Narrow Results

Mechanization, specialization, and shrewd steering have been regularly added to indoor fixtures commercial enterprises due to their ever-growing clinical and technological capacities. Competing in a marketplace in which technical layout competencies are required, the capability to think creatively and create novel shapes is what makes a product stand out and provides cost. Customers are no longer content material to buy fixtures in bulk, preferring as an alternative to choose pieces that are extra suitable to their own necessities and people in their homes, to higher fulfill primary demands and cope with problems like cramped quarters. The user’s needs for domestic product ergonomics inform the established order of a standards layer, which is then weighted to get its corresponding cost. In situations wherein competing gadgets provide similar features and overall performance, the aesthetics of the product grow to be more critical. This research explores the Pareto-primarily based Genetic Algorithm (PGA) inside this framework, with the goal of fostering innovation in product layout and the generation of novel ideas. Using more suitable $K$-means clustering and the Enhanced Least Squares Support Vector Machine (ELSSVM), this study gives a prediction technique for objectively assessing furniture comfort. Designers might be capable of focusing on aesthetics by using the findings as a springboard to research new shapes, way to the fact that the proposed version investigates several alternatives in pursuit of nice purposeful forms. Afterward, a mathematical version is used to symbolize the layout preference problems, and the layout scheme selection method is simulated. In the give-up, the advised model’s viability is checked by means of looking at how fixtures from the product’s layout picks had been carried out to assist designers give us fresh thoughts.

This paper studies the color expression method of fine arts painting based on multi-scale Retinex, so as to improve the color expression ability of fine arts painting works in digital application. The images of art paintings are converted into HSI and LAB color spaces, respectively. The images are binarized using the Otsu threshold method in the HSI color space, and segmented using the $K$-means clustering algorithm in the LAB color space. The image processing results from the two color spaces are mathematically combined (or calculated) to complete the segmentation of the art painting image. The multi-scale Retinex algorithm is used to map the tone of the segmented image. Based on the trichromatic theory, the color correction is carried out on the image after tone mapping, and the final color expression result of art painting is obtained. The experimental results show that the image processed by this method has high definition, rich color and detail information, and strong overall layering, which makes the color expression of art paintings more ideal.

Daily ${}^7\mathrm{\text{Be}}$ activity concentrations data across 21 global locations between 2010 and 2017 from the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) were analyzed using multifractal formalism. The multifractal detrended fluctuation analysis revealed that ${}^7\mathrm{\text{Be}}$ distribution across the 21 locations are multifractal ($0.17<\alpha <0.66$) with a wide range of fractal exponents. The observed multifractality was found to be statistically significant except at two locations (RN45 and RN47). The multifractal strength ($\alpha$) and Holder’s exponent (${\alpha }_0$) were used to group the locations into 3 clusters with K-means algorithm. The relationship between ${}^7\mathrm{\text{Be}}$ and five drivers (Southern Oscillation Index — SOI, North Atlantic Oscillation — NAO, Total Sunspot number — Tot_SN, Northern hemisphere Sunspot number — NH_SN and Southern hemisphere sunspot number — SH_SN) was investigated using multifractal detrended cross-correlation analysis. The multifractal cross-correlation between ${}^7\mathrm{\text{Be}}$ and drivers was found to be 0.06–0.21(SOI), 0.08–0.23 (NAO), 0.04–0.27 (Tot_SN), 0.05–0.25 (NH_SN) and 0.04–0.27 (SH_SN). NAO was found to be the strongest driver of ${}^7\mathrm{\text{Be}}$. The location, RN16 in Yellowknife Canada, showed strong cross-correlation with the five drivers.

In this paper, we present efficient geometric algorithms for the discrete constrained 1-D K-means clustering problem and extend our solutions to the continuous version of the problem. One key clustering constraint we consider is that the maximum difference in each cluster cannot be larger than a given threshold. These constrained 1-D K-means clustering problems appear in various applications, especially in intensity-modulated radiation therapy (IMRT). Our algorithms improve the efficiency and accuracy of the heuristic approaches used in clinical IMRT treatment planning.

In this paper, a fully complex-valued radial basis function (FC-RBF) network with a fully complex-valued activation function has been proposed, and its complex-valued gradient descent learning algorithm has been developed. The fully complex activation function, sech(.) of the proposed network, satisfies all the properties needed for a complex-valued activation function and has Gaussian-like characteristics. It maps Cⁿ → C, unlike the existing activation functions of complex-valued RBF network that maps Cⁿ → R. Since the performance of the complex-RBF network depends on the number of neurons and initialization of network parameters, we propose a K-means clustering based neuron selection and center initialization scheme. First, we present a study on convergence using complex XOR problem. Next, we present a synthetic function approximation problem and the two-spiral classification problem. Finally, we present the results for two practical applications, viz., a non-minimum phase equalization and an adaptive beam-forming problem. The performance of the network was compared with other well-known complex-valued RBF networks available in literature, viz., split-complex CRBF, CMRAN and the CELM. The results indicate that the proposed fully complex-valued network has better convergence, approximation and classification ability.

Task-related reorganization of functional connectivity (FC) has been widely investigated. Under classic static FC analysis, brain networks under task and rest have been demonstrated a general similarity. However, brain activity and cognitive process are believed to be dynamic and adaptive. Since static FC inherently ignores the distinct temporal patterns between rest and task, dynamic FC may be more a suitable technique to characterize the brain’s dynamic and adaptive activities. In this study, we adopted $k$-means clustering to investigate task-related spatiotemporal reorganization of dynamic brain networks and hypothesized that dynamic FC would be able to reveal the link between resting-state and task-state brain organization, including broadly similar spatial patterns but distinct temporal patterns. In order to test this hypothesis, this study examined the dynamic FC in default-mode network (DMN) and motor-related network (MN) using Blood-Oxygenation-Level-Dependent (BOLD)-fMRI data from 26 healthy subjects during rest (REST) and a hand closing-and-opening (HCO) task. Two principal FC states in REST and one principal FC state in HCO were identified. The first principal FC state in REST was found similar to that in HCO, which appeared to represent intrinsic network architecture and validated the broadly similar spatial patterns between REST and HCO. However, the second FC principal state in REST with much shorter “dwell time” implied the transient functional relationship between DMN and MN during REST. In addition, a more frequent shifting between two principal FC states indicated that brain network dynamically maintained a “default mode” in the motor system during REST, whereas the presence of a single principal FC state and reduced FC variability implied a more temporally stable connectivity during HCO, validating the distinct temporal patterns between REST and HCO. Our results further demonstrated that dynamic FC analysis could offer unique insights in understanding how the brain reorganizes itself during rest and task states, and the ways in which the brain adaptively responds to the cognitive requirements of tasks.

For computing the $k$-means clustering of the streaming and distributed big sparse data, we present an algorithm to obtain the sparse coreset for the $k$-means in polynomial time. This algorithm is mainly based on the explicit form of the center of mass and the approximate $k$-means. Because of the existence of the approximation, the coreset of the output inevitably has a factor, which can be controlled to be a very small constant.

Coarse-graining of complex networks is one of the important algorithms to study large-scale networks, which is committed to reducing the size of networks while preserving some topological information or dynamic properties of the original networks. Spectral coarse-graining (SCG) is one of the typical coarse-graining algorithms, which can keep the synchronization ability of the original network well. However, the calculation of SCG is large, which limits its real-world applications. And it is difficult to accurately control the scale of the coarse-grained network. In this paper, a new SCG algorithm based on K-means clustering (KCSCG) is proposed, which cannot only reduce the amount of calculation, but also accurately control the size of coarse-grained network. At the same time, KCSCG algorithm has better effect in keeping the network synchronization ability than SCG algorithm. A large number of numerical simulations and Kuramoto-model example on several typical networks verify the feasibility and effectiveness of the proposed algorithm.

In order to deeply analyze the quantitative relationship between traffic flow state and crash risk, a highway traffic safety state classification method based on multi-parameter fusion clustering was proposed. First, attribute data of highway traffic crashes and corresponding upstream and downstream traffic flow data were extracted, and matched with paired case-control method. Secondly, considering the different roles of traffic volume, speed and occupancy in traffic state classification, the weight optimization algorithm is introduced to calculate the weight of the three parameters. Therefore, the comprehensive evaluation index of traffic state with the fusion of three parameters is obtained and used as the input index of traffic safety state clustering. Finally, $k$-means clustering method is used to classify the highway traffic safety status. The result of the case study shows that the proposed method can achieve reasonable and effective traffic safety state division. The classification results are helpful to quantitatively evaluate highway crash risk levels under different traffic safety states.

In this paper, we propose a new face detection and tracking algorithm for real-life telecommunication applications, such as video conferencing, cellular phone and PDA. We combine template-based face detection and tracking method with color information to track a face regardless of various lighting conditions and complex backgrounds as well as the race. Based on our experiments, we generate robust face templates from wavelet-transformed lowpass and two highpass subimages at the second level low-resolution. However, since template matching is generally sensitive to the change of illumination conditions, we propose a new type of preprocessing method. Tracking method is applied to reduce the computation time and predict precise face candidate region even though the movement is not uniform. Facial components are also detected using k-means clustering and their geometrical properties. Finally, from the relative distance of two eyes, we verify the real face and estimate the size of facial ellipse. To validate face detection and tracking performance of our algorithm, we test our method using six different video categories of QCIF size which are recorded in dynamic environments.

A robust and efficient approach to estimate the fundamental matrix is proposed. The main goal is to reduce the computational cost involved in the estimation when robust schemas are applied. The backbone of the proposed technique is the conventional Least Median of Squares (LMedS) technique. It is well known that the LMedS is one of the most robust regressors for highly contaminated data and unstable models. Unfortunately, its computational complexity renders it useless for practical applications. To overcome this problem, a small number of low-dimensionality least-square problems are solved using well-selected subsets from the input data. The results of this initial approach are fed into the LMedS schema, which is applied to recover the final estimation of the Fundamental matrix. The complexity is substantially reduced by applying a selection process based on an effective statistical analysis of the inherent correlation of the input data. This analysis is used to define a suitable clustering of the data and to drive the subset selection aiming at the reduction of the search space in the LMedS schema. It is shown that avoiding redundancies better estimates can be obtained while keeping the computational cost low. Selected results of computer experiments were conducted to assess the performance of the proposed technique.

Lung tumor detection using computer-aided modeling improves the accuracy of detection and clinical recommendation precision. An optimal tumor detection requires noise reduced computed tomography (CT) images for pixel classification. In this paper, the butterfly optimization algorithm-based $K$-means clustering (BOAKMC) method is introduced for reducing CT image segmentation uncertainty. The introduced method detects the overlapping features for optimal edge classification. The best-fit features are first trained and verified for their similarity. The clustering process recurrently groups the feature matched pixels into clusters and updates the centroid based on further classifications. In this classification process, the uncertain pixels are identified and mitigated in the tumor detection analysis. The best-fit features are used to train local search instances in the BOA process, which influences the similar pixel grouping in the uncertainty detection process. The proposed BOAKMC improves accuracy and precision by 10.2% and 13.39% and reduces classification failure and time by 11.29% and 11.52%, respectively.

Background: The concept of tuberculosis diagnosis plays a significant role in the current world since, in accordance with the Global Tuberculosis (TB) Report in 2019, more than one million cases are reported per year in India. Various tests are available even then the chest X-ray is the most significant one, devoid of which the diagnosis will be incomplete. By the usage of computationally designed algorithms, various clinical, as well as diagnostic functions, were built in ancient poster anterior chest radiographs. The Digital image (X-ray) may be an essential medium for examining and annotating patient’s demographics coverage in the screening of TB via chest radiography. Results: Even though several medicines are available to cure TB, diagnosis with accuracy is a major challenge. So, we have introduced a fastened technique with the merged combination of Adaptive Boosting (AdaBoost) and learning vector quantization (LVQ) for determining TB in an easier way with the input chest X-ray image of a person with the aid of computer-aided diagnosis with greatest accuracy, precision, recall and F1 values. This finest technique got an accuracy of 94.73% when compared to the prior conventional methods used such as SVM and Convolutional Neural Network. Conclusions: Tuberculosis detection can be done in a meaningful way with the aid of MATLAB simulation using Computer Aided Diagnosis. The algorithms Adaboost and LVQ works best with the datasets for around 400 chest X-ray images for detecting the normal and abnormal images conditions for the detection of the disease for a patient suspected to have TB, in a fraction of seconds.

In data cleaning, the process of detecting and correcting corrupt, inaccurate or irrelevant records from the record set is a tedious task. Particularly, the process of “outlier detection” occupies a significant role in data cleaning that removes or eliminates the outlier’s that exist in data. Traditionally, more efforts have been taken to remove the outliers, and one of the promising ways is customizing clustering models. In this manner, this paper intends to propose a new outlier detection model via enhanced k-means with outlier removal (E-KMOR), which assigns all outliers into a group naturally during the clustering process. For assigning the point to be outliers, a new intra-cluster based distance evaluation is employed. The main contribution of this paper is to select cluster centroid optimally through a newly proposed hybrid optimization algorithm termed particle updated lion algorithm (PU-LA), which hybrids the concepts of LA and particle swarm optimization (PSO), respectively. Thereby, the proposed work is named as E-KMOR-PU-LA. Finally, the efficacy of the proposed E-KMOR-PU-LA model is proved through a comparative analysis over conventional models by concerning runtime and accuracy.

Social distance monitoring is of great significance for public health in the era of COVID-19 pandemic. However, existing monitoring methods cannot effectively detect social distance in terms of efficiency, accuracy, and robustness. In this paper, we proposed a social distance monitoring method based on an improved YOLOv4 algorithm. Specifically, our method constructs and pre-processes a dataset. Afterwards, our method screens the valid samples and improves the K-means clustering algorithm based on the IoU distance. Then, our method detects the target pedestrians using a trained improved YOLOv4 algorithm and gets the pedestrian target detection frame location information. Finally, our method defines the observation depth parameters, generates the 3D feature space, and clusters the offending aggregation groups based on the L2 parametric distance to finally realize the pedestrian social distance monitoring of 2D video. Experiments show that the proposed social distance monitoring method based on improved YOLOv4 can accurately detect pedestrian target locations in video images, where the pre-processing operation and improved K-means algorithm can improve the pedestrian target detection accuracy. Our method can cluster the offending groups without going through calibration mapping transformation to realize the pedestrian social distance monitoring of 2D videos.

Impulse noise is an image noise that degrades the quality of the image drastically. In this paper, k-means clustering has been incorporated with fuzzy-support vector machine (FSVM) classifier for classification of noisy and non-noisy pixels in removal of impulse noise from gray images. Here, local binary pattern (LBP) has been incorporated with previously used feature vector prediction error of the processing pixel, absolute difference between median value and processing pixel, median pixel, pixel under operation and mean value around the processing kernel. In this work, $k$-means clustering has been used for reducing the feature vector set, where features have been extracted from the images corrupted with 10%, 50%, and 90% impulse noise. If the pixel is depicted as noisy in testing phase, histogram adaptive fuzzy filter is processed over the noisy pixel under operation. It is seen that the proposed filter offers improved performance over some of the state-of-the-art filter in terms of different image quality measures likely PSNR, SSIM, MSE, FSIM, etc. It is observed that performance is increased by $\sim$2–5dB than baseline filters likely SVM fuzzy filter, and artificial neural network based adaptive sized mean filter (ANNASMF) especially at high density noise.

Parallel discovery of inherent clusters using massively threaded architectures is the solution for handling computational challenges raised by fat datasets in cluster analysis. The Graphics Processing Unit and Compute Unified Device Architecture form a convincing platform to parallelize clustering algorithms. The parallel K-means algorithm aims at increasing the speedup, but often faces the hitch of falling into local minima. The heuristic search procedure to discover the global optima in the solution space is known as Tabu Search. The K-means clustering solution is fine-tuned by applying parallel implementation of Tabu Search K-means clustering in order to increase efficacy. The aim is to combine optimization characteristic of Tabu Search for calculation of centroids with clustering attitude of K-means and to enhance the solution using processing power of GPU. The parallelization strategy used exhibits increase in speedup. The parallel Tabu-KM algorithm is tested on standard datasets, and performance is compared with sequential K-means, parallel K-means and sequential Tabu K-means algorithms. The experimental results confirm yet another parallelization technique to unravel data clustering problems.

Rectilinear Steiner Tree (RST) construction is a fundamental problem in very large scale integration (VLSI) physical design. Its applications include placement and routing in VLSI physical design automation (PDA) where wire length and timing estimations for signal nets are obtained. In this paper, a pseudo-Boolean satisfiability (PB-SAT)-based approach is presented to solve rectilinear Steiner tree problem. But large nets are a bottleneck for any SAT-based approach. Hence, to deal with large nets, a region-partitioning-based algorithm is taken into consideration, which eventually achieves a reasonable running time. Furthermore, a clustering-based approach is also explored to improve the partitioning of nets by identifying clusters and then applying a heuristic-based approach to get the minimum wire length for each set of the clusters. Experimental results obtained by these techniques show that the proposed algorithm can solve the RST problem very effectively even on large circuits and it outperforms the widely used RST algorithm FLUTE with 3$\times$to 9$\times$speedups.

Hardware Trojans (HT) are tiny, malicious circuits intentionally designed by an adversary. The existing works found in the literature on gate-level netlists are mainly based on supervised classification with few attempts at unsupervised clustering. However, the over-reliance on pre-defined structural features used in these supervised classification methods makes them vulnerable to the new Trojan attacks, whereas most unsupervised methods ignore this feature completely. This work presents an unsupervised approach for HT net detection based on the structural features required for small rare-event triggered HTs irrespective of the payload. The proposed work uses k-means clustering on these features to reduce the search space. A new metric based on combinational controllability is applied next to detect the possible trigger net. Experimental results of fifteen selected Trust-HUB benchmarks show the capability of the proposed technique against different types of HT triggers. Results show that the proposed approach reduces the search space massively (up to 99%) while running within a reasonable time frame.

In this report, a clustering approach is presented to detect dynamical nonlinearity in a stationary time series. The one-dimensional time series sampled from a dynamical system is mapped on to the m-dimensional phase space by the method of delays. The vectors in the phase space are partitioned by k-means clustering technique. The local trajectory matrix for the vectors in each of the clusters is determined. The eigenvalues of the local trajectory matrices represent the variation along the principal directions and are obtained by singular value decomposition (SVD). The product of the eigenvalues represents the generalized variance and is a measure of the local dispersion in the phase space. The sum of the local dispersions (g^T) is used as the discriminant statistic to classify data sets obtained from deterministic and stochastic settings. The surrogates were generated by the Iterated Amplitude Adjusted Fourier Transform (IAAFT) technique.