Unsupervised Classification Method Research Articles

Gini Coefficient-Based Feature Learning for Unsupervised Cross-Domain Classification with Compact Polarimetric SAR Data

Remote sensing image classification usually needs many labeled samples so that the target nature can be fully described. For synthetic aperture radar (SAR) images, variations of the target scattering always happen to some extent due to the imaging geometry, weather conditions, and system parameters. Therefore, labeled samples in one image could not be suitable to represent the same target in other images. The domain distribution shift of different images reduces the reusability of the labeled samples. Thus, exploring cross-domain interpretation methods is of great potential for SAR images to improve the reuse rate of existing labels from historical images. In this study, an unsupervised cross-domain classification method is proposed that utilizes the Gini coefficient to rank the robust and stable polarimetric features in both the source and target domains (GRFST) such that an unsupervised domain adaptation (UDA) can be achieved. This method selects the optimal features from both the source and target domains to alleviate the domain distribution shift. Both fully polarimetric (FP) and compact polarimetric (CP) SAR features are explored for crop-domain terrain type classification. Specifically, the CP mode refers to the hybrid dual-pol mode with an arbitrary transmitting ellipse wave. This is the first attempt in the open literature to investigate the representing abilities of different CP modes for cross-domain terrain classification. Experiments are conducted from four aspects to demonstrate the performance of CP modes for cross-data, cross-scene, and cross-crop type classification. Results show that the GRFST-UDA method yields a classification accuracy of 2% to 12% higher than the traditional UDA methods. The degree of scene similarity has a certain impact on the accuracy of cross-domain crop classification. It was also found that when both the FP and circular CP SAR data are used, stable, promising results can be achieved.

Feature extraction and pattern recognition in time-lapse pressure transient responses

Monitoring of modern wells equipped with permanent downhole gauges and flowmeters provides large datasets of pressure, temperature and flowrate measurements. On a limited scale, selected events from these datasets are traditionally used in well performance analysis and monitoring, reservoir simulation, and many other tasks employing physics-based models. New methods that combine model- and data-driven approaches for full-scale analysis of these large datasets have recently attracted interest, suggesting new metrics for knowledge extraction. Most recent studies have used a combination of Pressure Transient Analysis (PTA), which is a key reservoir engineering tool, with hybrid methods, such as PTA-metrics and AI-powered models. The emergence of hybrid methodologies combining model- and data-driven approaches has opened new perspectives for comprehensive analysis and knowledge extraction from big well monitoring data sets, accumulated in the industry. Despite their potential, these methodologies often face challenges in reliability, effective data processing, interpretation and feature engineering.This paper introduces a novel PTA-feature extraction and pattern recognition methods for analysis of time-lapse pressure transient responses and their Bourdet derivatives. The pattern recognition method builds on an unsupervised classification method to extract PTA-features, associated with different flow regimes commonly used in PTA. Each pressure transient in a time-lapse series is first processed by an autonomously fine-tuned algorithm that measures the signal's distance to an ensemble of physically meaningful responses defined by PTA-feature library, thus breaking the transient into a series of likely PTA-features. A set of hyperparameters is used in the unsupervised classification, where an optimization procedure is employed for automated tuning of the hyperparameters to a particular transient. Subsequently, recognition of underlying patterns governed by sequences of these PTA-features in the time-lapse transient responses is performed. Testing of the combination of these new methods through synthetic and field cases is further carried out with verification via comparison with expert’ interpretation results. Added value to the previously introduced PTA metrics, which provide on-the-fly well and reservoir performance analysis, is finally demonstrated. The proposed pattern recognition method improves reliability and automates the calculation of the PTA metrics.The developed methodology serves as a new tool for knowledge extraction from big well monitoring datasets, available in the companies operating in the oil and gas industry, as well as the emerging industries such as carbon capture and storage and geothermal energy production. The article concludes with a discussion of the main advantages and limitations of the suggested feature extraction and pattern recognition methods. Besides the combined use of the methods described in this article, these methods may also be integrated with conventional physics-based approaches widely used in the industry for well data interpretation and reservoir simulations, improving their performance and efficiency for big data sets.

Analysis of gender gap in financial competence: an analysis using unsupervised classification methods

Analysis of gender gap in financial competence: an analysis using unsupervised classification methods

Combining supervised and unsupervised analyses to quantify behavioral phenotypes and validate therapeutic efficacy in a triple transgenic mouse model of Alzheimer's disease.

Behavioral testing is an essential tool for evaluating cognitive function and dysfunction in preclinical research models. This is of special importance in the study of neurological disorders such as Alzheimer's disease. However, the reproducibility of classic behavioral assays is frequently compromised by interstudy variation, leading to ambiguous conclusions about the behavioral markers characterizing the disease. Here, we identify age- and genotype-driven differences between 3xTg-AD and non-transgenic control mice using a low-cost, highly customizable behavioral assay that requires little human intervention. Through behavioral phenotyping combining both supervised and unsupervised behavioral classification methods, we are able to validate the preventative effects of the immunosuppressant cyclosporine A in a rodent model of Alzheimer's disease, as well as the partially ameliorating effects of candidate drugs nebivolol and cabozantinib.

An association study of m6A methylation with major depressive disorder

ObjectiveTo find the relationship between N6-methyladenosine (m6A) genes and Major Depressive Disorder (MDD).MethodsDifferential expression of m6A associated genes between normal and MDD samples was initially identified. Subsequent analysis was conducted on the functions of these genes and the pathways they may affect. A diagnostic model was constructed using the expression matrix of these differential genes, and visualized using a nomogram. Simultaneously, an unsupervised classification method was employed to classify all patients based on the expression of these m6A associated genes. Following this, common differential genes among different clusters were computed. By analyzing the functions of the common differential expressed genes among clusters, the role of m6A-related genes in the pathogenesis of MDD patients was elucidated.ResultsDifferential expression was observed in ELAVL1 and YTHDC2 between the MDD group and the control group. ELAVL1 was associated with comorbid anxiety in MDD patients. A linear regression model based on these two genes could accurately predict whether patients in the GSE98793 dataset had MDD and could provide a net benefit for clinical decision-making. Based on the expression matrix of ELAVL1 and YTHDC2, MDD patients were classified into three clusters. Among these clusters, there were 937 common differential genes. Enrichment analysis was also performed on these genes. The ssGSEA method was applied to predict the content of 23 immune cells in the GSE98793 dataset samples. The relationship between these immune cells and ELAVL1, YTHDC2, and different clusters was analyzed.ConclusionAmong all the m6A genes, ELAVL1 and YTHDC2 are closely associated with MDD, ELAVL1 is related to comorbid anxiety in MDD. ELAVL1 and YTHDC2 have opposite associations with immune cells in MDD.

Change Detection Study Of Al Razaza Lake Region Utilizing RemoteSensing And GIS Technique

Remote sensing techniques are now widely used for environment studies, surveysand monitoring. It became a need to study Razaza lake and the surrounding areasespecially when the Antiquities Authority of the Ministry of Tourism announced,that the discovery of the effects under the waters of Lake Razzaza in Karbalaprovince. The aim procedures carried out in this study, the first classified the studyarea by using unsupervised classification method to produce the land cover and landuse ,deriving the elevations as a vector layer using TIN for the whole surroundingregion of lake, the second part, the output is a continuous surface to be created byapplying Kriging interpolation, the third step, obtaining maps of contour lines alsoused to show lines of constant density or brightness by implementations the contourlines , converting the result of interpolation method to a vector values that can becalculated, then the implementation of the classified process and calculating the areaaround the lake, all these steps are done by using ArcGIS. Finally a classificationprocess for the same scene on a different date is applied and then compare it withthe first scene using ERDAS.

Atomic Force Microscopy beyond Topography: Chemical Sensing of 2D Material Surfaces through Adhesion Measurements.

Developing new functionalities of two-dimensional materials (2Dms) can be achieved by their chemical modification with a broad spectrum of molecules. This functionalization is commonly studied by using spectroscopies such as Raman, IR, or XPS, but the detection limit is a common problem. In addition, these methods lack detailed spatial resolution and cannot provide information about the homogeneity of the coating. Atomic force microscopy (AFM), on the other hand, allows the study of 2Dms on the nanoscale with excellent lateral resolution. AFM has been extensively used for topographic analysis; however, it is also a powerful tool for evaluating other properties far beyond topography such as mechanical ones. Therefore, herein, we show how AFM adhesion mapping of transition metal chalcogenide 2Dms (i.e., MnPS3 and MoS2) permits a close inspection of the surface chemical properties. Moreover, the analysis of adhesion as relative values allows a simple and robust strategy to distinguish between bare and functionalized layers and significantly improves the reproducibility between measurements. Remarkably, it is also confirmed by statistical analysis that adhesion values do not depend on the thickness of the layers, proving that they are related only to the most superficial part of the materials. In addition, we have implemented an unsupervised classification method using k-means clustering, an artificial intelligence-based algorithm, to automatically classify samples based on adhesion values. These results demonstrate the potential of simple adhesion AFM measurements to inspect the chemical nature of 2Dms and may have implications for the broad scientific community working in the field.

Unsupervised Classification of Landsat-8 Satellite Imagery-Based on ISO Clustering

Remote sensing, specifically satellite imagery, is gaining prominence in computer science nowadays, in the era of artificial intelligence, in an attempt to deliver more precise information. The satellite images of Earth are gathered, evaluated, and processed for use in civil and military applications with a military aim. Satellite images do have a wide range of services. The areas of study of agriculture, fishery, oceanography, and meteorology include geology, biodiversity, cartography, land use planning, and armed conflict. Transformation is the goal of the categorization of satellite images. Transformation of satellite images into information that can be used rather than having an image of a location. This paper classified a scene of the Landsat-8 satellites with specifications (Path=168 and Row=38). This scene was classified into four categories (Water, Vegetation, bare land, and Build-up) based on the unsupervised classification method (ISO Clustering). The ISO Clustering method is found in the Arc Map program. The results regarding classification accuracy are a good percentage compared to unsupervised Classification.

Multitemporal Analysis of Seagrass Dynamics on Derawan Island (2003–2021) Using Remote Sensing Techniques

The shallow waters around Derawan Island are renowned for their beauty, attracting a significant number of tourists. Since the 2008 National Sports Week (PON) in East Kalimantan, the construction of inns and jetties has enhanced both accommodation and accessibility on the island. However, this development has also impacted the seagrass beds in the surrounding shallow waters. This study examines the changes in the area and density of seagrass beds from 2003 (prior to the PON activities) through to 2011 (a few years post-PON) and in 2021 (the most recent conditions), assessing the effects of lodging and jetty construction on these beds. Data were collected via field surveys using the photo transect method, and the benthic habitat map was created using Landsat 8 OLI Imagery, applying the Lyzenga water column correction algorithm and unsupervised classification method. The Normalized Difference Building Index (NDBI) algorithm and land digitization were utilized to track the development of the inns and jetties, revealing a rapid, widespread increase in construction throughout the island's southern region (R-square = 0.59). The study findings indicate a significant degradation of seagrass meadows between 2003 and 2021, particularly near populated areas on the southern coast, resulting in decreased density levels.

Classification of Red Blood Cells From a Geometric Morphometric Study

Sickle cell disease causes the deformation of erythrocytes into sickle cells. The study of this process using digital images of peripheral blood smears can help specialists to quantify the number of deformed cells in order to gauge the severity of the illness. A new method for classifying red blood cells into three categories: healthy, sickle cell disease, and other deformations is proposed. This method does not require obtaining the contour of each cell but instead utilizes information obtained from a small number of points, obtained through appropriate geometric sampling and the use of stereological formulas. The parameters utilized for classification are the bending energy times length (E) and the circular shape factor (F). In normal cells, which exhibit an almost circular shape, these parameters typically have values close to (1,1). To assess the effectiveness of classification using the parameters (E,F), a synthetic curve dataset and a dataset of red blood cells are employed, applying various supervised and unsupervised classification methods.

Finite mixture models for multiscale dynamics in Chinese financial markets

A novel statistical machine learning technique for decomposing, analyzing, and modeling complex time series is presented in this paper. This methodology is based on unsupervised classification methods using finite mixtures, which have been widely used in factor analysis to extract essential information from multivariate static data in a parsimonious manner. Here, they are applied in a purely dynamic framework to create a new family of autoregressive models. The selection of the number of components and appropriate geometric template in these models is based on a set of information criteria that require solving a multiple objective optimization problem. To ensure the most adequate models are selected, the problem is augmented with a simulation instruction, providing a robust criterion for assignment. Experiments on real financial data sampled over the COVID-19 pandemic demonstrate the potential of this approach as a promising alternative to the well-known Box-Jenkins methodology. The proposed method offers flexibility, accuracy, and stability, making it suitable for a variety of applications in financial modeling and other fields where complex time series data analysis is required.

Potentiality of Landsat-9 for early-season mapping of winter garlic and winter wheat

ABSTRACT Landsat-9 is the latest satellite of the Landsat program and provides useful agricultural remote sensing data. The earlier the crop distribution data is obtained, the greater the value of the data. However, whether Landsat-9 image is suitable for the early-season mapping of winter garlic and winter wheat is a problem worthy of attention. Therefore, this study evaluates the potential to use two Landsat-9 images, acquired on 13 December 2021 and 30 January 2022, for the early-season identification of winter garlic and winter wheat in China. According to J-M (Jeffries-Matusita) distances, we evaluated the separability of winter garlic, winter wheat, and other ground objects based on the two Landsat-9 images. Then, winter garlic and winter wheat were extracted by using unsupervised classification method, i.e. the IsoData and K-means clustering algorithms, and supervised classification method, i.e. the Random Forest (RF), and Support Vector Machine (SVM) algorithms. The separability between garlic and wheat in January is stronger than that in December. The classification overall accuracy based on the Landsat-9 image on 30 January 2022 is 92.03% with a kappa coefficient of 0.87 using the SVM algorithm. This period is about 4 months earlier than the crop harvest period. Landsat-9 images have good potentiality for early-season mapping of winter garlic and winter wheat.

An unsupervised mechanical fault classification method under the condition of unknown number of fault types

An unsupervised mechanical fault classification method under the condition of unknown number of fault types

Learning adaptive Grassmann neighbors for image-set analysis

Representing image sets as subspaces on Grassmann manifold and leveraging the Riemannian geometry of this space has proven to be highly effective in various visual recognition tasks. However, the computational cost associated with operating on Riemannian manifolds, especially of high-dimensional ones, poses a significant limitation to the applicability of existing techniques. Moreover, the image sets with labels are scarce and valuable in real-world scenarios, where the manual annotation is time-consuming and costly. To tackle these issues, in this paper, we propose an unsupervised Grassmannian dimensionality reduction method for image set classification and clustering, which simultaneously performs local structure learning and dimensionality reduction under the subspace view. Specifically, instead of relying on the pre-defined similarity matrix that obtained from raw data and may contain much noise, we adaptively learn the local structures from the projected low-dimensional Grassmann manifold. The refined structures then provide more discriminative power to learn the accurate manifold-to-manifold mapping. To better preserve the global structure of the data, this mapping is constrained to satisfy the orthogonal condition. In addition, our model can be compatible with F-norm OLPP for Grassmann manifold when the squared F-norm is transformed into projection metric. A re-weighted iterative scheme is carefully developed to optimize the problem. Experimental results on several benchmark datasets for image set classification and clustering tasks clearly demonstrate that our model consistently outperforms baselines, yielding significant improvements.

Rapid identification of bacterial isolates using microfluidic adaptive channels and multiplexed fluorescence microscopy.

We demonstrate the rapid capture, enrichment, and identification of bacterial pathogens using Adaptive Channel Bacterial Capture (ACBC) devices. Using controlled tuning of device backpressure in polydimethylsiloxane (PDMS) devices, we enable the controlled formation of capture regions capable of trapping bacteria from low cell density samples with near 100% capture efficiency. The technical demands to prepare such devices are much lower compared to conventional methods for bacterial trapping and can be achieved with simple benchtop fabrication methods. We demonstrate the capture and identification of seven species of bacteria with bacterial concentrations lower than 1000 cells per mL, including common Gram-negative and Gram-positive pathogens such as Escherichia coli and Staphylococcus aureus. We further demonstrate that species identification of the trapped bacteria can be undertaken in the order of one-hour using multiplexed 16S rRNA-FISH with identification accuracies of 70-98% with unsupervised classification methods across 7 species of bacteria. Finally, by using the bacterial capture capabilities of the ACBC chip with an ultra-rapid antimicrobial susceptibility testing method employing fluorescence imaging and convolutional neural network (CNN) classification, we demonstrate that we can use the ACBC chip as an imaging flow cytometer that can predict the antibiotic susceptibility of E. coli cells after identification.

Applying Geo-Spatial Techniques to Determine the Temporal Variation and Conversion Pattern of Land-Use Surrounding Negombo Lagoon

Remote Sensing technology and Geographic Information Systems are the major partners in geospatial technology and they play a vital role in surface, subsurface and above earth’s studies. It provides a better platform for researchers specially in the fields of regional planning, disaster management, natural resources management etc. Observing land-use patterns in a particular area forms a basis for decision making in these fields. The study has the purpose of applying geo-spatial techniques to determine the temporal variation and conversion pattern of land-use surrounding Negombo lagoon. For that, the study has used Landsat imageries of 2003, 2009, 2015 and 2021 from the United States Geological Survey. ISO Cluster Unsupervised Classification method and data conversion tools were used to create initial layers for analysing the land-use in selected four years by using Arc GIS 10.4. software. Intersect and field calculators were used to get a better understanding of the variation of land-use through classified Landsat imageries of respective years. The results will support for measuring area of the conversion of different land-use into another and / or within the same category in a particular area throughout the time. The study concluded that, the used techniques could measure only the temporal variation of major land use types; water bodies, vegetation, built-up areas and cultivations but the minor categories of land uses were unable to detect. However, the better spatial and temporal resolution data enables to improve results to avoid such issues.

SPATIO-TEMPORAL DYNAMICS OF URBAN THERMAL ENVIRONMENT IN UDAIPUR CITY, RAJASTHAN, INDIA

Abstract. The Udaipur urban agglomeration was selected to analyse the urban heat island (UHI) effect in the area from 2005 to 2017. Landsat 7 ETM+ sensor data was used to derive land surface temperatures (LST) and map landscape characteristics. The agglomeration was classified into seven land use land cover classes including agriculture cropped, agriculture fallow, barren, built up, scrub, vegetation and water using unsupervised classification method. The LST results were obtained using NDVI method. Urban heat island intensities were mapped using z-score method. Regions with UHI values of above 2 standard deviations were considered to reflect UHI effect. Results show an overall decrease of 10.2 percent in agricultural cropped and fallow lands. The scrub class also shows a moderate decline of 2.3 percent. Increase in area were observed for built up, barren and vegetation classes by 8.8, 2.1 and 1.9 percent respectively. The dominant LULC change was the transformation of agricultural lands to built-up class. The class-wise mean LST observations increase in order of water, vegetation, cropped, fallow, built up, scrub and barren land. The mean LST records an increase of 1.7° C from 30.1° C to 31.8° C over the study period. Results suggest that UHI effects are more prominent in urban fringe area corresponding with the built up and barren land cover classes contrary to its core area. Built up areas surrounded by barren lands left vacant for city’s sprawl in the future exhibit highest UHI intensities.

Unsupervised domain adaptation for SAR target classification based on domain- and class-level alignment: From simulated to real data

Synthetic Aperture Radar (SAR) target classification methods based on convolutional neural networks have attracted wide attention recently. Typically, these methods necessitate a substantial quantity of labeled data to train the deep model, which is an obstacle for SAR target classification since annotating SAR images is costly and time-consuming. Generating large-scale labeled data in simulation to train the deep model is a feasible alternative. However, cross-domain recognition is challenging due to significant differences in distribution between simulated and real domains. In this paper, we propose an unsupervised domain adaptation method for SAR target classification by aligning the distribution between the simulated and unlabeled real domains. Specifically, we encode the dependencies across different granularity perspectives including domain-, and class-levels simultaneously to align two domains finely. On the one hand, a gradient-weighted adversarial alignment method is proposed to align the global distributions at the domain-level. On the other hand, a novel class-level alignment method based on the prototypical network is proposed to align the finer class structure. Technically, we respectively model the cross-domain prototype-prototype relations and cross-domain prototype-instance relations with contrastive learning to achieve intra-class prototype compact as well as inter-class prototype separation among cross-domain categories, where the prototypes of the real domain are learned with pseudo labels. To improve the quality of pseudo-labels, a pseudo-label filtering strategy combined with the attribute scattering center is further designed. The experiments on two simulation-to-reality datasets indicate that the proposed method outperforms state-of-the-art unsupervised domain adaptive methods, such as pixelfeat (+4.45% and +7.6% in accuracy on SAMPLE and S2M datasets respectively). Code and data will be available in https://github.com/YuShi1213/Sim2Real-Unsupervised-SAR-Target-Classification.

Unsupervised Classification under Uncertainty: The Distance-Based Algorithm

This paper presents a method for unsupervised classification of entities by a group of agents with unknown domains and levels of expertise. In contrast to the existing methods based on majority voting (“wisdom of the crowd”) and their extensions by expectation-maximization procedures, the suggested method first determines the levels of the agents’ expertise and then weights their opinions by their expertise level. In particular, we assume that agents will have relatively closer classifications in their field of expertise. Therefore, the expert agents are recognized by using a weighted Hamming distance between their classifications, and then the final classification of the group is determined from the agents’ classifications by expectation-maximization techniques, with preference to the recognized experts. The algorithm was verified and tested on simulated and real-world datasets and benchmarked against known existing algorithms. We show that such a method reduces incorrect classifications and effectively solves the problem of unsupervised collaborative classification under uncertainty, while outperforming other known methods.

Gold mining's environmental footprints, drivers, and future predictions in Ghana

The last two decades have seen a surge in gold mining operations around the world. Despite mining occupying a smaller geographical area compared to other land use/land cover (LULC) classes, it exhibits strong interconnections with various land uses and serves as a major driver for changes in mining landscapes. Understanding and evaluating historical and potential future LULC changes in these landscapes are crucial in assessing the environmental impact of mining. Traditionally, these assessments heavily rely on geospatial techniques, with limited emphasis on projecting future LULC trends. This research aims to monitor, analyse the drivers of change, and predict future changes in LULC under two scenarios: the “business as usual” scenario and the "remedial measures" scenarios. Utilising the CA-Markov model, this article predicts LULC changes and offers comprehensive insights into the environmental impacts of mining, combining geospatial and social research methodologies. The investigation spanned a 34-year period (1986–2020) and employed a blend of supervised and unsupervised image classification methods, complemented by interviews, focus groups, and field observations. The findings reveal substantial land degradation, water pollution, and a significant loss of forest cover, accounting for 27,333 ha (36%). Continuation of current mining practices is predicted to lead to further ecological deterioration.