Gaussian Mixture Model Clustering Research Articles

Shoreline Extraction Methods from Sentinel-2 and PlanetScope Images

Abstract. This work aims to compare and assess the performance of certain methodologies for shoreline mapping based on the use of medium (10 m) and high resolution (3 m) multi-spectral imagery, provided by Sentinel-2 (S2) and PlanetScope (PS), respectively. Being Sentinel-2 part of the Copernicus missions, its data are freely available. PS imagery are also freely available for scientific research, upon approval by the European Space Agency of a related project proposal. Several spectral indices, including Normalized Difference Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI), Automated Water Extraction Index (AWEI), and Water Index (WI), were used for shoreline detection. In particular, two unsupervised classification techniques, the Gaussian Mixture Model (GMM) and K-means clustering were deployed as shoreline extraction methods. The outcomes of such approaches were validated using reference shorelines derived from aerial orthomosaics, generated from images acquired as close as possible to the satellite imagery dates, and the ”baseline and transect” approach for accuracy verification. Three tide-less Mediterranean beaches were used as study cases for comparison: the beach between Castelldefels and Gava in Spain, Feniglia and Marina di Grosseto in Italy. The results demonstrated sub-pixel accuracy in shoreline extraction, with Mean Absolute Distances ranging from 2 m to 5 m for S2 data and 1.5 m to 2 m for PS data. These findings highlight the potential of freely available satellite data for semi-automatic shoreline detection. Results obtained by using the combination of different indices and methodologies show that the best option may change depending on the considered context, hence future investigations should be dedicated to the development of a procedure for automatically determining the context-based (close to) optimal index-classifier combination.

Reconstruction of measurement data with multiple outliers using novel domain-based RBF

Due to the high accuracy of computation, radial basis function (RBF) is widely recognized as a versatile and effective method for interpolating and approximating discrete points in various fields. However, RBF is quite sensitive to outliers, which can easily lead to distorted results. In this article, a novel overlapped domain-based RBF (ODRBF) method is proposed, in which the concept of effective domain is introduced to build a moving model, and Student’s t-regression and Gaussian mixture model (GMM) clustering are used for dealing with local anomalies. By introducing the effective domain, the estimated points and domain radius are constructed and the global model can be transformed into local estimation models. In each effective domain, a series of estimation models are iteratively generated through Student’s t-regression, and based on the distances between the estimation model and discrete points, GMM clustering is used to subsequently select the data as the input of the next regression. This iterative strategy in each effective domain ensures the removal of multiple outliers. Then, the preserved points in the processed effective domain are used to obtain local estimated value by RBF. The proposed method demonstrates strong robustness to highly contaminated dataset in the reconstruction of the simulation and experimental datasets.

Data-driven distributionally robust day-ahead dispatch for active distribution networks based on improved conditional generative adversarial network

With the increasing penetration of renewable energy generation (REG), strong uncertainty poses a significant challenge to the rationality of day-ahead dispatch plans in distribution systems. We propose a data-driven, distributionally robust day-ahead dispatch method for active distribution networks (ADNs) based on an improved conditional generative adversarial network (CGAN). First, an improved CWGAN-GP model is constructed based on three-dimensional convolution (Conv3D) and the Wasserstein distance to obtain the day-ahead REG scenarios, considering the correlation features of photovoltaic (PV) and wind turbines (WTs), which reduces the conservatism of scenario generation. Accordingly, the Gaussian mixture model (GMM) clustering method based on moment information was used to construct an ambiguity set. Moreover, a two-stage distributionally robust day-ahead dispatch model for an ADN is established, which realizes the fusion of the data-driven method and distributionally robust optimization (DRO) mechanism model. Finally, the validity of the data-driven DRO (DDRO) method is verified through a case analysis.

Applying Gaussian Mixture Models for Enhanced Characterization of Featured Surfaces and Mixed Lubrication Analysis

Abstract Understanding surface topography is vital for optimizing the performance of engineering components. Featured surfaces, with distinct patterns and textures, have garnered attention for their potential to reduce friction and wear. However, accurately describing their topography poses challenges, necessitating effective segmentation methods in many applications. This paper proposes utilizing the Gaussian Mixture Models (GMMs) clustering method as a novel approach for surface metrology analysis of featured surfaces. The GMMs provide an approach to identify and analyze specific surface features and enhance comprehension of their contributions to functionality. The paper presents a comprehensive methodology involving surface characterization, GMMs clustering, plateau reference plane location, and calculation of essential topography parameters. Results from four different surfaces are discussed, demonstrating the effectiveness of the proposed GMMs-based methodology in segmenting plateau regions, grooves, and porosity.

A bridge damage monitoring and assessment strategy based on Gaussian mixture modelling

For the bridges equipped with health monitoring systems, it is essential to precisely assess the health status of the bridge from the data with noise and outliers. An improved Gaussian mixture model clustering algorithm is used to process the obtained bridge strain sensor monitoring data and generate the characteristic data of clustering in order to improve the accuracy of data analysis of bridge health monitoring system. The cluster and feature data are obtained by Expectation Maximization process, and the isolated clusters are filtered out by the threshold parameters of weight and the Euclidean distance between clusters center. Based on the feature data, a scoring strategy is established to assess the bridge health and sensor status. The proposed strategy is used to analyze the monitoring data of a bridge in western China, and the analysis results show the availability of this strategy. Compared with the directly collected data, the processed bridge health scoring curve variation is smooth, which can reduce the influence of noise and abnormal data. The sensor status scoring curve can track changes in collected data and respond to different transformation scenarios. This shows that the proposed strategy can evaluate the state of this bridge and the sensors. provided a reference for bridge maintenance decision during the operational period.

Capacity Demand Analysis of Rural Biogas Power Generation System with Independent Operation Considering Source-Load Uncertainty

With the greatly increased penetration rate of wind power, photovoltaic, and other new energy sources in the power system, the proportion of controllable units gradually decreased, resulting in increased system uncertainty. The biogas power generation system can effectively alleviate the pressure caused by source-load uncertainty in such high-permeability systems of new energy sources such as wind power and photovoltaic. Hence, from the perspective of the power system, this paper introduces a capacity demand analysis method for a rural biogas power generation system capable of independent operation amidst source-load uncertainty. To enhance the depiction of pure load demand uncertainty, a scene set generation method is proposed, leveraging quantile regression analysis and Gaussian mixture model clustering. Each scene’s data and probability of occurrence elucidate the uncertainty of pure load demand. An integrated optimal operation model for new energy and biogas-generating units, free from energy storage capacity constraints, is established based on the generated scenario set. Addressing considerations such as biogas utilization rate and system operation cost, a biogas storage correction model, utilizing the gas storage deviation degree index and the cost growth rate index, is developed to determine biogas demand and capacity. The example results demonstrate the significant reduction in gas storage construction costs and charging and discharging imbalances achieved by the proposed model while ensuring systemic operational cost effectiveness.

Multi-objective evolutionary algorithm with evolutionary-status-driven environmental selection

The hardly dominated boundary (HDB) is a common feature of multi-objective optimization problems (MOPs). Previous studies have proposed several multi-objective evolutionary algorithms (MOEAs) to deal with the problem characterized by HDBs. Nevertheless, these methods lack consideration of the evolutionary status, potentially resulting in low efficiency when addressing problems that entail a confluence of HDB and other intricate characteristics. This paper proposes an MOEA with evolutionary-status-driven environmental selection (MOEA-ESD) to address such an issue. Specifically, in each generation, the current and historical information are used to evaluate the evolutionary status. Based on the estimated evolutionary status, the proposed MOEA-ESD algorithm adaptively uses three types of environmental selection: the traditional environmental selection (TES) based on Pareto-dominance and crowding distance, a convergence-first environmental selection (CFES) based on Gaussian mixture model clustering, and a diversity-first environmental selection (DFES) based on outlier detection and extreme solutions. In this way, inferior solutions situated on the HDB can be effectively eliminated, thereby facilitating the convergence of the population while concurrently preserving a commendable degree of diversity. Moreover, a set of new benchmark problems with different objective magnitudes and complicated Pareto sets is developed to enrich the features of HDB-MOPs and to verify the algorithm's performance. Our experimental results on 22 HDB-MOPs show the promising performance of the proposed algorithm. The source code of MOEA-ESD is available at https://github.com/CIAM-Group/EvolutionaryAlgorithm_Codes/tree/main/MOEA-ESD.

Nucleus segmentation of white blood cells in blood smear images by modeling the pixels' intensities as a set of three Gaussian distributions.

The precise segmentation of white blood cells (WBCs) within blood smear images is a significant challenge with implications for both medical research and image processing. Of particular importance is the often neglected task of accurately segmenting WBC nuclei, an aspect that currently lacks dedicated methodologies. This paper introduces a straightforward and efficient method designed to fill this critical gap, providing an effective solution for the efficient segmentation of WBC nuclei. In blood smear imagery, the distinctive coloration of WBCs contrasts with the hues of other blood components. The inherent obscurity of WBCs prompts their segmentation by isolating pixels with minimal intensities. To streamline this process, our proposed method employs the Laplacian pyramid technique to decorrelate pixels in blood smear images, thereby amplifying the contrast. Subsequently, the intensities of pixels constituting blood cells, encompassing WBCs and the background, are modeled using three Gaussian random variables. Capitalizing on this feature, we implement the Gaussian mixture model (GMM) clustering method to determine the optimal threshold value, facilitating a highly precise segmentation of WBC nuclei. The proposed method demonstrates the capability to process images containing a single WBC as well as effectively functioning with images containing multiple cells of this type. Evaluation of the method on the ALL-IDB, ALL-IDB2, CellaVision, and JTSC datasets yielded accuracy values of 0.9802, 0.9725, 0.9772, and 0.9730, respectively. Comparative analysis with state-of-the-art methods revealed a notably comparable performance, underscoring the effectiveness of the proposed approach. The method presented in this article is highly competitive for segmenting the nuclei of WBCs compared to state-of-the-art methods. The three main advantages of our method are its ability to process images containing one or more WBCs, the automatic calculation of threshold values for each processed image, eliminating the need for manual parameter adjustments. Lastly, the method is efficient, as its algorithmic complexity is approximately .

A Richer Vocabulary of Chinese Personality Traits: Leveraging Word Embedding Technology for Mining Personality Descriptors.

This study uses a data-driven approach to mine the distribution of personality traits among Chinese people in the Chinese social context. Based on the hypothesis of personality lexicology, word embedding technology was employed in machine learning to mine personality vocabulary from Tencent's word embedding database. More than 10,000 Chinese personality descriptors were extracted and analyzed using Gaussian Mixture Model Cluster and Hierarchical clustering analysis. The data was collected from 658 Chinese people randomly from all parts of China through an online questionnaire method. The results reveal six personality traits in the Chinese context, expanding the personality thesaurus and providing examples to illustrate each trait. The findings coincide with previous research on the five-factor model, which partially describes the personality traits of Chinese people, but does not offer a complete explanation of their typical social behavior patterns. Additionally, the study supports the notion of cultural particularity in personality traits. The approach used in this study offers a richer personality vocabulary than traditional personality mining methods, and word embedding technology captures richer semantic information in Chinese. The six Chinese personality traits identified in this study will also be used to explore how to quantify and evaluate personality traits based on word embedding and personality descriptors.

Specular highlight removal of light field image combining dichromatic reflection with exemplar patch filling

In this paper, a new method based on dichromatic reflection model (DRM) and exemplar patch is proposed to remove highlight pixels for light field images. Firstly, a Gaussian mixture model clustering method with strong generalization performance combined with depth information is used to classify saturated highlight pixels and unsaturated highlight pixels. A confidence strategy based on DRM is proposed to remove unsaturated highlight pixels, and an exemplar patch matching method based on gradient combined with the sum of square of color difference is designed to remove saturated highlight pixels. Meanwhile, a method named SSIME based on information entropy with structural similarity index measure is designed to quantitatively evaluate the effectiveness of proposed method. Experiments show that our proposed method not only effectively detects and removes specular highlights, but also applies to objects with a large range of highlights, which breakthroughs the limitation that large-region specular highlight cannot be removed due to the short baseline of the light field camera.

Damage evolution analysis of concrete based on multi-feature acoustic emission and Gaussian mixture model clustering

In this work, a novel method for damage evolution analysis of concrete under uniaxial compression is proposed based on the multi-feature acoustic emission (AE) and the Gaussian mixture model (GMM) clustering. The hierarchical clustering algorithm is adopted to select optimal AE parameters, while multiple features of these parameters are generated through the principal component analysis (PCA). Then the concrete damage signals are separated by using the GMM clustering with multiple features. Meanwhile, the waveform signals associated with concrete damage in each cluster are validated using the Fast Fourier transform (FFT) and the continuous wavelet transform (CWT). Finally, the damage evolution process of concrete under uniaxial compression is divided based on the variation of AE characteristics. The results show that the frequency ranges of signals for micro-cracks, mixed cracks and friction of concrete under uniaxial compression are 20–30 kHz, 20–65 kHz, and 100–120 kHz, respectively. The damage process of concrete under uniaxial compression is divided into three stages according to the inflection points of the cumulative energy curve, namely the initial compaction stage, the crack formation stage and the failure stage.

A comprehensive evaluation of OPTICS, GMM and K-means clustering methodologies for geochemical anomaly detection connected with sample catchment basins

The process of data-driven clustering to uncover geochemical anomalies linked to sample catchment basins (SCBs) includes a comprehensive framework to discern areas exhibiting unique geochemical attributes within a specified study area. The Ordering Points to Identify the Clustering Structure (OPTICS) method can serve as a robust methodology for detecting geochemical anomalies in SCBs. This is attributed to its capacity to effectively manage varying cluster densities, adaptively identify cluster numbers, exhibit resilience to noise, and display minimum sensitivity to parameters. A comparison was conducted in this research between the outcomes of the OPTICS clustering algorithm and two traditional clustering techniques, namely the Gaussian Mixture Model (GMM) and K-means clustering. In the following, the Expectation-Maximization (EM) technique is employed to train the GMM for clustering. Moreover, the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) as two validate statistical metrics implemented to ascertain the optimal number of components (clusters) belong to the GMM. It should be noted that the effectiveness of the clustering algorithms was further assessed using the Calinski-Harabasz (CH) index and the success-rate curves. OPTICS, a density-based clustering approach, was confirmed to be more effective than K-means and GMM for identifying MVT PbZn anomalies in Varcheh district, western Iran. Furthermore, the specified anomalies show a geo-spatial correspondence with the geological facts, and it has been observed that strong anomalies are more discoverable in close proximity to MVT PbZn occurrences. This work suggests a novel anomaly detection approach based on OPTICS, which exhibits superior performance and data-modeling efficiency. The main emphasis is on effectively distinguishing geochemical anomalies from sample data originating from populations with uncertain distributions.

Interpretable Lost Circulation Analysis: Labeled, Identified, and Analyzed Lost Circulation in Drilling Operations

Summary Lost circulation (LC) is a serious problem in drilling operations, as it increases nonproductive time and costs. It can occur due to various complex factors, such as geological parameters, drilling fluid properties, and operational drilling parameters, either individually or in combination. Therefore, studying the types, influencing factors, and causes of LC is crucial for effectively improving prevention and plugging techniques. Currently, the expert diagnosis of LC types relies heavily on the experience and judgment of experts, which may lead to inconsistencies and biases. Additionally, difficulties in obtaining data or missing important data can affect the efficiency and timeliness of diagnosis. Traditional physical modeling methods struggle to analyze complex factor correlations, and conventional machine learning techniques have limited interpretability. In this paper, we propose an interpretable lost circulation analysis (ILCA) framework that provides a new method for analyzing LC. First, we use Gaussian mixture model (GMM) clustering to analyze the LC characteristics of regional case data, efficiently and accurately labeling 296 LC events. Second, we establish the relationship between geological features, drilling fluid properties, operational drilling parameters, and LC types using the XGBoost algorithm. This enables timely identification of LC types during drilling operations using real-time data, with a precision greater than 85%. Finally, we use interpretable machine learning techniques to conduct a comprehensive quantitative analysis of influencing factors based on the established XGBoost model, providing a clear explanation for the identification model. This enables drilling engineers to gain deeper insights into the factors influencing LC events. In summary, the proposed ILCA framework is capable of efficiently labeling LC types based on regional case data, identifying LC types in a timely manner using real-time data, and conducting quantitative analysis of the factors and causes of LC. This approach addresses the limitations of traditional methods and offers valuable insights for drilling engineers.

PV power prediction based on feature selection and GMM clustering

The increase in the installed capacity of photovoltaics (PV) brings great challenges to the stable operation of the power system. To improve the accuracy of PV power prediction, Spearman’s rank correlation coefficient and Grey Relation analysis (GRA) are combined to analyze and screen each meteorological feature to ensure the effectiveness and reasonableness of the input features. Based on the analysis of meteorological features, the Gaussian Mixture Models (GMM) clustering method is proposed to cluster the historical data sets. The historical data are clustered into two categories: simple PV power change and complex PV power change, which facilitates the model to learn the power change characteristics under different weather conditions. A Bidirectional Gated Recurrent Unit (BiGRU) prediction model is built to mine the timing properties of the two types of data and trained and tested separately. By setting several error evaluation indexes for comprehensive judgment, the methods in this paper are better than several other classical prediction methods and have a strong adaptability to different weather conditions.

Generating Multi-Codebook Neural Network by Using Intelligent Gaussian Mixture Model Clustering Based on Histogram Information for Multi-Modal Data Classification

One of the open challenges in machine learning is multi-modal data classification. A classifier model needs to be enhanced to deal with multi-modal data. This study is proposed to develop multi-codebook neural networks using intelligent Gaussian mixture model clustering for multi-modal data classification. The intelligent Gaussian mixture model clustering is developed in this study prior to the development of multi-codebook models. The method analyzes the gradient of input data histogram to find the number of generated mixtures and cluster the data. The proposed multi-codebook neural network model has three variants based on the rules to fund the number of clusters. The experiment result showed that the proposed three variants of multi-codebook models performed well in the synthetic and benchmark datasets. The proposed model improved the original method by 24.14%, 15.97%, and 3.71% accuracy and 0.3510, 0.0487, and 0.2031 kappa for synthetic datasets, benchmark datasets, and overall datasets respectively. By using the ANOVA test, we have proved that all three variants of the proposed multi-codebook neural network were proved to have significant improvements compared to the original version.

Correlative Scan Matching Position Estimation Method by Fusing Visual and Radar Line Features

Millimeter-wave radar and optical cameras are one of the primary sensing combinations for autonomous platforms such as self-driving vehicles and disaster monitoring robots. The millimeter-wave radar odometry can perform self-pose estimation and environmental mapping. However, cumulative errors can arise during extended measurement periods. In particular scenes where loop closure conditions are absent and visual geometric features are discontinuous, existing loop detection methods based on back-end optimization face challenges. To address this issue, this study introduces a correlative scan matching (CSM) pose estimation method that integrates visual and radar line features (VRL-SLAM). By making use of the pose output and the occupied grid map generated by the front end of the millimeter-wave radar’s simultaneous localization and mapping (SLAM), it compensates for accumulated errors by matching discontinuous visual line features and radar line features. Firstly, a pose estimation framework that integrates visual and radar line features was proposed to reduce the accumulated errors generated by the odometer. Secondly, an adaptive Hough transform line detection method (A-Hough) based on the projection of the prior radar grid map was introduced, eliminating interference from non-matching lines, enhancing the accuracy of line feature matching, and establishing a collection of visual line features. Furthermore, a Gaussian mixture model clustering method based on radar cross-section (RCS) was proposed, reducing the impact of radar clutter points online feature matching. Lastly, actual data from two scenes were collected to compare the algorithm proposed in this study with the CSM algorithm and RI-SLAM.. The results demonstrated a reduction in long-term accumulated errors, verifying the effectiveness of the method.

Deep federated learning hybrid optimization model based on encrypted aligned data

Federated learning can achieve multi-party data-collaborative applications while safeguarding personal privacy. However, the process often leads to a decline in the quality of sample data due to a substantial amount of missing encrypted aligned data, and there is a lack of research on how to improve the model learning effect by increasing the number of samples of encrypted aligned data in federated learning. Therefore, this paper integrates the functional characteristics of deep learning models and proposes a Variational AutoEncoder Gaussian Mixture Model Clustering Vertical Federated Learning Model (VAEGMMC-VFL), which leverages the feature extraction capability of the autoencoder and the clustering and pattern discovery capabilities of Gaussian mixture clustering on diverse datasets to further explore a large number of potentially usable samples. Firstly, the Variational AutoEncoder is used to achieve dimensionality reduction and sample feature reconstruction of high-dimensional data samples. Subsequently, Gaussian mixture clustering is further employed to partition the dataset into multiple potential Gaussian-distributed clusters and filter the sample data using thresholding. Additionally, the paper introduces a labeled sample attribute value finding algorithm to fill in attribute values for encrypted unaligned samples that meet the requirements, allowing for the full recovery of encrypted unaligned data. In the experimental section, the paper selects four sets of datasets from different industries and compares the proposed method with three federated learning clustering methods in terms of clustering loss, reconstruction loss, and other metrics. Tests on precision, accuracy, recall, ROC curve, and F1-score indicate that the proposed method outperforms similar approaches.

Reconstructing palaeodiet from the skeleton with stable isotopes - on the interpretation of multi-dimensional datasets.

The analysis of stable isotopes in archaeological human and animal skeletal finds is a routine method in the field of archaeometry. Multi-dimensional datasets are this way easily generated what, however, demands new interpretive approaches. Cluster analyses are capable of detecting multi-dimensional similarities among the isotopic signatures and are thus superior over traditional uni- and bivariate evaluations. This study aims to demonstrate the potential of Gaussian Mixture Model (GMM) clustering for palaeodiet reconstruction. We reevaluate previously published isotopic datasets obtained on early medieval skeletal finds from Upper Bavaria (Germany), also including some individuals from "special burials", e.g., from separated burial sites or circular enclosures. This manageable and easy-to-understand dataset contains four-dimensional isotope signatures (δ13Ccollagen, δ15N, δ13Ccarbonate, δ18Ocarbonate) of 27 individuals. We illustrate the iterative process of cluster analysis and the additional benefit of its outcome for the interpretation of the data. This includes the recognition of noisy isotope data and of those isotope systems that are essential. In our example, three isotope systems were required for a correct palaeodiet reconstruction. While δ13Ccollagen and δ15N values are commonly used for palaeodiet reconstructions, δ13Ccarbonate values, including additional information on dietary carbohydrates and fats, are required for a meaningful cluster separation and, thus, the interpretation of the individuals' diet. No general correlation between dietary quality and grave goods was detectable.

Integrated Spatio-Temporal Deep Clustering (ISTDC) for cognitive workload assessment

Traditional high-dimensional electroencephalography (EEG) features (spectral or temporal) may not always attain satisfactory results in cognitive workload estimation. In contrast, deep representation learning (DRL) transforms high-dimensional data into cluster-friendly low-dimensional feature space. Therefore, this paper proposes an Integrated Spatio-Temporal Deep Clustering (ISTDC) model that uses DRL followed by a clustering method to achieve better clustering performance. The proposed model is illustrated using four Algorithms and Variational Bayesian Gaussian Mixture Model (VBGMM) clustering method. Temporal and spatial Variational Auto Encoder (VAE) models (mentioned in Algorithm 2 and Algorithm 3) learn temporal and spatial latent features from sequence-wise EEG signals and scalp topographical maps using the Long short-term memory and Convolutional Neural Network models. The concatenated spatio-temporal latent feature (mentioned in Algorithm 4) is passed to the VBGMM clustering method to efficiently estimate workload levels of n-back task. For the 0-back vs. 2-back task, the proposed model achieves the maximum mean clustering accuracy of 98.0%, and it improves by 11.0% over the state-of-the-art method. The results also indicate that the proposed multimodal approach outperforms temporal and spatial latent feature-based unimodal models in workload assessment.

A clustering (DBSCAN+GMM) investigation of the young open cluster NGC 6649

ABSTRACT We investigate the spatial structure and dynamical properties of the young (∼60 Myr), richly populated open cluster NGC 6649 based on reliable cluster members from Gaia-DR3 data. The density-based spatial clustering of applications with noise is used to estimate membership probabilities for 25045 sample stars in the field of the cluster. More than 1000 likely cluster members down to G ∼ 20 mag can be identified based on the membership probabilities. The cluster members of NGC 6649 extend up to a radial distance of ${\sim} 21\,\mathrm{ arcmin}$ (12 pc), which is much larger than previously thought. We estimate a core radius of $R_{\mathrm{ c}}=2\overset{^{\prime }}{.}38\pm 0\overset{^{\prime }}{.}06$ (1.4 pc) and a tidal radius of $R_{\mathrm{ t}}=35\overset{^{\prime }}{.}85\pm 6\overset{^{\prime }}{.}64$ (20.9 pc) for the cluster, indicating that the cluster has formed a dense core and an extended halo. We find that mass segregation has taken place within the cluster. We also detect two small and sparse clusters (C1 and C2) around NGC 6649. C1 has recently been reported, but C2 is detected for the first time. The Gaussian mixture model clustering method is used to identify 184 and 60 members for C1 and C2, respectively. We find that NGC 6649, C1, and C2 are located at similar distances. Moreover, we estimate a small spatial separation of ∼11.6 pc between C1 and NGC 6649, which is close to or slightly smaller than the detectable radius of NGC 6649. This indicates that NGC 6649 and C1 are physically connected, and there may exist strong tidal interactions and mass transfer between the two clusters. C2 shows a significant lack of faint members (G > 17 mag), probably due to star evaporation and tidal stripping. Our results show that NGC 6649 may be the dominant cluster in a triple-cluster system (NGC 6649+C1+C2).