectability of a target in the physical world is reduced due to the contrast reduction by the overlaid AR display image (AUC less than 0.87 for all the contrast levels evaluated). We propose an image quality optimization scheme to optimize the AR display configurations to match observer detection performance for targets in both the digital and physical worlds. The image quality optimization procedure is validated using both simulation and bench measurements of a chest radiography image with digital and physical targets for various imaging configurations.

Uncertainty-guided hierarchical frequency domain Transformer for image restoration

Existing convolutional neural network (CNN)-based and vision Transformer (ViT)-based image restoration methods are usually explored in the spatial domain. However, we employ Fourier analysis to show that these spatial domain models cannot perceive the entire frequency spectrum of images, i.e., mainly focus on either high-frequency (CNN-based models) or low-frequency components (ViT-based models). This intrinsic limitation results in the partial missing of semantic information and the appearance of artifacts. To address this limitation, we propose a novel uncertainty-guided hierarchical frequency domain Transformer named HFDT to effectively learn both high and low-frequency information while perceiving local and global features. Specifically, to aggregate semantic information from various frequency levels, we propose a dual-domain feature interaction mechanism, in which the global frequency information and local spatial features are extracted by corresponding branches. The frequency domain branch adopts the Fast Fourier Transform (FFT) to convert the features from the spatial domain to the frequency domain, where the global low and high-frequency components are learned with Log-linear complexity. Complementarily, an efficient convolution group is employed in the spatial domain branch to capture local high-frequency details. Moreover, we introduce an uncertainty degradation-guided strategy to efficiently represent degraded prior information, rather than simply distinguishing degraded/non-degraded regions in binary form. Our approach achieves competitive results in several degraded scenarios, including rain streaks, raindrops, motion blur, and defocus blur.

Machine learning based analysis of stroke lesions on mouse tissue sections

An unbiased, automated and reliable method for analysis of brain lesions in tissue after ischemic stroke is missing. Manual infarct volumetry or by threshold-based semi-automated approaches is laborious, and biased to human error or biased by many false -positive and -negative data, respectively. Thereby, we developed a novel machine learning, atlas-based method for fully automated stroke analysis in mouse brain slices stained with 2% Triphenyltetrazolium-chloride (2% TTC), named “StrokeAnalyst”, which runs on a user-friendly graphical interface. StrokeAnalyst registers subject images on a common spatial domain (a novel mouse TTC- brain atlas of 80 average mathematical images), calculates pixel-based, tissue-intensity statistics (z-scores), applies outlier-detection and machine learning (Random-Forest) models to increase accuracy of lesion detection, and produces volumetry data and detailed neuroanatomical information per lesion. We validated StrokeAnalyst in two separate experimental sets using the filament stroke model. StrokeAnalyst detects stroke lesions in a rater-independent and reproducible way, correctly detects hemispheric volumes even in presence of post-stroke edema and significantly minimizes false-positive errors compared to threshold-based approaches (false-positive rate 1.2–2.3%, p < 0.05). It can process scanner-acquired, and even smartphone-captured or pdf-retrieved images. Overall, StrokeAnalyst surpasses all previous TTC-volumetry approaches and increases quality, reproducibility and reliability of stroke detection in relevant preclinical models.

Estimation of slope length gradient (LS) factor for the sub-watershed areas of Juri River in Tripura

The slope length gradient (LS) factor expresses the effect of topography, i.e., combining effects of slope length (L) and slope steepness (S) on rate of soil erosion. The longer the slope length, greater is the amount of cumulative runoff and similarly steeper the slope of the land the higher is the velocities of the runoff that contributes erosion. The LS factor is the most important parameter for estimation of soil loss and soil management using Universal Soil Loss Equation (USLE) and Revised Universal Soil Loss Equation (RUSLE). A precise inventory and assessment of vulnerable areas includes the formulation of correct soil management for property development. Juri is a peri-urban watershed situated in Tripura, India, having high undulating topography, which leads to increase soil erosion and land slide problem. Therefore, it has become essential to quantify the LS factor to understand the characteristics of the topographic factor, which will help the researchers for estimation of soil erosion and preparation of management plan. The present study focused to evaluate the LS factor for the sub-watershed areas of Juri River of Tripura. Two methods namely (1) suggested by Moore and Burch (1986) and (2) proposed by Wischmeier and Smith (1978) were used to calculate the LS factor. Both the methods use Digital Elevation Model (DEM) in spatial domain to estimate the LS factor. The LS factor of the study area ranges from 0 to 982.71 by method-1 and 0.66 to 285.58 by method-2. The study indicates that, the method-2 is more apposite than method-1 because it gives LS factor value, which distributed uniformly in spatial domain; whereas method-1 resulted highest LS factor value along the flow direction and it had almost same value of LS factor in the remaining areas. This study will help for prediction of soil erosion and preparation of management plan by reducing the slope length of the study area. This findings and the methodology employed can be widely used in similar mountainous to hilly watersheds around the world for calculation of soil loss.

Uncertainty assessment of spatial domain models in early stage mining projects – A review

• Performance of feasibility studies in mining projects is poor which could be due to omission of quantitative uncertainty assessments in public reporting. • JORC 2012 standards require quantitative uncertainty assessment for geological and spatial domain models. • Current methodologies of quantitative uncertainty assessment and communication exist and can be used in early-stage mining projects. • Professional development and incorporating software and process workflows could be the way to go to improve current reporting practice. Sustainability of mining projects is linked to informed investment decisions based on public reporting of exploration and mineral resource estimation results. In Australia, guidelines for public reporting are established by the Joint Ore Reserves Committee reporting code through the JORC Code (2012). Although assessment of uncertainty in the results reported is a requirement, this is often communicated in a qualitative manner and subjectively evaluated. This can become a liability if not communicated effectively, particularly in early stages of mining projects when spatial domains of geological interpretation and mineralization envelopes influence reliability of resource estimates. This review describes methodologies for quantitative uncertainty assessment and communication, and explores how they could be applied in public reporting practice. The complexity, cost and additional work of doing a quantitative assessment could hinder a straightforward implementation. This could be overcome if mining companies budget for quantitative uncertainty assessment and associated professional development. A compulsory requirement for inclusion of uncertainty assessments in public reporting or use of standardized subjective language would improve industry practice.

A fast and effective image steganalysis model based on convolutional neural network

In recent years, the performance of deep learning in image steganalysis applications has become more and more outstanding, but at the same time the training time has also greatly increased. Some models need to be trained for several days, and the research efficiency is very low. In this article, we propose an image steganalysis model in spatial domain based on a three-layer convolutional neural network. The model does not use a pooling layer, and uses the global average pooling layer instead of the fully connected layer. Experimental results show that the training time of the model is greatly shortened, and the accuracy of detecting the three steganography algorithms with an embedding rate of 0.4bpp exceeds 85%.

3D Gravity Inversion on Unstructured Grids

Compared with structured grids, unstructured grids are more flexible to model arbitrarily shaped structures. However, based on unstructured grids, gravity inversion results would be discontinuous and hollow because of cell volume and depth variations. To solve this problem, we first analyzed the gradient of objective function in gradient-based inversion methods, and a new gradient scheme of objective function is developed, which is a derivative with respect to weighted model parameters. The new gradient scheme can more effectively solve the problem with lacking depth resolution than the traditional inversions, and the improvement is not affected by the regularization parameters. Besides, an improved fuzzy c-means clustering combined with spatial constraints is developed to measure property distribution of inverted models in both spatial domain and parameter domain simultaneously. The new inversion method can yield a more internal continuous model, as it encourages cells and their adjacent cells to tend to the same property value. At last, the smooth constraint inversion, the focusing inversion, and the improved fuzzy c-means clustering inversion on unstructured grids are tested on synthetic and measured gravity data to compare and demonstrate the algorithms proposed in this paper.

The Frequency Discrepancy Between Real and Generated Images

Despite the success of Generative Adversarial Networks (GANs), little work has focused on the discrepancy between real and generated images in frequency domain. In this work, we provide a systematic analysis on this topic. We first demonstrate the general existence of the frequency discrepancy and further perform extensive experiments both on datasets with various frequency distributions and models with different upsampling methods to reveal the sources of the discrepancy. Experimental results show that: resize-convolution is not a perfect alternative to deconvolution, and natural images and unnatural images should be treated separately during training. Based on these studies, we provide some novel solutions to reduce the discrepancy. Finally, we further show the effectiveness of our solutions on Variational Auto Encoders (VAEs). We hope that the community should pay equal attention to the performance of generative models both in spatial and frequency domain.

Quality factor estimation of JPEG images using a statistical model

This paper proposes a new approach to quality factor identification for a JPEG images. This approach is based on a spatial domain model of the variance of 8×8 blocks of JPEG images. The variance model follows a Gamma distribution, which is characterized by two parameters. These parameters are considered as a unique fingerprint for determining the value of the quality factor. To demonstrate the benefits of this approach, a likelihood ratio (LR) is established by exploiting the variance distribution. This ratio enables the quality factor to be obtained. Experiments on simulated images and large image databases with different image sizes and quality factors highlight the high accuracy of the proposed method.

Study on Image De-Noising and Technique

Noise in an image may be unwanted signals, elements or any natural involvement. De-noising the noisy image is a very crucial task, for which there are number of different methods, algorithms and transforms available. In this paper, a review of different types of image noise and methods of de-noising has been discussed. The merits and demerits of the techniques have also been discussed. The filters and domains for different types of noise is discuss like in Salt and Pepper NSM Filter show good result, Gaussian noise Linear and Non-Filter, Speckle noise spatial and frequency domain and Poisson ROF model use. Comparison of domain, filter and their result is done in different types of noise. The focus of the work is to identify the advantages and disadvantages of different algorithms.

Estimation of mechanical material properties using an isotropic three-dimensional elastic plate model

This paper develops an inverse method to simultaneously estimate complex Young’s and shear moduli of a rectangular-shaped solid material. This new method exploits a relatively simple experiment and is designed for a frequency range of approximately 1 kHz–10 kHz. A fully elastic spatial domain model of a plate is derived and corresponding experimental measurements are collected. An error function is defined using the residual between the model displacements and the measured displacements, which allows a metric for a search over a space where the complex model values of Young’s and shear moduli are allowed to vary. When this error function is minimized, the parameters used to create the model are considered the best estimate possible. An experiment is included to illustrate the estimation process, and it is shown that the real parts of both moduli compare favorably to previously available measurements. Finally, the model is compared to the experimental measurements to illustrate its accuracy when it is formulated using the material property estimates obtained using this method.

Markov Random Field Based Model In Spatial Domain for Denoising of Microarray Images

International Journal of Computer Sciences and Engineering (A UGC Approved and indexed with DOI, ICI and Approved, DPI Digital Library) is one of the leading and growing open access, peer-reviewed, monthly, and scientific research journal for scientists, engineers, research scholars, and academicians, which gains a foothold in Asia and opens to the world, aims to publish original, theoretical and practical advances in Computer Science,Information Technology, Engineering (Software, Mechanical, Civil, Electronics & Electrical), and all interdisciplinary streams of Computing Sciences. It intends to disseminate original, scientific, theoretical or applied research in the field of Computer Sciences and allied fields. It provides a platform for publishing results and research with a strong empirical component. It aims to bridge the significant gap between research and practice by promoting the publication of original, novel, industry-relevant research.

Spatial and Time Domain Model for LAA/Wi-Fi Coexistence

In Release 13 of the 3rd Generation Partnership Project standard, the deployment of long-term evolution (LTE)/ LTE-advanced to the unlicensed spectrum called licensed-assisted access (LAA) was introduced to satisfy the high demand of traffic data with a new type of services mostly multimedia streams. The unlicensed spectrum is a shared resource between different radio access terminals such as Wi-Fi and radar system, and listen before talk (LBT) in LTE side makes this coexistence possible. In this letter, we propose a new framework to model and evaluate the performance of LAA/Wi-Fi coexistence based on marked Poisson point process (MPPP) and Markov chain model. The proposed model captures the back-off procedure in the spatial and time domain to provide a more realistic analysis of the channel busy probability and the signal to interference plus noise ratio coverage probability. Numerical results show the validity of the proposed model with the overestimation of the LAA and Wi-Fi nodes density in MPPP model.

Watermarking Techniques for Three Dimensional (3D) Mesh Authentication in Spatial Domain

Due to lightning growth in multimedia processing technologies and frequent usage of Internet, media contents often get exposed to malicious interventions. Hence it is very important to verify the authenticity and integrity of the digital contents by using proper security mechanisms. Thanks to the advancement in computer graphics processing technologies due to which higher dimensional rendering of real life objects has become apparent. Lately, three dimensional (3D) representations are more preferred in different application domains for better visualization and understanding. While talking about security related to 3D models, digital watermarking has come out to be an effective and prominent solution among various other authentication mechanisms. The purpose of this study is to present an in-depth review and investigation of the authentication based watermarking techniques for 3D models in spatial domain. The methods are classified based on various attributes that effect the design and working of an authentication protocol. The study also discusses different types of geometrical attacks that may occur to a 3D model and methodologies that help in detecting and locating the attacks. The paper concludes with a set of remarks and some of the research issues and challenges which are needed to be studied further.

Multisource learning for skeleton-based action recognition using deep LSTM and CNN

Human action recognition based on information concerning the skeleton of the human body has been widely studied because such information can simply and clearly express features related to actions, and is unaffected by physical features of the body. Therefore, this paper discusses action recognition based on three-dimensional skeletal information obtained from RGB-D videos. We propose a multisource action recognition model that combines features of the temporal and spatial domains. Our model focuses on action features from three levels: global-level, local-level, and detail-level, since different actions concern different parts of the body. For temporal features, we adopt long short-term memory to create a model that analyzes skeleton sequence. For features of the spatial domain, we analyze the effects of three features on action recognition: skeleton joint coordinate, pairwise relative position, and speed of movement. Finally, the temporal and spatial domain models are combined into a multisource model for improving the accuracy of action recognition. Experiments show that our model brings about considerable improvement in the recognition of a variety of general actions and interactive activities.

A Comprehensive Study of Retinal Vessel Classification Methods in Fundus Images

Nowadays, it is obvious that there is a relationship between changes in the retinal vessel structure and diseases such as diabetic, hypertension, stroke, and the other cardiovascular diseases in adults as well as retinopathy of prematurity in infants. Retinal fundus images provide non-invasive visualization of the retinal vessel structure. Applying image processing techniques in the study of digital color fundus photographs and analyzing their vasculature is a reliable approach for early diagnosis of the aforementioned diseases. Reduction in the arteriolar-venular ratio of retina is one of the primary signs of hypertension, diabetic, and cardiovascular diseases which can be calculated by analyzing the fundus images. To achieve a precise measuring of this parameter and meaningful diagnostic results, accurate classification of arteries and veins is necessary. Classification of vessels in fundus images faces with some challenges that make it difficult. In this paper, a comprehensive study of the proposed methods for classification of arteries and veins in fundus images is presented. Considering that these methods are evaluated on different datasets and use different evaluation criteria, it is not possible to conduct a fair comparison of their performance. Therefore, we evaluate the classification methods from modeling perspective. This analysis reveals that most of the proposed approaches have focused on statistics, and geometric models in spatial domain and transform domain models have received less attention. This could suggest the possibility of using transform models, especially data adaptive ones, for modeling of the fundus images in future classification approaches.

A novel combination of second-order statistical features and segmentation using multi-layer superpixels for salient object detection

Salient object detection is one of the outstanding capabilities of the human visual system (HVS). The researcher community aims at developing a salient object detection model that matches the detection accuracy as well as computation time taken by the humans. These models can be developed in either spatial domain or frequency domain. Spatial domain models provide good detection accuracy at the cost of high computational time while frequency domain models offer fast computational speed to meet real-time requirements at the cost of poor detection accuracy. In order to induce a trade-off between computational time and accuracy, we propose a model which provides high detection accuracy without taking much of computation time. To detect the salient object with an accurate shape, we first segment the given image by utilizing a bipartite graph partitioning approach which aggregates multi-layer superpixels in a principled and effective manner. Second, the saliency of each segmented region is computed based on a hypercomplex Fourier transform (HFT) saliency map reconstructed using amplitude spectrum, filtered at an appropriate scale chosen using statistical features extracted from grey- level co-occurrence matrix and original phase spectrum. Finally, a saliency map is generated by taking average of the HFT coefficients of each region in the segmented image and then using the average HFT intensity value of the entire image as a threshold to clearly separate salient object from the background. The performance of the proposed model is evaluated in terms of F---measure, area under curve (AUC), and computation time using six publicly available image datasets. Both qualitative and quantitative evaluations on six publicly available datasets demonstrate the robustness and efficiency of the proposed model against twenty popular state-of-the-art methods.