Spatial Dimension Reduction Research Articles

An Adaptive Atrous Spatial Pyramid Pooling Network for Hyperspectral Classification

Hyperspectral imaging (HSI) offers rich spectral and spatial data, beneficial for a variety of applications. However, challenges persist in HSI classification due to spectral variability, non-linearity, limited samples, and a dearth of spatial information in conventional spectral classifiers. While various spectral–spatial classifiers and dimension reduction techniques have been developed to mitigate these issues, they are often constrained by the utilization of handcrafted features. Deep learning has been introduced to HSI classification, with pixel- and patch-level deep learning (DL) classifiers gaining substantial attention. Yet, existing patch-level DL classifiers encounter difficulties in concentrating on long-distance dependencies and managing category areas of diverse sizes. The proposed Self-Adaptive 3D atrous spatial pyramid pooling (ASPP) Multi-Scale Feature Fusion Network (SAAFN) addresses these challenges by simultaneously preserving high-resolution spatial detail data and high-level semantic information. This method integrates a modified hyperspectral superpixel segmentation technique, a multi-scale 3D ASPP convolution block, and an end-to-end framework to extract and fuse multi-scale features at a self-adaptive rate for HSI classification. This method significantly enhances the classification accuracy of HSI with limited samples.

An ENSO Prediction Model Based on Backtracking Multiple Initial Values: Ordinary Differential Equations–Memory Kernel Function

This article presents a new prediction model, the ordinary differential equations–memory kernel function (ODE–MKF), constructed from multiple backtracking initial values (MBIV). The model is similar to a simplified numerical model after spatial dimension reduction and has both nonlinear characteristics and the low-cost advantage of a time series model. The ODE–MKF focuses on utilizing more temporal information and includes machine learning to solve complex mathematical inverse problems to establish a predictive model. This study first validates the feasibility of the ODE–MKF via experiments using the Lorenz system. The results demonstrate that the ODE–MKF prediction model could describe the nonlinear characteristics of complex systems and exhibited ideal predictive robustness. The prediction of the El Niño-Southern Oscillation (ENSO) index further demonstrates its effectiveness, as it achieved 24-month lead predictions and effectively improved nonlinear problems. Furthermore, the reliability of the model was also tested, and approximately 18 months of prediction were achieved, which was verified with the Clouds and the Earth’s Radiant Energy System (CERES) Energy Balanced and Filled (EBAF) radiation fluxes. The short-term memory index Southern Oscillation (SO) was further used to examine the applicability of ODE–MKF. A six-month lead prediction of the SO trend was achieved, indicating that the predictability of complex systems is related to their inherent memory scales.

Joint species distribution models with imperfect detection for high-dimensional spatial data.

Determining the spatial distributions of species and communities is a key task in ecology and conservation efforts. Joint species distribution models are a fundamental tool in community ecology that use multi-species detection-nondetection data to estimate species distributions and biodiversity metrics. The analysis of such data is complicated by residual correlations between species, imperfect detection, and spatial autocorrelation. While many methods exist to accommodate each of these complexities, there are few examples in the literature that address and explore all three complexities simultaneously. Here we developed a spatial factor multi-species occupancy model to explicitly account for species correlations, imperfect detection, and spatial autocorrelation. The proposed model uses a spatial factor dimension reduction approach and Nearest Neighbor Gaussian Processes to ensure computational efficiency for data sets with both a large number of species (e.g., >100) and spatial locations (e.g., 100,000). We compared the proposed model performance to five alternative models, each addressing a subset of the three complexities. We implemented the proposed and alternative models in the spOccupancy software, designed to facilitate application via an accessible, well documented, and open-source R package. Using simulations, we found that ignoring the three complexities when present leads to inferior model predictive performance, and the impacts of failing to account for one or more complexities will depend on the objectives of a given study. Using a case study on 98 bird species across the continental US, the spatial factor multi-species occupancy model had the highest predictive performance among the alternative models. Our proposed framework, together with its implementation in spOccupancy, serves as a user-friendly tool to understand spatial variation in species distributions and biodiversity while addressing common complexities in multi-species detection-nondetection data.

FIRE IMAGE DETECTION USING CNN

Fire detection is a critical task in ensuring the safety of human lives and property. In recent years, deep learning techniques, such as Convolutional Neural Networks (CNNs), have shown promising results in fire detection due to their ability to automatically learn relevant features from images. In this project, we propose a real-time fire detection system using CNNs to detect fires in images or video streams. The proposed system consists of a CNN-based architecture that includes multiple convolutional layers for feature extraction, activation functions for introducing non-linearity, pooling layers for spatial dimension reduction, and fully connected layers for global pattern learning. The architecture is trained on a labelled dataset of fire and non-fire images or videos using optimization algorithms to minimize the loss function. To achieve real-time fire detection, the trained CNN model is integrated into a fire detection system that can process images or video streams in real-time. The system captures images or video frames from a live feed, pre-processes the data, and passes it through the CNN model for fire detection. The system can generate alerts or notifications in real-time when fire is detected, allowing for prompt response and mitigation measures. The performance of the proposed fire detection system is evaluated using metrics such as accuracy, precision, recall, and F1-score. The system is tested on different datasets and under various conditions to assess its accuracy, robustness, and real-time processing capabilities. The results of this project demonstrate the effectiveness of CNNs for real-time fire detection and highlight the potential of the proposed system for enhancing fire safety in various applications, including surveillance systems, smart buildings, and industrial settings. The developed system can be used as an early warning tool for fire detection, enabling timely response and mitigation, and helping to prevent or minimize the damage caused by fires. Keywords: Fire

Simplicial Complex-Enhanced Manifold Embedding of Spatiotemporal Data for Structural Health Monitoring

Structural Health Monitoring requires the continuous assessment of a structure’s operational conditions, which involves the collection and analysis of a large amount of data in both spatial and temporal domains. Conventionally, both data-driven and physics-based models for structural damage detection have relied on handcrafted features, which are susceptible to the practitioner’s expertise and experience in feature selection. The limitations of handcrafted features stem from the potential for information loss during the extraction of high-dimensional spatiotemporal data collected from the sensing system. To address this challenge, this paper proposes a novel, automated structural damage detection technique called Simplicial Complex Enhanced Manifold Embedding (SCEME). The key innovation of SCEME is the reduction of dimensions in both the temporal and spatial domains for efficient and information-preserving feature extraction. This is achieved by constructing a simplicial complex for each signal and using the resulting topological invariants as key features in the temporal domain. Subsequently, curvature-enhanced topological manifold embedding is performed for spatial dimension reduction. The proposed methodology effectively represents both intra-series and inter-series correlations in the low-dimensional embeddings, making it useful for classification and visualization. Numerical simulations and two benchmark experimental datasets validate the high accuracy of the proposed method in classifying different damage scenarios and preserving useful information for structural identification. It is especially beneficial for structural damage detection using complex data with high spatial and temporal dimensions and large uncertainties in reality.

Real-time long-term voltage stability assessment based on eGBDT for large-scale power system with high renewables penetration

• A data-driven voltage stability assessment framework is established, which can be performed utilizing observed PMU measurements and is completely independent from system modeling. • Based on the complete datasets, considering the random and uncertainty of renewable energy in multiple scenarios, we developed a refined voltage state classification ( secure, warning and emergency ) for high renewables penetration power systems. • Dramatic speedup by virtue of GOSS and HEEFB techniques enables the traditional GBDT-based learning algorithms to be effective for handling massive high-dimensional data, realizing the real-time assessment of voltage stability. Real-time voltage stability assessment is a critical issue for ensuring the safety of large-scale renewable energy connected power system. In this paper, a data-driven approach based on the enhanced Gradient Boost Decision Tree (eGBDT) for long-term voltage stability assessment from massive high-dimensional PMU measurements was proposed. This approach effectively improves the ability of the traditional GBDT algorithm to handle high-dimensional observed data by utilizing gradient-based one-side sampling (GOSS) and histogram-enhanced exclusive feature bundling (HEEFB) techniques, realizing spatial dimension reduction and feature compression. The proposed method exploits both the accuracy of the GBDT algorithm and the fast computational capacity of GOSS and HEEFB techniques, which effectively reduces the calculation time. Case studies on the Nordic32 system and IEEE 118-bus test system demonstrated the accuracy and the speed of the proposed approach. Extensive studies on the robustness of dealing with missing data illustrate that the proposed method is practically suitable for real-time application.

Multispectral LiDAR point cloud highlight removal based on color information.

With the rapid development of light detection and ranging (LiDAR) technology, multispectral LiDAR (MSL) can realize three-dimensional (3D) imaging of the ground object by acquiring rich spectral information. Although color restoration has been achieved on the basis of the full-waveform data of MSL, further improvement of the visual effect of color point clouds still faces many challenges. In this paper, a highlight removal method for MSL color point clouds is proposed to explore the potential of 3D visualization. First, the MSL reflection model are introduced according to radar equation and Phong model, and the restored color of the MSL point clouds is determined to comprise diffuse and specular components. Second, a data conversion method is proposed to improve the massive point cloud processing efficiency by spatial dimension reduction and data compression. Then, the visual saliency map after color denoising is used to obtain the highlight region, the unknown information of which is recovered based on the global or local color information. Finally, three representative targets are selected and evaluated by qualitative and quantitative validation, which verifies that the method can effectively recover the high-quality highlight-free point clouds of MSL.

A sophisticated offline network developed for recognizing handwritten Chinese character efficiently

In this paper, we designed an efficient network structure by combining multiple scale convolution shuffle (MSCS) module and attention features spatial aggregation (ASA) module. MSCS module can obtain local and global multiple receptive fields feature information, and it contains the structure idea of the lightweight network convolution block. Moreover, we have proposed ASA module to reducing the information loss of different features in spatial dimension reduction. Meanwhile, we introduce the joint loss function to strengthen the ability of the model to distinguish the difference of attention features between inter-class and intra-class. The experimental results tested on offline HCCR competition dataset ICDAR-2013 show that our network model trained only on handwritten dataset only takes 3.97 ms to recognize a character image and achieves 97.63% accuracy just requires 22.9 MB for storage. Therefore, our method reaches the state-of-the-art in efficient single networks particularly in view of the storage space and inference time.

Quantifying the Human Influence on the Intensity of Extreme 1- and 5-Day Precipitation Amounts at Global, Continental, and Regional Scales

Abstract This study provides a comprehensive analysis of the human contribution to the observed intensification of precipitation extremes at different spatial scales. We consider the annual maxima of the logarithm of 1-day (Rx1day) and 5-day (Rx5day) precipitation amounts for 1950–2014 over the global land area, four continents, and several regions, and compare observed changes with expected responses to external forcings as simulated by CanESM2 in a large-ensemble experiment and by multiple models from phase 6 of the Coupled Model Intercomparison Project (CMIP6). We use a novel detection and attribution analysis method that is applied directly to station data in the areas considered without prior processing such as gridding, spatial or temporal dimension reduction, or transformation to unitless indices and uses climate models only to obtain estimates of the space–time pattern of extreme precipitation response to external forcing. The influence of anthropogenic forcings on extreme precipitation is detected over the global land area, three continental regions (the western Northern Hemisphere, western Eurasia, and eastern Eurasia), and many smaller IPCC regions, including central North America, East Asia, east-central Asia, eastern Europe, eastern North America, northern Europe, and western Siberia for Rx1day, and central North America, eastern Europe, eastern North America, northern Europe, the Russian Arctic region, and western Siberia for Rx5day. Consistent results are obtained using forcing response estimates from either CanESM2 or CMIP6. Anthropogenic influence is estimated to have substantially decreased the approximate waiting time between extreme annual maximum events in regions where anthropogenic influence has been detected, which has important implications for infrastructure design and climate change adaptation policy. Significance Statement All previous detection and attribution studies of observed changes in extreme precipitation (i) use station data that has been heavily processed via gridding, transformation, and spatial and temporal averaging or other dimension reduction approaches, as well as using climate models to estimate the responses to external forcing, and (ii) also use models to estimate the unforced natural variability of extreme precipitation. Both aspects reduce user confidence in detection and attribution results. This study uses station data directly and avoids difficult to verify model-based estimates of the unforced variability of precipitation extremes. Results confirm findings from previous studies, and extend them to a number of subcontinental regions, thus substantially increasing confidence in detection and attribution findings precipitation.

Windowed least-squares model reduction for dynamical systems

This work proposes a windowed least-squares (WLS) approach for model reduction of dynamical systems. The proposed approach sequentially minimizes the time-continuous full-order-model residual within a low-dimensional space–time trial subspace over time windows. The approach comprises a generalization of existing model reduction approaches, as particular instances of the methodology recover Galerkin, least-squares Petrov–Galerkin (LSPG), and space–time LSPG projection. In addition, the approach addresses key deficiencies in existing model reduction techniques, e.g., the dependence of LSPG and space–time LSPG projection on the time discretization and the exponential growth in time exhibited by a posteriori error bounds for both Galerkin and LSPG projection. We consider two types of space–time trial subspaces within the proposed approach: one that reduces only the spatial dimension of the full-order model, and one that reduces both the spatial and temporal dimensions of the full-order model. For each type of trial subspace, we consider two different solution techniques: direct (i.e., discretize then optimize) and indirect (i.e., optimize then discretize). Numerical experiments conducted using trial subspaces characterized by spatial dimension reduction demonstrate that the WLS approach can yield more accurate solutions with lower space–time residuals than Galerkin and LSPG projection.

Double-layer Bose-Einstein condensates: A quantum phase transition in the transverse direction, and reduction to two dimensions.

We revisit the problem of the reduction of the three-dimensional (3D) dynamics of Bose-Einstein condensates, under the action of strong confinement in one direction (z), to a 2D mean-field equation. We address this problem for the confining potential with a singular term, viz., V_{z}(z)=2z^{2}+ζ^{2}/z^{2}, with constant ζ. A quantum phase transition is induced by the latter term, between the ground state (GS) of the harmonic oscillator and the 3D condensate split in two parallel noninteracting layers, which is a manifestation of the "superselection" effect. A realization of the respective physical setting is proposed, making use of resonant coupling to an optical field, with the resonance detuning modulated along z. The reduction of the full 3D Gross-Pitaevskii equation (GPE) to the 2D nonpolynomial Schrödinger equation (NPSE) is based on the factorized ansatz, with the z -dependent multiplier represented by an exact GS solution of the 1D Schrödinger equation with potential V_{z}(z). For both repulsive and attractive signs of the nonlinearity, the 2D NPSE produces GS and vortex states, that are virtually indistinguishable from the respective numerical solutions provided by full 3D GPE. In the case of the self-attraction, the threshold for the onset of the collapse, predicted by the 2D NPSE, is also virtually identical to its counterpart obtained from the 3D equation. In the same case, stability and instability of vortices with topological charge S=1, 2, and 3 are considered in detail. Thus, the procedure of the spatial-dimension reduction, 3D → 2D, produces very accurate results, and it may be used in other settings.

An Examination of Grouping and Spatial Organization Tasks for High-Dimensional Data Exploration

How do analysts think about grouping and spatial operations? This overarching research question incorporates a number of points for investigation, including understanding how analysts begin to explore a dataset, the types of grouping/spatial structures created and the operations performed on them, the relationship between grouping and spatial structures, the decisions analysts make when exploring individual observations, and the role of external information. This work contributes the design and results of such a study, in which a group of participants are asked to organize the data contained within an unfamiliar quantitative dataset. We identify several overarching approaches taken by participants to design their organizational space, discuss the interactions performed by the participants, and propose design recommendations to improve the usability of future high-dimensional data exploration tools that make use of grouping (clustering) and spatial (dimension reduction) operations.

Restricted Spatial Regression Methods: Implications for Inference

The issue of spatial confounding between the spatial random effect and the fixed effects in regression analyses has been identified as a concern in the statistical literature. Multiple authors have offered perspectives and potential solutions. In this article, for the areal spatial data setting, we show that many of the methods designed to alleviate spatial confounding can be viewed as special cases of a general class of models. We refer to this class as restricted spatial regression (RSR) models, extending terminology currently in use. We offer a mathematically based exploration of the impact that RSR methods have on inference for regression coefficients for the linear model. We then explore whether these results hold in the generalized linear model setting for count data using simulations. We show that the use of these methods have counterintuitive consequences which defy the general expectations in the literature. In particular, our results and the accompanying simulations suggest that RSR methods will typically perform worse than nonspatial methods. These results have important implications for dimension reduction strategies in spatial regression modeling. Specifically, we demonstrate that the problems with RSR models cannot be fixed with a selection of “better” spatial basis vectors or dimension reduction techniques. Supplementary materials for this article are available online.

Orthotopic-filtering-based fault diagnosis algorithms for nonlinear systems with slowly varying faults

A novel nonlinear fault diagnosis method for detecting various slowly varying faults is proposed. The idea of linear programming binding constraints is adopted and the orthotopic spaces are recursively calculated to approximate the exact feasible parameter set, and a parameter global expansion filtering fault diagnosis algorithm is derived. Moreover, to reduce the amount of calculation and improve the efficiency of fault diagnosis, a parameter directional expansion filtering fault diagnosis algorithm is proposed based on the spatial dimension reduction. Finally, various simulation examples are provided to demonstrate the accuracy and effectiveness of the proposed algorithms.

Flowfield characteristics in sidewall compression inlets

In this work, the flow fields in three-dimensional (3D) sidewall compression inlets are investigated theoretically and numerically. Combining spatial dimension reduction theory with the shock dynamics theory, a detailed analysis of the flowfield in the compression part of an inlet is proposed. Using this analytical approach, wave configurations are determined and flowfield parameters are solved; numerical simulations are conducted to validate these theoretical results. The effects of the attack angle and yaw angle on the pressure ratio, wave structures, and total pressure recovery performance are discussed in detail. Findings offer an effective way to study shock/shock interaction (SSI) in sidewall inlets, which is essential to the designing of such inlets.

Medical Healthcare System Based on Wireless Body Area Networks: The Importance of Anomaly Detection

In this research paper, a modern framework is presented to detect anomaly in medical wireless body sensor network systems that are incorporated in distant observation of patient’s vital signs. The suggested framework effects analysis of data in a sequential manner using a mini gateway utilized as a root station to discover abnormal alterations and to deal with inaccurate computations in gathered medical information minus advance awareness of irregular occurrences or consistent data patterns. The suggested perspective relies on Principal Component Analysis (PCA) utilized in spatial analysis and dimension reduction for gathered medical details. The key goal is distinguishing defective calculations from clinical dangers for reduction of false alarms prompted by incorrect computations or ill-behaved sensors. The result from the experiments on real medical datasets reveal that the suggested PCA perspective is able to attain good discovery perfection with lesser falt alarm rate when contrasted with other perspectives that fail to minimize the excessive dimension of gathered more information so in multivariate Wireless Body Sensor Networks (WBSN) implementations, and dynamic streaming nature of sensor data, mostly in medical implementations.

Decoding Speech from Single Trial MEG Signals Using Convolutional Neural Networks and Transfer Learning.

Decoding speech directly from the brain has the potential for the development of the next generation, more efficient brain computer interfaces (BCIs) to assist in the communication of patients with locked-in syndrome (fully paralyzed but aware). In this study, we have explored the spectral and temporal features of the magnetoencephalography (MEG) signals and trained those features with convolutional neural networks (CNN) for the classification of neural signals corresponding to phrases. Experimental results demonstrated the effectiveness of CNNs in decoding speech during perception, imagination, and production tasks. Furthermore, to overcome the long training time issue of CNNs, we leveraged principal component analysis (PCA) for spatial dimension reduction of MEG data and transfer learning for model initialization. Both PCA and transfer learning were found to be highly beneficial for faster model training. The best configuration (50 principal coefficients + transfer learning) led to more than 10 times faster training than the original setting while the speech decoding accuracy remained at a similarly high level.

Shock/shock interactions between bodies and wings

This paper examines the Shock/Shock Interactions (SSI) between the body and wing of aircraft in supersonic flows. The body is simplified to a flat wedge and the wing is assumed to be a sharp wing. The theoretical spatial dimension reduction method, which transforms the 3D problem into a 2D one, is used to analyze the SSI between the body and wing. The temperature and pressure behind the Mach stem induced by the wing and body are obtained, and the wave configurations in the corner are determined. Numerical validations are conducted by solving the inviscid Euler equations in 3D with a Non-oscillatory and Non-free-parameters Dissipative (NND) finite difference scheme. Good agreements between the theoretical and numerical results are obtained. Additionally, the effects of the wedge angle and sweep angle on wave configurations and flow field are considered numerically and theoretically. The influences of wedge angle are significant, whereas the effects of sweep angle on wave configurations are negligible. This paper provides useful information for the design and thermal protection of aircraft in supersonic and hypersonic flows.

Three-dimensional shock wave configurations induced by two asymmetrical intersecting wedges in supersonic flow

This study explores the three-dimensional (3D) wave configurations induced by 3D asymmetrical intersecting compression wedges in supersonic and hypersonic inviscid flows. By using the “spatial dimension reduction” approach, the problem of 3D steady shock/shock interaction is converted to that of the interaction of two moving shock waves in the characteristic two-dimensional (2D) plane. Shock polar theory is used to analyze the shock configurations in asymmetrical situations. The results show that various shock configurations exist in 3D asymmetrical shock wave interactions, including regular interaction, transitioned regular interaction, single Mach interaction, inverse single Mach interaction, transitional double Mach interaction, weak shock interaction, and weak single Mach interaction. All of the above 3D steady shock/shock interactions have their corresponding 2D moving shock/shock interaction configurations. Numerical simulations are performed by solving the 3D inviscid Euler equations with the non-oscillatory, non-free parameters, dissipative (NND) numerical scheme, and good agreement with the theoretical analysis is obtained. Furthermore, the comparison of results show that the concept of the “virtual wall” in shock dynamics theory is helpful for understanding the mechanism of two-dimensional shock/shock interactions.

Theoretical solutions to three-dimensional asymmetrical shock/shock interaction

This paper explores theoretical solutions to the three-dimensional (3D) shock/shock interaction induced by 3D asymmetrical intersecting compression wedges in supersonic inviscid flows. For Mach interactions, an analytical method known as spatial-dimension reduction, which transforms the problem of 3D steady shock/shock interaction into a two-dimensional (2D) pseudo-steady problem on cross sections, is used to obtain the solutions in the vicinity of the Mach stem. The theoretical solutions include the pressure, temperature, density, Mach number behind the Mach stem, and total pressure recovery coefficient. Numerical simulations are performed to validate the theoretical results. Here, the NND scheme is employed by solving 3D inviscid Euler equations, and good agreements are obtained. The asymmetry of the flow characteristics induced by the wedge angle and sweep angle are thoroughly discussed.