Estimation Of Distribution Research Articles

Distance-aware network for physical-world object distribution estimation and counting

Crowd counting has become popular due to its applications in congested scenes. Current methods excel with specialized datasets but often ignore the density distribution affected by perspective. To investigate the true distribution of crowd density in the physical world, this paper introduces a framework leveraging head size variations to estimate real crowd distribution and count crowds. First, a convolutional neural network is designed to generate predicted density maps and corresponding bounding boxes across different datasets. Second, a filter kernel is employed to identify the most crowded area in the input image based on the obtained boxes. Finally, the vanishing point is computed by calculating the intersection of the two generated lines, which is then used to obtain the inverse perspective of the predicted density map. The Grid Mean Relative Error (GMRE) metric is proposed for evaluating transformation accuracy. In comparison with the Grid Average Mean Absolute Error (GAME), GMRE is distance-aware and more suitable for evaluating differences between distinct coordinate systems. Additionally, extensive experiments are conducted to validate the counting capabilities of the proposed network. Experimental results show the network’s competitive counting ability and reduced transformation error.

Advancing unsupervised anomaly detection with normalizing flow and multi-scale ensemble learning

Visual anomaly detection plays a crucial role in manufacturing to ensure product quality by identifying image patterns that deviate from the expected ones. Existing methods that rely on distribution estimation struggle with the complexity of real-world images, resulting in complex and inefficient procedures. This study leverages normalizing flow techniques to address the cold start anomaly detection problem, where no prior examples of anomalies are available during the training phase. In such scenarios, models must learn exclusively from defect-free images and still accurately identify anomalies. We propose a novel unsupervised multi-scale and multi-semantic normalizing flow model, enhanced with an ensemble of neural networks, to detect anomalies based on their feature distributions. Our model estimates the likelihood of non-defective features, identifying anomalies as out-of-distribution values. Extensive experiments on three state-of-the-art anomaly detection datasets demonstrate that our proposal achieves superior AUROC performance and improves computational efficiency compared to existing approaches. Furthermore, we validate the robustness and adaptability of our proposal through low-shot training experiments using only 20% of available training data, highlighting its potential as an efficient solution for cold start anomaly detection.

Risk profiling of tobacco epidemic and estimated number of smokers living in China: a cross-sectional study based on PBICR

BackgroundEvidence on the prevalence of smoking in China remains insufficient, with most previous studies focusing on a single region. However, smoking prevalence exhibits significant inequalities across the entire country. This study aimed to evaluate the risk of tobacco prevalence across the country, taking into account spatial inequalities.MethodsThe data used in this study were collected in 23 provinces, 5 autonomous regions, and 4 municipalities directly under the central government in 2022. Large population survey data were used, and a Bayesian geostatistical model was employed to investigate smoking prevalence rates across multiple spatial domains.FindingsSignificant spatial variations were observed in smokers and exposure to secondhand smoke across China. Higher levels of smokers and secondhand smoke exposure were observed in western and northeastern regions. Additionally, the autonomous region of Tibet, Shanghai municipality, and Yunnan province had the highest prevalence of smokers, while Tibet, Qinghai province, and Yunnan province had the highest prevalence of exposure to secondhand smoke.ConclusionWe have developed a model-based, high-resolution nationwide assessment of smoking risks and employed rigorous Bayesian geostatistical models to help visualize smoking prevalence predictions. These prediction maps provide estimates of the geographical distribution of smoking, which will serve as strong evidence for the formulation and implementation of smoking cessation policies.HighlightsOur study investigated the prevalence of smokers and exposure to secondhand smoke in different spatial areas of China and explored various factors influencing the smoking prevalence. For the first time, our study applied Bayesian geostatistical modeling to generate a risk prediction map of smoking prevalence, which provides a more intuitive and clear understanding of the spatial disparities in smoking prevalence across different geographical regions, economic levels, and development status. We found significant spatial variations in smokers and secondhand smoke exposure in China, with higher rates in the western and northeastern regions.

Risk Assessment and Distribution Estimation for UAV Operations with Accurate Ground Feature Extraction Based on a Multi-Layer Method in Urban Areas

Risk Assessment and Distribution Estimation for UAV Operations with Accurate Ground Feature Extraction Based on a Multi-Layer Method in Urban Areas

Bias Reduction of Maximum Likelihood Estimation in the Inverse Xgamma Distribution

A distribution for modeling survival data that accounts for flexibility in modeling data with upside-down bathtub-shaped hazard rate functions is the inverse Xgamma distribution (IXG). The maximum likelihood method (MLE) is the most often used technique for parameter estimation of the IXG distribution. Conversely, the MLE is infamously biased for small sample sizes. This motivates us to produce almost unbiased estimators for IXG parameter. More precisely, we minimize MLE biases to the second degree of bias using two techniques for bias correction: bootstrap and analytical approaches. Two actual data applications and Monte Carlo simulations are used to compare the performances of these methods.

Discriminative Hamiltonian variational autoencoder for accurate tumor segmentation in data-scarce regimes

Deep learning has gained significant attention in medical image segmentation. However, the limited availability of annotated training data presents a challenge to achieving accurate results. In efforts to overcome this challenge, data augmentation techniques have been proposed. However, the majority of these approaches primarily focus on image generation. For segmentation tasks, providing both images and their corresponding target masks is crucial, and the generation of diverse and realistic samples remains a complex task, especially when working with limited training datasets. To this end, we propose a new end-to-end hybrid architecture based on Hamiltonian Variational Autoencoders (HVAE) and a discriminative regularization to improve the quality of generated images. Our method provides an accurate estimation of the joint distribution of the images and masks, resulting in the generation of realistic medical images with reduced artifacts and off-distribution instances. As generating 3D volumes requires substantial time and memory, our architecture operates on a slice-by-slice basis to segment 3D volumes, capitalizing on the richly augmented dataset. Experiments conducted on two public datasets, BRATS (MRI modality) and HECKTOR (PET modality), demonstrate the efficacy of our proposed method on different medical imaging modalities with limited data.

Bundle-specific tractogram distribution estimation using higher-order streamline differential equation

Streamline tractography locally traces peak directions extracted from fiber orientation distribution (FOD) functions, lacking global information about the trend of the whole fiber bundle. Therefore, it is prone to producing erroneous tracks while missing true positive connections. In this work, we propose a new bundle-specific tractography (BST) method based on a bundle-specific tractogram distribution (BTD) function, which directly reconstructs the fiber trajectory from the start region to the termination region by incorporating the global information in the fiber bundle mask. A unified framework for any higher-order streamline differential equation is presented to describe the fiber bundles with disjoint streamlines defined based on the diffusion vectorial field. At the global level, the tractography process is simplified as the estimation of BTD coefficients by minimizing the energy optimization model, and is used to characterize the relations between BTD and diffusion tensor vector under the prior guidance by introducing the tractogram bundle information to provide anatomic priors. Experiments are performed on simulated Hough, Sine, Circle data, ISMRM 2015 Tractography Challenge data, FiberCup data, and in vivo data from the Human Connectome Project (HCP) for qualitative and quantitative evaluation. Results demonstrate that our approach reconstructs complex fiber geometry more accurately. BTD reduces the error deviation and accumulation at the local level and shows better results in reconstructing long-range, twisting, and large fanning tracts.

Thermal Analysis of Parabolic and Fresnel Linear Solar Collectors Using Compressed Gases as Heat Transfer Fluid in CSP Plants

This study introduces the use of compressed air as a heat transfer fluid in small-scale, concentrated linear solar collector technology, evaluating its possible advantages over traditional fluids. This work assumes the adoption of readily available components for both linear parabolic trough and Fresnel collectors and the coupling of the solar field with Brayton cycles for power generation. The aim is to provide a theoretical analysis of the applicability of this novel solar plant configuration for small-scale electricity generation. Firstly, a lumped thermal model was developed in a MatLab® (v. 2023a) environment to assess the thermal performance of a PT collector with an evacuated receiver tube. This model was then modified to describe the performance of a Fresnel collector. The resulting optical–thermal model was validated through literature data and appears to provide realistic estimates of temperature distribution along the entire collector length, including both the receiver tube surface and the Fresnel collector’s secondary concentrator. The analysis shows a high thermal efficiency for both Fresnel and parabolic collectors, with average values above 0.9 (in different wind conditions). Th5s study also shows that the glass covering of the Fresnel evacuated receiver, under the conditions considered (solar field outlet temperature: 550 °C), reaches significant temperatures (above 300 °C). Furthermore, due to the presence of the secondary reflector, the temperature difference between the upper and the lower part of the glass envelope can be very high, well above 100 °C in the final part of the collector string. Differently, in the case of PTs, this temperature difference is quite limited (below 30 °C).

Structural characterization of gallium fluoride phosphate glasses by advanced solid‐state NMR methods and correlation with photophysical properties

AbstractGallium fluoride phosphate glasses feature low refractive index, high energy radiation resistance, wide transmission range, and favorable emission characteristics of rare‐earth dopants. For the development of optimized glass compositions, a fundamental understanding of these properties in terms of glass structure is sought. We report nuclear magnetic resonance (NMR) structural studies of glasses in the system xGa(PO3)3–(40 − x)GaF3–20BaF2–20ZnF2–20SrF2 (x = 5, 10, 15, 20, and 25 mol%). 31P NMR results with 71Ga recoupling show that the network structure is dominated by P–O–Ga linkages, and no P–O–P linkages exist. 71Ga NMR results show that Ga is mainly six‐coordinated featuring a mixed fluoride/phosphate coordination. Quantitative estimates of this ligand distribution around gallium were obtained by 71Ga{31P} spin echo double resonance (REDOR) measurements. Photophysical properties suggest changes in the Eu(III) ligand distribution toward a fluoride‐dominated environment at low P/F ratio while the glass network is largely sustained by bridging oxygen atoms via P–O–Ga linkages.

An optimal transport approach to estimating causal effects via nonlinear difference-in-differences

Abstract We propose a nonlinear difference-in-differences (DiD) method to estimate multivariate counterfactual distributions in classical treatment and control study designs with observational data. Our approach sheds a new light on existing approaches like the changes-in-changes estimator and the classical semiparametric DiD estimator, and it also generalizes them to settings with multivariate heterogeneity in the outcomes. The main benefit of this extension is that it allows for arbitrary dependence between the coordinates of vector potential outcomes and includes higher-dimensional unobservables, something that existing methods cannot provide in general. We demonstrate its utility on both synthetic and real data. In particular, we revisit the classical Card & Krueger dataset, which reports fast food restaurant employment before and after a minimum wage increase. A reanalysis with our methodology suggests that these restaurants substitute full-time labor with part-time labor on aggregate in response to a minimum wage increase. This treatment effect requires estimation of the multivariate counterfactual distribution, an object beyond the scope of classical causal estimators previously applied to this data.

The Immigrant Mortality Advantage Among Over-65 Middle Eastern and North African Immigrants to the United States.

Immigrants to the United States often have longer life expectancies than their U.S.-born counterparts, however it is unclear whether a similar "immigrant advantage" exists for immigrants from the Middle East and North Africa (MENA). This study uses a novel machine-learning name classifier to offer one of the first national-level examinations of MENA mortality patterns by nativity in the United States. A recurrent neural network model was developed to identify MENA individuals based on given name and surname characteristics. The model was trained on more than 2.5 million mortality-linked social security records in the Berkeley Unified Numident Mortality Database (BUNMD). Mortality rates and life expectancy were estimated using a Gompertz distribution and maximum likelihood estimation, focusing on high-coverage years between 1988 and 2005 and deaths over age 65. Foreign-born MENA men over 65 showed a significant immigrant mortality advantage with a hazard ratio (HR) of 0.64 and an estimated 3.13 additional years of life expectancy at age 65 compared to U.S.-born counterparts. Foreign-born MENA women also exhibited an advantage, with a HR of 0.71 and an additional 2.24years of life expectancy at age 65. This study is one of the first national-level analyses of mortality outcomes among the over-65 MENA population in the United States, finding a MENA immigrant mortality advantage. The results suggest further research is needed to identify and disaggregate the MENA population in health research.

Estimation of porosity distribution from formation image logs and comparing the results with NMR log analysis: Implications for the heterogeneity analysis of carbonate reservoirs with an example from the Asmari reservoir, southwest Iran

Estimating heterogeneity in carbonate rocks is the most important parameter in identifying reservoir and non-reservoir areas. Diagenetic and sedimentary processes directly control reservoir heterogeneity. In this regard, the high hydrocarbon production rate and the reduction of the complexity of the carbonate reservoirs require the evaluation of a new architecture of the reservoir, in which the simultaneous impact of the diagenetic-sedimentary processes on the petrophysical logs is investigated. The current study studies the Oligocene-Miocene Asmari Formation Iran's giant carbonate reservoir. Based on petrographic studies, sedimentary facies, depositional environment and diagenetic processes were characterized. Formation evaluation was conducted to determine lithology, water saturation, fluid volume, and effective porosity. The estimated porosity from the evaluation of the conventional logs (full suite) was compared with the results of estimating the porosity from the NMR T2 distribution curve and the porosity histogram of the image logs. Lorenz's coefficient of heterogeneity was used to understand the heterogeneity of the reservoir. Diagenetic-sedimentary processes and petrophysical logs were used to investigate the pore type and their distribution. The results of petrographic studies, NMR T2 distribution curve, and porosity histogram show that the diagenetic-sedimentary process is the most important factor enhancing the reservoir heterogeneity. Diagenetic and sedimentary processes were traced by evaluating NMR and Image logs. Based on the results, the porosity distribution using the T2 curve and Image show a good agreement. Diagenetic processes mainly dissolution and dolomitization increased the heterogeneity of the main production zones (zones 3 and 4) of the Asmari reservoir that was confirmed using NMR and Image logs.

Brain Tumor Detection using Hybrid CNN in fMRI images

One of the deadliest brain tumors in the world. Both adults and children are at risk for the disease. Depending on their position, appearance and shape, it has a low survival rate and is available in different varieties. Incorrect classification of the brain tumor will have negative consequences. Early detection of the tumor type and stage is necessary for selecting a specific treatment. Identifying brain tumors can be done effectively by examining fMRI, or functional magnetic resonance imaging. Due to the various information and various types of tumors in brain, the manual process takes lot of time to identify, and there may be chances of human error. Therefore, it is necessary to have an intelligent diagnostic system. The recent evolution of image categorization systems has been influenced by the use of deep convolution neural networks (CNN). In this research work, a new hybrid paradigm with images combines a convolution neural network (CNN) and a masked autoencoder for distribution estimation (MADE). Three different forms of brain tumors were tested using T1-weighted contrast enhanced images. The CNN-MADE hybrid's excellent classification performance is demonstrated by the results, despite the limited availability of medical images. A new layered model was proposed to categorize brain tumors through MRI. The proposed approach is more successful than the alternatives based on the experiments' results.

Fitting COVID-19 datasets to a new statistical model

This paper discussed gull alpha power Weibull distribution with a three-parameter. Different statistical inference methods of Gull Alpha Power Weibull distribution parameters have been obtained, estimated, and evaluated. Then, the results are compared to find a suitable model. The unknown parameters of the published Gull Alpha Power Weibull distribution are analyzed. Seven estimation methods are maximum likelihood, Anderson–Darling, right-tail Anderson–Darling, Cramér–von Mises, ordinary least-squares, weighted least-squares, and maximum product of spacing. In addition, the performance of this distribution is computed using the Monte Carlo method, and the limited sample features of parameter estimates for the proposed distribution are analyzed. In light of the importance of heavy-tailed distributions, actuarial approaches are employed. Applying actuarial criteria such as value at risk and tail value at risk to the suggested distribution shows that the model under study has a larger tail than the Weibull distribution. Two real-world COVID-19 infection datasets are used to evaluate the distribution. We analyze the existence and uniqueness of the log-probability roots to establish that they represent the global maximum. We conclude by summarizing the outcomes reported in this study.

Ocean Wave Directional Distribution from GPS Buoy Observations off the West Coast of Ireland: Assessment of a Wavelet-Based Method

Abstract Knowledge of the directional distribution of a wave field is crucial for a better understanding of complex air–sea interactions. However, the dynamic and unpredictable nature of ocean waves, combined with the limitations of existing measurement technologies and analysis techniques, makes it difficult to obtain precise directional information, leading to a poor understanding of this important quantity. This study investigates the potential use of a wavelet-based method applied to GPS buoy observations as an alternative approach to the conventional methods for estimating the directional distribution of ocean waves. The results indicate that the wavelet-based estimations are consistently good when compared to the framework of widely used parameterizations for the directional distribution. The wavelet-based method presents advantages in comparison with the conventional methods, including being purely data-driven and not requiring any assumptions about the shape of the distribution. In addition, it was found that the wave directional distribution is narrower at the spectral peak and broadens asymmetrically at higher and lower scales, particularly sharply for frequencies below the peak. The directional spreading appears to be independent of the wave age across the entire range of frequencies, implying that the angular width of the directional spectrum is primarily controlled by nonlinear wave–wave interactions rather than by wind forcing. These results support the use of the wavelet-based method as a practical alternative for the estimation of the wave directional distribution. In addition, this study highlights the need for continued innovation in the field of ocean wave measuring technologies and analysis techniques to improve our understanding of air–sea interactions. Significance Statement This study presents a wavelet-based technique for obtaining the directional distribution of ocean waves applied to GPS buoy. This method serves as an alternative to conventional methods and is relatively easy to implement, making it a practical option for researchers and engineers. The study was conducted in a highly energetic environment characterized by high wind speeds and large waves, providing a valuable dataset for understanding the dynamics of marine environment in extreme conditions. This research has implications for improving our understanding of directional characteristics of ocean waves, which is crucial for navigation, offshore engineering, weather forecasting, and coastal hazard mitigation. This study also highlights the challenges associated with understanding wave directionality and emphasizes a need for further observations.

Enhancing subsurface seismic profiling with distributed acoustic sensing and optimization algorithms

The distribution of shear-wave velocities in the subsurface is generally used to assess the potential for seismic liquefaction and soil amplification effects and to classify seismic sites. Newly developed distributed acoustic sensing (DAS) technology enables estimation of the shear-wave distribution as a high-density seismic observation system. This technology is characterized by low maintenance costs, high-resolution outputs, and real-time data transmission capabilities, albeit with the challenge of managing massive data generation. Rapid and efficient interpretation of data is the key to advancing application of the DAS technology. In this study, field tests were carried out to record ambient noise over a short period using DAS technology, from which the surface-wave dispersion curves were extracted. In order to reduce the influence of directional effects on the results, an unsupervised clustering method is used to select appropriate clusters to extract the Green's function. A combination of a genetic algorithm and Monte Carlo (GA-MC) simulation is proposed to invert the subsurface velocity structure. The stratigraphic profiles obtained by the GA-MC method are in agreement with the borehole profiles. Compared to other methods, the proposed optimization method not only improves the solution quality but also reduces the solution time.

The SPAR Model: A New Paradigm for Multivariate Extremes: Application to Joint Distributions of Metocean Variables

Abstract This paper presents the application of a new multivariate extreme value model for the estimation of metocean variables. The model requires fewer assumptions about the forms of the marginal distributions and dependence structure compared to existing approaches, and provides a flexible and rigorous framework for modeling multivariate extremes. The method involves a transformation of variables to polar coordinates. The tail of the radial variable is then modeled using the generalized Pareto distribution, with parameters conditional on angle, providing a natural extension of univariate theory to multivariate problems. The resulting model is referred to as the semi-parametric angular-radial (SPAR) model. We consider the estimation of the joint distributions of (1) wave height and wave period, and (2) wave height and wind speed. We show that the SPAR model provides a good fit to the observations in terms of both the marginal distributions and dependence structures. The use of the SPAR model for estimating long-term extreme responses of offshore structures is discussed, using some simple response functions for floating structures and an offshore wind turbine with monopile foundation. We show that the SPAR model is able to accurately reproduce response distributions, and provides a realistic quantification of uncertainty.

Using joint species distribution modelling to predict distributions of seafloor taxa and identify vulnerable marine ecosystems in New Zealand waters

Effective ecosystem-based management of bottom-contacting fisheries requires understanding of how disturbances from fishing affect seafloor fauna over a wide range of spatial and temporal scales. Spatial predictions of abundance for 67 taxa were developed, using an extensive dataset of faunal abundances collected using a towed camera system and spatially explicit predictor variables including bottom-trawl fishing effort, using a Joint Species Distribution Model (JSDM). The model fit metrics varied by taxon: the mean tenfold cross-validated AUC score was 0.70 ± 0.1 (standard deviation) for presence–absence and an R2 of 0.11 ± 0.1 (standard deviation) for abundance models. Spatial predictions of probability of occurrence and abundance (individuals per km2) varied by taxon, but there were key areas of overlap, with highest predicted taxon richness in areas of the continental shelf break and slope. The resulting joint predictions represent significant advances on previous predictions because they are of abundance, allow the exploration of co-occurrence patterns and provide credible estimates of taxon richness (including for rare species that are often not included in more commonly used single-species distribution modelling). Habitat-forming taxa considered to be Vulnerable Marine Ecosystem (VME) indicators (those taxa that are physically or functionally fragile to anthropogenic impacts) were identified in the dataset. Spatial estimates of likely VME distribution (as well as associated estimates of uncertainty) were predicted for the study area. Identifying areas most likely to represent a VME (rather than simply VME indicator taxa) provides much needed quantitative estimates of vulnerable habitats, and facilitates an evidence-based approach to managing potential impacts of bottom-trawling.

Exploring the Transport Path of Oceanic Microplastics in the Atmosphere.

Microplastics (MP) have been recognized as an emerging atmospheric pollutant, yet uncertainties persist in their emissions and concentrations. With a bottom-up approach, we estimate 6-hourly MP fluxes at the ocean-atmosphere interface, using as an input the monthly ocean surface MP concentrations simulated by the global oceanic model (NEMO/PISCES-PLASTIC, Nucleus for European Modeling of the Ocean, Pelagic Interaction Scheme for Carbon and Ecosystem Studies), a size distribution estimate for the MP in the micrometer range, and a sea salt emission scheme. The atmospheric dispersion is then simulated with the Lagrangian model FLEXPART. We identify hotspot sources in the tropical regions and highlight the seasonal variability of emissions, atmospheric concentrations, and deposition fluxes both on land and ocean surfaces. Due to the variability of MP concentration during the year, the MP flux from the sea surface appears to follow a seasonality opposite to that of sea salt aerosol emissions. The comparison with existing observations of MP in the marine atmosphere suggests an underestimation of one to 2 orders of magnitude in our current knowledge of the MP in the oceans' surface. In addition, we show that the MP in the micrometer range is transported efficiently around the globe and can penetrate and linger in the stratosphere over time scales of months. The interaction of these particles with the chemistry and physics of the atmosphere is still mostly unknown and deserves to be further investigated.

Adaptive minimax estimation of service time distribution in the Mt/G/∞\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$M_t/G/\\infty $$\\end{document} queue from departure data

This article deals with the problem of estimating the service time distribution of the Mt/G/∞\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$M_t/G/\\infty $$\\end{document} queue from observation of the departure epochs. We develop minimax optimal estimators of G and study behavior of the minimax pointwise risk over a suitable family of service time distribution functions. In addition, we address the problem of adaptive estimation and propose a data–driven estimation procedure that adapts to unknown smoothness of the service time distribution function G. Lastly, a numerical study is presented to illustrate practical performance of the developed adaptive procedure.