Adaptive K-means Clustering Research Articles

IC-IE-AKS-O: an automatic recognition method for coastal slope landslide areas

Automatically and accurately identifying the deformation zone of coastal slope landslides is crucial for exploring the mechanism of landslides and predicting landslide disasters. To this end, this study proposes an integrated automatic recognition method combining Image Clipping (IC), Image Information Enhancement (IE), Adaptive K-means Clustering Segmentation (AKS), and Optimization (O): IC-IE-AKS-O, which achieves precise extraction of the deformation area in coastal slope landslide images. Firstly, due to the more complex natural environment of field slopes, to extend the monitoring duration, we introduce a hierarchical operation algorithm based on the HSV color model, which effectively mitigates the impact of sunlight, rain, and foggy weather on image recognition accuracy. Secondly, this study proposes a 2D landslide image segmentation technique that combines K-means clustering with global threshold segmentation for landslide images, enabling the segmentation of small image regions with precision. Finally, we combine image information enhancement technology with image segmentation technology. To verify its effectiveness, we identify a landslide image of a coastal slope in Pingtan. The method displays an average relative error of 5.20% and 5.14% in the X and Y directions, respectively. Its advantages are threefold: (1) The combination of image information enhancement and segmentation techniques can more accurately identify landslide areas that appear blurred in the image; (2) expanding the temporal dimension of coastal slope monitoring; (3) providing excellent boundary conditions and segmentation results. The practical application of this method ensures the stable and accurate operation of the coastal slope monitoring system, providing a safeguard for the sustainable development of marine safety.

Integration of Probabilistic Multi-Class Labeling and Adaptive K-Means Clustering with KNN Classification: Application to Weather Data

Integration of Probabilistic Multi-Class Labeling and Adaptive K-Means Clustering with KNN Classification: Application to Weather Data

Online detection method for variable load conditions and anomalous sound of hydro turbines using correlation analysis and PCA-adaptive-K-means

The hydro turbine is the critical component of hydropower stations. Sound analysis provides a versatile non-contact approach for detecting anomalies or faults of hydro turbines. However, sound analysis methods such as Fourier transform and AI-based pattern recognition algorithms have limitations in achieving online detection due to potential fault frequency aliasing, non-obvious fault characteristics, lack of anomalous samples, and irregular occurrence of anomalous sounds. To overcome these challenges, the proposed method initially employs online correlation analysis to distinguish variable load conditions, followed by the detection of anomalous sounds under stable load conditions using principal component analysis (PCA) and adaptive K-means clustering (K-means). In the method, time–frequency analysis and wavelet packet decomposition are incorporated to extract signal features, with PCA employed to highlight anomalous features. Subsequently, feature correlation analysis and online adaptive clustering are used to accurately detect variable load conditions and anomalous sounds. The proposed method does not rely on prior knowledge of anomalous data, making it flexible and applicable to most practical hydro turbines. Additionally, the method aims to establish a robust foundation for fault diagnosis utilizing sound signals in hydro turbines, providing real-time detection of variable load conditions or anomalous sounds every second. To validate the effectiveness of the proposed method, detection experiments are conducted in practical hydropower station.

A Cluster-based Undersampling Technique for Multiclass Skewed Datasets

Imbalanced data classification is a demanding issue in data mining and machine learning. Models that learn with imbalanced input generate feeble performance in the minority class. Resampling methods can handle this issue and balance the skewed dataset. Cluster-based Undersampling (CUS) and Near-Miss (NM) techniques are widely used in imbalanced learning. However, these methods suffer from some serious flaws. CUS averts the impact of the distance factor on instances over the majority class. Near-miss method discards the inter-class data within the majority of class elements. To overcome these flaws, this study has come up with an undersampling technique called Adaptive K-means Clustering Undersampling (AKCUS). The proposed technique blends the distance factor and clustering over the majority class. The performance of the proposed method was analyzed with the aid of an experimental study. Three multiminority datasets with different imbalance ratios were selected and the models were created using K-Nearest Neighbor (kNN), Decision Tree (DT), and Random Forest (RF) classifiers. The experimental results show that AKCUS can attain better efficacy than the benchmark methods over multiminority datasets with high imbalance ratios.

Complete, fully-automatic extraction, classification and image registration of repeating printed fabric patterns and their derivatives

Repeating pattern recognition and extraction are important processes in the production and analysis of printed fabrics. In this study, the repeating patterns on printed fabric were extracted and innovatively analyzed using gray-scale image transform, two-dimensional discrete wavelet transform, fast Fourier transform and adaptive K-means clustering methods. The gray-scale image and two-dimensional discrete wavelet transforms were applied to compress the image. Image spectral characteristics were analyzed with a fast Fourier transform (FFT), the true fundamental frequency was extrapolated from its multiples, and frequency grouping was achieved using adaptive k-means clustering. The groups with the lowest frequencies in the image rows and columns were taken as giving the extraction frequency, and then the extraction frequency was mapped back onto the uncompressed fabric image to obtain the repeating pattern of the printed fabric. The similarity between patterns was used as a criterion for judging pattern design derivativeness. Similarity analysis was divided into two stages to reduce the calculation time and the amount of calculation required. In the first stage, repeating patterns were extracted and compared with existing patterns in an image database. In the second phase similarity analysis was carried out. The experimental results show the feasibility of the application of fast Fourier transforms to establishing a database of repeating patterns of printed fabrics. The average automatic repeated pattern extraction time is 10 s per image. The similarity analysis developed in the study has an accuracy rate of 98.0%, a sensitivity of 94.4% and a specificity of 98.5%. It is suitable for the analysis of all printing patterns, and meets the needs of the industry and printing pattern designers for data sharing and economical, competitive product management, over the total lifecycle of a fabric product from development to delivery, so satisfying market and societal needs.

Segmentation and classification of brain tumors from MRI images based on adaptive mechanisms and ELDP feature descriptor

Generally, soft tissue information of the brain is analyzed using Magnetic Resonance Imaging (MRI). An unusual growth of cells (tumor) in or around the brain may affect its overall functionality. Among brain tumors, gliomas that appear frequently in adults pose various challenges to radiologists. If they are not recognized properly in the initial phases, patients may face severe health issues including death. So, early and precise diagnosis of gliomas is significant for increasing the survival rate of patients. Recently, several computer-aided diagnosis (CAD) approaches have been developed for the detection and classification of gliomas. However, it is a challenging issue due to the noise as well as low sensitive boundary pixels appearing in gliomas. Considering this idea, we present a new approach based on adaptive methodologies and local texture descriptors. Here, primarily, we apply a median filter to remove noise from brain MRI images. Then, we employ adaptive decomposition strategies, namely bi-dimensional empirical mode decomposition (BEMD), and modified quasi-bivariate variational mode decomposition (MQBVMD) to attain meaningful sub-images. Later, we extract hidden texture details from the sub-images using an ELDP feature descriptor and then use a support vector machine (SVM) to distinguish brain MRI images as low-grade (LG) and high-grade (HG). Finally, we identify the infected region of gliomas with the help of adaptive K-means clustering and morphological operations. Using these sequences of operations, the proposed framework obtained 94.44% classification accuracy and 83.14% of dice similarity coefficient (DSC) value which is relatively high when compared to existing approaches.

Optimized Level Set Method for Segmentation of SAR Images using Adaptive Fuzzy K Means Clustering

The complications that occurred in remote sensing image information and analysis algorithms growth of a large scale image segmentation haven’t kept a place with the requirement for the methods which to develop the final accuracy of object detection as well as the recognition. Traditional Level set segmentation methods which are Chan-Vese (CV), Image and Vision Computing (IVC) 2010, ACM with SBGFRLS, and Online Region-Based ACM (ORACM) are suffered from more amounts of time complexity, as well as low segmentation accuracy due to large intensity homogeneities and the noise at which the region based segmentation is impossible. So this is the reason, we proposed a navel hybrid methodology called adaptive particle swarm optimization (PSO) based Fuzzy K-Means clustering algorithm. The proposed approach is diversified into two stages; in stage one, pre-processing the input image to improve the clustering efficiency and overcome the obstacles present in traditional methods by using particle swarm optimization (PSO) and Adaptive Fuzzy K-means clustering algorithm. With the help of the PSO algorithm, we getithe “optimum” pixels values are extracted from the input SAR images, these optimum values are automatically acted as clusters centers for Adaptive Fuzzy K-Means Clustering instead of random initialization from the original image. The pre-processing segmentation result improved the clustering efficiency but suffers from few drawbacks such as boundary leakages and outliers even particle Swarm optimization is used. To overcome the above drawbacks post-processing is needed to facilitate the superior segmentation results with the help of the level set method. It utilizes an efficient curve deformation driven by external and internal forces to capture the important structures (usual iedges) in an image. The combined approach of both pre-processing and post-processingwhich is called Particle Swarm Optimization based Adaptive Fuzzy-K-Means (AFKM) clustering via the level set method. The proposed approach is successfully implemented on large

Normalization-Based Validity Index of Adaptive K-Means Clustering for Multi-Solution Application

Validity evaluation aims to analyze the quality of the clustering algorithm with different measurement criteria. A variety of assessment methods have been introduced in the application of pattern recognition and computer vision. Although it is well known that mining information of massive data is essential, most of the validity indices only provide a single partitioning scheme for clustering validation. Moreover, the conventional evaluation algorithm is susceptible to the density and dimension of the dataset, which leads to assessment failure. In this paper, a normalization-based validity index (NbVI) is proposed for validity evaluation of the adaptive K-means clustering from a multi-solution perspective. According to the concept of high-compact within clusters and high-separation among groups, NbVI attempts to find the maximum relative ratio between normalized inter-distance and normalized intra-distance. The experimental results demonstrate that the proposed NbVI method exhibits excellent performance for the clustering of the density-unbalanced dataset for multi-solution applications. Moreover, the NbVI validation shows high versatility using different clustering algorithms.

An Automatic Head Surface Temperature Extraction Method for Top-View Thermal Image with Individual Broiler.

Surface temperature variation in a broiler’s head can be used as an indicator of its health status. Surface temperatures in the existing thermograph based animal health assessment studies were mostly obtained manually. 2185 thermal images, each of which had an individual broiler, were captured from 20 broilers. Where 15 broilers served as the experimental group, they were injected with 0.1mL of pasteurella inoculum. The rest, 5 broilers, served as the control group. An algorithm was developed to extract head surface temperature automatically from the top-view broiler thermal image. Adaptive K-means clustering and ellipse fitting were applied to locate the broiler’s head region. The maximum temperature inside the head region was extracted as the head surface temperature. The developed algorithm was tested in Matlab® (R2016a) and the testing results indicated that the head region in 92.77% of the broiler thermal images could be located correctly. The maximum error of the extracted head surface temperatures was not greater than 0.1 °C. Different trend features were observed in the smoothed head surface temperature time series of the broilers in experimental and control groups. Head surface temperature extracted by the presented algorithm lays a foundation for the development of an automatic system for febrile broiler identification.

Intelligent Estimation of Vitrinite Reflectance of Coal from Photomicrographs Based on Machine Learning

The accurate measurement of vitrinite reflectance (especially for mean maximum vitrinite reflectance, MMVR) is an important issue in the fields of coal mining and processing. However, the application of MMVR has been somewhat hampered by the subjective and the time-consuming characteristic of manual measurements. Semi-automated methods that are oversimplified might affect the accuracy in measuring MMVR values. To address these concerns, we propose a novel MMVR measurement strategy based on machine learning (MMVRML). Considering the complex nature of coal, adaptive K-means clustering is firstly employed to automatically detect the number of clusters (i.e., maceral groups) in photomicrographs. Furthermore, comprehensive features along with a support vector machine are utilized to intelligently identify the regions with vitrinite. The largest region with vitrinite in each photomicrograph is gridded for further regression analysis. Evaluations on 78 photomicrographs show that the model based on random forest and 15 simplified grayscale features achieves the state-of-the-art root mean square error of 0.0424. In addition, to facilitate the usage of petrologists without strong expertise in the machine learning domain, we released the first non-commercial standalone software for estimating MMVR.

Adaptive attenuation correction during H-scan ultrasound imaging using K-means clustering

H-scan ultrasound (US) imaging (where the ‘H’ stands for Hermite) is a novel non-invasive, low cost and real-time technology. Like traditional US, H-scan US suffers from frequency-dependent attenuation that must be corrected to have acceptable image quality for tissue characterization. The goal of this research was to develop a novel attenuation correction method based on adaptive K-means clustering. To properly isolate these signals, a lateral moving window approach applied to adaptively adjust GH filters based on the changing of RF vector spectrums. Then the signal isolated via the same filter will be combined together via overlap-add technology to keep the information loss minimum. Experimental data was collected using a Verasonics 256 US scanner equipped with a L11-4v linear array transducer. In vivo data indicates that H-scan US imaging after adaptive attenuation correction can optimally re-scale the GH kernels and match to the changing spectrum undergoing attenuation (i.e. high frequency shift). This approach produces H-scan US images with more uniform spatial intensity and outperforms global attenuation correction strategies. Overall, this approach will improve the ability of H-scan US imaging to estimate acoustic scatterer size and will improve its clinical use for tissue characterization when imaging complex tissues.

Identification of plasmodium falciparum and plasmodium vivax on digital image of thin blood films gf

<span>Observing presence of <em>Plasmodium</em> parasite of stained thick or thin blood films through microscopic examination is a gold standard for malaria diagnosis. Although the microscopic examination has been extensively used, misidentification might occur caused by human factors. In order to overcome misidentification problem, several studies have been conducted to develop a computer-aided malaria diagnosis (CADx) to assist paramedics in decision-making. This study proposes an approach to identify species and stage of <em>Plasmodium falciparum</em> and <em>Plasmodium vivax</em> on thin blood films collected from the Laboratory of Parasitology, Faculty of Medicine, Universitas Gadjah Mada. Adaptive k-means clustering is applied to segment <em>Plasmodium </em>parasites. A total of 39 features consisting of shape and texture features are extracted and then selected by using wrapper-based forward and backward directions. Classification is evaluated in two schemes. The first scheme is to classify the species of parasite into two classes. The second scheme is to classify the species and stage of parasite into six classes. Three classifiers applied are k-nearest neighbour (KNN), support vector machine (SVM) and multi-layer perceptron (MLP). Furthermore, to facilitate the multiclass classification, one-versus-one (OVO) and one-versus-all (OVA) methods are implemented. The first scheme achieves the accuracy of 88.70% based on MLP classifier using three selected features. While the accuracy gained by the second scheme is 95.16% based on OVO and MLP classifier using 29 selected features. These results indicate that the proposed approach successfully identifies the species and stage of parasite on thin blood films and has potential to be implemented in the CADx system for assisting paramedics in diagnosing malaria.</span>

Automatic Counting for Live and Dead Cells from Trypan Blue-Stained Images by Image Analysis Based on Adaptive K-Means Clustering

Computer-assisted image analysis can be employed to reduce the time consumed in the routine task such as cell counting. This study aimed to establish a method to perform this routine task based on an image analysis to automatically count live and dead cells after staining with trypan blue dye. Gray scale conversion and morphological operation were applied to the input images to enhance the image quality before image segmentation, then adaptive k-means clustering was applied to classify the groups of live and dead cells. Circular Hough transform and object labelling were carried out to identify the number of each cell type. The counting results from the proposed method were compared with the counting of three experts and the ImageJ software. The results showed that the proposed method had very high correlation with the results of the three experts in counting live cells (R2>0.95) and was better than the counting results achieved by ImageJ. The number of dead cells counted by our program was in good agreement with the experts’ counting (R2>0.64). In conclusion, this study suggests that using new image analysis program can be confidently substituted for a manual counting in routine cell counting

Roadside vegetation segmentation with Adaptive Texton Clustering Model

Automatic roadside vegetation segmentation is important for various real-world applications and one main challenge is to design algorithms that are capable of representing discriminative characteristics of vegetation while maintaining robustness against environmental effects. This paper presents an Adaptive Texton Clustering Model (ATCM) that combines pixel-level supervised prediction and cluster-level unsupervised texton occurrence frequencies into superpixel-level majority voting for adaptive roadside vegetation segmentation. The ATCM learns generic characteristics of vegetation from training data using class-specific neural networks with color and texture features, and adaptively incorporates local properties of vegetation in every test image using texton based adaptive K-means clustering. The adaptive clustering groups test pixels into local clusters, accumulates texton frequencies in every cluster and calculates cluster-level class probabilities. The pixel- and cluster-level probabilities are integrated via superpixel-level voting to determine the category of every superpixel. We evaluate the ATCM on three real-world datasets, including the Queensland Department of Transport and Main Roads, the Croatia, and the Stanford background datasets, showing very competitive performance to state-of-the-art approaches.

K-Means Cloning: Adaptive Spherical K-Means Clustering

We propose a novel method for adaptive K-means clustering. The proposed method overcomes the problems of the traditional K-means algorithm. Specifically, the proposed method does not require prior knowledge of the number of clusters. Additionally, the initial identification of the cluster elements has no negative impact on the final generated clusters. Inspired by cell cloning in microorganism cultures, each added data sample causes the existing cluster ‘colonies’ to evaluate, with the other clusters, various merging or splitting actions in order for reaching the optimum cluster set. The proposed algorithm is adequate for clustering data in isolated or overlapped compact spherical clusters. Experimental results support the effectiveness of this clustering algorithm.

Harmonics extraction based speech recovery for underdetermined mixing systems

It is an intractable task to achieve high-efficiency and high-quality speech recovery for the existing underdetermined systems. To solve this problem, this paper proposes a harmonics extraction based underdetermined speech recovery algorithm, which consists of 4 stages. In the 1st stage, spectrum correction technique is adopted to extract the harmonic components from the mixtures’ short time Fourier transform (STFT); In the 2nd stage, a phase-coherence criterion is applied on these harmonic components to identify the single source components; In the 3rd stage, these single source patterns are further categorized into multiple groups by means of the adaptive k-means clustering, from which the mixing matrix is estimated; In the last stage, this estimated matrix is further combined with the subspace projection algorithm, which resultantly yields the final source recovery. The high efficiency lies in that the harmonics extraction is properly combined with single source component identification. The speech recovery experiment demonstrated that, compared to the original subspace projection algorithm, the proposed method can acquire a higher recovery quality, which presents a potential application in other harmonic related fields.

Eavesdrop with PoKeMon: Position free keystroke monitoring using acoustic data

Keyboard input is an important way to enter massive personal information into computing systems and the Internet in our social life. The large amount of acoustic data from the keystroke tapping makes it possible for eavesdroppers to snoop the sensitive and private input remotely. Keystroke monitoring using the acoustic signals, which extracts the user information directly from the keystroke acoustics, is a critical security and privacy issue and has recently attracted much research attention. The existing works use smartphones to collect and infer the keystrokes with various data analysis algorithms. However, most of them rely on the assumption that the phone-keyboard relative position is known in advance, which is often unrealistic for real-world scenarios such as adversary eavesdropping. To address this problem and make keystroke monitoring free to use, in this paper, we propose a Po sition free Ke ystroke Mon itoring scheme using a single smartphone. By exploiting the reliable linguistic accountings and the layout of a certain keyboard, we develop a geometrical approach to estimate the relative positions and angles of the smartphone to the keyboard. Our proposed scheme can self-calibrate the position estimation with the continuous acoustic signals. After that, effective keystroke snooping is established based on the Mel-Frequency Cepstral Coefficient and adaptive k-means clustering. We conduct experiments with Samsung S4 smartphones and simulations. The results show that the positioning scheme achieves a high accuracy by 93%, and the keystroke snooping accuracy is improved by 32.6% compared to the existing works.

Beat-to-beat P-wave morphology as a predictor of paroxysmal atrial fibrillation

Background and objectivesAtrial Fibrillation (AF) is the most common cardiac arrhythmia. The initiation and the perpetuation of AF is linked with phenomena of atrial remodeling, referring to the modification of the electrical and structural characteristics of the atrium. P-wave morphology analysis can reveal information regarding the propagation of the electrical activity on the atrial substrate. The purpose of this study is to investigate patterns on the P-wave morphology that may occur in patients with Paroxysmal AF (PAF) and which can be the basis for distinguishing between PAF and healthy subjects. MethodsVectorcardiographic signals in the three orthogonal axes (X, Y and Z), of 3–5 min duration, were analyzed during SR. In total 29 PAF patients and 34 healthy volunteers were included in the analysis. These data were divided into two distinct datasets, one for the training and one for the testing of the proposed approach. The method is based on the identification of the dominant and the secondary P-wave morphology by combining adaptive k-means clustering of morphologies and a beat-to-beat cross correlation technique. The P-waves of the dominant morphology were further analyzed using wavelet transform whereas time domain characteristics were also extracted. Following a feature selection step, a SVM classifier was trained, for the discrimination of the PAF patients from the healthy subjects, while its accuracy was tested using the independent testing dataset. ResultsIn the cohort study, in both groups, the majority of the P-waves matched a main and a secondary morphology, while other morphologies were also present. The percentage of P-waves which simultaneously matched the main morphology in all three leads was lower in PAF patients (90.4 ± 7.8%) than in healthy subjects (95.5 ± 3.4%, p= 0.019). Three optimal scale bands were found and wavelet parameters were extracted which presented statistically significant differences between the two groups. Classification between the two groups was based on a feature selection process which highlighted 7 features, while an SVM classifier resulted a balanced accuracy equal to 93.75%. The results show the virtue of beat-to-beat analysis for PAF prediction. ConclusionThe difference in the percentage of the main P-wave-morphology and in the P-wave time-frequency characteristics suggests a higher electrical instability of the atrial substrate in patients with PAF and different conduction patterns in the atria.

Spatiotemporal Video Denoising Based on Adaptive Thresholding and Clustering

In this paper we propose a novel video denoising method based on adaptive thresholding and K-means clustering. In the proposed method the adaptive thresholding is applied rather than the conventional hard-thresholding of the VBM3D method. The adaptive thresholding has a high ability to adapt and change according to the amount of noise. More specifically, hard-thresholding is applied on the higher noise areas while soft-thresholding is applied on the lower noise areas. Consequently, we can successfully remove the noise effectively and at the same time preserve the edges of the image, because the clustering approach saves more computation time and is more capable of finding relevant patches than the block-matching approach. So, the K-means clustering method in the final estimate in this paper is adopted instead of the block-matching method in the VBM3D method in order to restrict the search of the candidate patches within the region of the reference patch and therefore improve the grouping. Experimental results emphasize the superiority of the new method over the reference methods in terms of visual quality, Peak Signal-to-Noise Ratio (PSNR), and Image Enhancement Factor (IEF). Execution time of the proposed algorithm consumes less time in denoising than that in the VBM3D algorithm.

Adaptive Image Enhancement Using Image Properties and Clustering

Image Enhancement is the method of improving the visibility of given image. Image Properties are used for the analysis of Quality of the given image. Various image Properties considered to improve the quality of the image. The Classification or grouping of images can be made by applying unsupervised image Classification algorithm. In our proposed method, various image properties are studied and an Adaptive K-means Clustering method is applied for Fractal image with Entropy Properties. The images are to enhanced on the basis of its grouping automatically. The resulted Classification can be acceptable by the user since the grouping is made on the type of the image i.e., Good Visible, Moderate and Blur images.