High quality voice conversion based on ISODATA clustering algorithm

Automated demarcation of the homogeneous domains of trace distribution within a rock mass based on GLCM and ISODATA

The extended-range electric vehicle (E-REV) can solve the problems of short driving range and long charging time of pure electric vehicles, but it is necessary to control the engine working points and allocate the power of the energy sources reasonably. In order to improve the fuel economy of the vehicle, an energy management strategy (EMS) that can adapt to the daily driving characteristics of the driver and adjust the control parameters online is proposed in this paper. Firstly, through principal component analysis (PCA) and iterative self-organizing data analysis techniques algorithm (ISODATA) of historical driving data, a typical driving cycle which can describe driving characteristics of the driver is constructed. Then offline optimization of control parameters by adaptive simulated annealing under each typical driving cycle and online recognition of driving cycles by extreme learning machine (ELM) are applied to the adaptive multi-workpoints energy management strategy (A-MEMS) of E-REV. In the end, compared with traditional rule-based control strategies, A-MEMS achieves good fuel-saving and emission-reduction result by simulation verification, and it explores a new and feasible solution for the continuous upgrade of the EMS.

Ensemble fuzzy radial basis function neural networks architecture driven with the aid of multi-optimization through clustering techniques and polynomial-based learning

Aiming at the counterintuitive phenomena of the Dempster–Shafer method in combining the highly conflictive evidences, a combination method of evidences based on the clustering analysis is proposed in this paper. At first, the cause of conflicts is disclosed from the point of view of the internal and external contradiction. And then, a new similarity measure based on it is proposed by comprehensively considering the Pignistic distance and the sequence according to the size of the basic belief assignments over focal elements. This measure is used to calculate the commonality function of evidences to amend the evidence sources; Meanwhile, the Iterative Self-organizing Data Analysis Techniques Algorithm (ISODATA) method based on the new measure is used for clustering according to the clustering characters of the original evidences. The Dempster rule is applied to combining all the evidences in each clustering into an evidential representative, and the reliability is calculated based on the commonality and the occurrence frequency of the evidences in the clustering. At last, Murphy's method is used to combine these evidential representatives of the different clusterings. The experimental results through a series of numeric examples show that the method proposed in this paper is more effective and superior to others.

Federated learning (FL) offers a promising paradigm that empowers numerous Internet of Things (IoT) devices to implement distributed learning on the premise of ensuring user privacy and data security. However, since FL adopts a synchronous distributed training mode, the heterogeneity of participating IoT devices and limited communication resources make FL encounter serious issues of low training efficiency in actual deployment. In this article, we propose an excellent FL policy for the heterogeneous IoT-edge FL system to improve distributed training efficiency. Specifically, first, by borrowing the idea of clustering, we explore an iterative self-organizing data analysis techniques algorithm (ISODATA)-based heterogeneous-aware client scheduling strategy to alleviate the issue of low training efficiency incurred by the heterogeneity of clients. Subsequently, to tackle the challenge of limited communication resources in FL, we first analyze the characteristics of the optimal resource block allocation solution theoretically and then introduce a mixed-integer linear programming (MILP)-based strategy to judiciously allocate resource blocks for scheduled clients. Comprehensive experimental results demonstrate that, compared with benchmarking strategies, our proposed FL policy can achieve up to 55.22% accuracy improvement in a relaxed time scenario, and attain up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3.62\times $ </tex-math></inline-formula> acceleration for reaching the specific expected accuracy.

PDF HTML阅读 XML下载 导出引用 引用提醒 中国人均灰水生态足迹变化驱动效应测度及时空分异 DOI: 10.5846/stxb201706221136 作者: 作者单位: 辽宁师范大学城市与环境学院,辽宁师范大学 作者简介: 通讯作者: 中图分类号: 基金项目: 国家社会科学基金重点项目（16AJY009） Driving effect measurements and spatial-temporal variation of the per capita gray water ecological footprint in China Author: Affiliation: College of Urban and Environment,Liaoning Normal University,College of Urban and Environment,Liaoning Normal University Fund Project: Key Projects of National Social Sci-ence Fund, No. 16AJY009. 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:将传统灰水足迹和水生态足迹方法相结合，运用扩展的Kaya恒等式和LMDI指数分解方法对中国各省市的人均灰水生态足迹变化的驱动因素进行测度分析，充分考虑了资本和劳动力因素，选取经济活度效应，资本深化效应，资本效率效应，足迹强度效应，环境效率效应5个效应对人均灰水生态足迹变化的影响，结合ISODATA聚类模型对各效应进行空间聚类，从而分析各效应的空间特征。结果显示：中国人均灰水生态足迹产出变化是这5种因素共同作用的结果，资本深化效应和经济活度效应具有增量效应特点，而环境效率效应、足迹强度效应、资本效率效应呈减量效应特点；在各驱动效应的的强弱对比中，资本深化效应和足迹强度效应的特征较为明显。经济发展带动了科技进步也使得用水效率不断提高是足迹强度效应呈减量效应的主要原因；而工业化阶段经济向资本密集型转变是资本效率下降的主要原因。研究对中国灰水生态变化与资本要素之间的关系进行了探讨，对环保政策的调整及水资源可持续利用研究具有一定的参考价值。 Abstract:In this study, the authors apply water ecological footprint measurement methods to the field of gray water footprint research and provide a method for calculating gray water ecological footprints. Based on the extended Kaya identity and the Logarithmic Mean Divisia Index (LMDI) technique decomposition methods, we calculated the gray water ecological footprint of 31 provinces and the per capita gray water ecological footprint in China from 2000 to 2014. We then measured and decomposed the values of the driving effects of per capita gray water ecological footprint change. From these results, we selected five of the more important economic and environmental factors to apply as capital factors to gray water research. This method included measurements of the effects of the working population, capital stocks, capital output coefficients, water ecological footprint intensity, and gray water ecological footprint emission coefficients. The five driving effects were economic activity, capital deepening, capital efficiency, footprint intensity, and environmental efficiency. The values of the driving effects on changes during the time period are discussed and analyzed based on the Iterative Self-Organizing Data Analysis Techniques Algorithm (ISODATA) clustering model for spatial clustering of the effects. The results showed that the Chinese per capita gray water ecological footprint output changed because of the interaction of these five factors. Among them, the incremental capital deepening effect was obvious, as a large amount of capital has been invested to promote rapid development of the regional economy and to employ a rapidly growing population. Economic activity can also promote the characteristics and the size of regional economies as population growth will increase regional demands for water. The other capital deepening effect is a decline in the output efficiency of capital. With the rapid development of its economy, China is in a period of industrialization and urbanization. The economic structure has gradually changed from a population-intensive to a capital-intensive structure. Industry, especially heavy industry, had a higher initial investment and a lower output efficiency, which is why the effect of capital efficiency continues to decrease. The most obvious effect on footprint intensity was the effect of per capita gray water ecological footprint reduction. With the improvement of economic activity and advances in science and technology, the water ecological footprint intensity has been greatly reduced. As a result of the effective control of pollutant emissions, the per capita gray water ecological footprint in most provinces and cities has shown a decreasing trend. A reduction in environmental efficiency is an inevitable consequence of a decrease in the water ecological footprint intensity. With a decrease in water ecological footprint intensity, more water is reused and we can improve water efficiency in most provinces to reduce the gray water ecological footprint. This study discusses the relationship between gray water ecological changes and capital factors; the reported results have value as a reference for the adjustment of environmental policies and the sustainable utilization of water resources. 参考文献 相似文献 引证文献

Blood cell analysis is becoming more and more attainable and worthy for blood diseases diagnosis. And blood cell segmentation is an important step for automated blood cells analysis based on image processing. This paper presented an improved support vector machine algorithm for blood cell segmentation for molecular hyperspectral images processing. This algorithm combined SVM with iterative self-organizing data analysis techniques algorithm to correct the miss-identified pixels by putting them into the minimum Euclidean distance cluster. Satisfactory performance can be seen from the experimental results that the proposed algorithm is superior to ISODATA and SVM algorithms in blood cell segmentation, because the new algorithm combines spatial and spectral information of blood cells.

Logistics location is an important component of logistics planning that affects traffic pressure and vehicle emissions. To date, there has not been an adequate study of the integration of big data into the location for a complicated logistics system. This study developed a decision support system that can address location problems for complicated logistics systems, e.g., a multilevel urban underground logistics system (ULS), using logistics big data. First, information needed in the logistics location, such as the traffic performance index (TPI) and the origin/destination (OD) matrix, was collected and calculated using a big data platform, and this information was digitized and represented based on a geographic information system (GIS) tool. Second, a two‐stage location model for a ULS was designed to balance the construction costs and traffic congestion. The first stage is establishing a set‐covering model to identify optimum locations for secondary hubs based on the ant colony optimization algorithm, and the second stage is clustering of the secondary hubs to determine locations for primary hubs using the iterative self‐organizing data analysis technique algorithm (ISODATA). Finally, the Xianlin district of Nanjing, China, was chosen as a case study to validate the effectiveness of the proposed system. The system can be used to facilitate logistics network planning and to promote the application of big data in logistics.

A comparison of land unit delineation techniques for land evaluation in the Western Cape, South Africa

The atomic stress tensor at a given continuum point is a spatial average value of some volume near the point. Recent progresses in multiscale modeling include the dealing of the optimal number and the size of these volumes. In this paper, we motivate the application of Iterative self-organizing data analysis technique algorithm to estimate volume numbers. The size of these space averaging volumes then could be got using Gaussian mixture model. Reduced computation complexity is offered by this method. Atomistic simulations are conducted to analyze the stress of a stone-wales defect graphene sheet to validate the method. Other multiscale values could also be determined using this method.

Land surface water is a key part in the global ecosystem balance and hydrological cycle. Remote sensing has become an effective tool for its spatio-temporal monitoring. However, remote sensing results exemplified in so-called water indices are subject to several limitations. This paper proposes a new and effective water index called the Sentinel Multi-Band Water Index (SMBWI) to extract water bodies in complex environments from Sentinel-2 satellite imagery. Individual tests explore the effectiveness of the SMBWI in eliminating interference of various special interfering cover features. The Iterative Self-Organizing Data Analysis Technique Algorithm (ISODATA) method and confusion matrix along with the derived accuracy evaluation indicators are used to provide a threshold reference when extracting water bodies and evaluate the accuracy of the water body extraction results, respectively. The SMBWI and eight other commonly used water indices are qualitatively and quantitatively compared through vision and accuracy evaluation indicators, respectively. Here, the SMBWI is proven to be the most effective at suppressing interference of buildings and their shadows, cultivated lands, vegetation, clouds and their shadows, alpine terrain with bare ground and glaciers when extracting water bodies. The overall accuracy in all tests was consistently greater than 96.5%. The SMBWI is proven to have a high ability to identify mixed pixels of water and non-water, with the lowest total error among nine water indices. Most notably, better results are obtained when extracting water bodies under interfering environments of cover features. Therefore, we propose that our novel and robust water index, the SMBWI, is ready to be used for mapping land surface water with high accuracy.

Wind power is a new energy source, for which the forecasting accuracy is related to the feasibility of various decision-making schemes for the power grid. The random, fluctuating and uncertain characteristics of wind power make accurate point forecasting very difficult and result in low practical application value. Therefore, a novel wind-power interval prediction method is proposed in this paper. First, variational model decomposition (VMD) is applied to the historical time sequence for wind power, followed by using an extreme learning machine (ELM) to establish a prediction model for the intrinsic mode function (IMF). An autoregressive integrated moving average (ARIMA) is used to establish a prediction model for the residual components, and the predicted IMF and residual components are reconstructed to predict the wind power. Finally, iterative self-organizing data analysis technique algorithm (ISODATA) are used with the error sequence to perform clustering segmentation, and adaptive diffusion kernel density estimation (ADKDE) is used to fit the probability density function for the prediction error of each segment and thereby obtain the cumulative distribution function and the final interval prediction under a given confidence level (CL). The results of a simulation based on real wind farm data from Hunan Province, China, from January to April 2020 show that the proposed method can significantly improve interval coverage.

Voice conversion with SI-DNN and KL divergence based mapping without parallel training data

Multispectral aerial photographs from 1986, 1989 and 1997 were used for mapping and assessment of changes in sub-alpine vegetation of the Krkonose Mts Plateau (1300 to 1400 m a. s. l.), Czech Republic. Scanned aerial photographs were digitally processed by various remote sensing methods, such as histogram equalization, principal component transformation, Normalised Difference Vegetation Index, and classification (both unsupervised, i.e., Iterative Self-Organizing Data Analysis Technique, and supervised, i.e., Parallelepiped and Maximum Likelihood algorithm). The results were geometrically corrected by registration into orthophoto images. Field data from 1991 assisted in class labelling and vegetation description. A detailed vegetation analysis was conducted using 1989 multispectral photographs since the field-verification and comparison to ground-based vegetation map could be done with only two-year difference. The unsupervised classification resulted in six classes (three of them mixed) with a 60.6% overall accuracy of unmixed classes. Some important vegetation types could not be separated. The supervised classification resulted in nine classes with the overall accuracy of 81.1%. The most common vegetation types were: mat-grass communities (40.4% of the total area) and tall grass and herb communities (23.1%). Pine stands, the best separated of vegetation types, were evaluated on the multispectral photographs from 1986 and 1997 and the P. mugo stands doubled in area over 10 years. The aerial-photograph based maps were compared with an available vegetation map obtained by a field survey. The correspondence between both ranged from 24 – 75%, with the highest values obtained for pine stands and mat-grass communities. In comparison to the field survey, aerial photography provided less detail and vegetation was characterized mostly by dominant species. The approach failed to map small patches of heterogeneous or atypical vegetation and to differentiate vegetation types with dominants of similar appearance. Nevertheless since this method is comparably less time- and labour-consuming, it seems suitable for mapping distinct vegetation types, surveying larger areas and identifying changes over time.

In this paper, we present a simple but novel framework to train a nonparallel many-to-many voice conversion (VC) model based on the encoder-decoder architecture. It is observed that an encoder-decoder text-to-speech (TTS) model and an encoder-decoder VC model have the same structure. Thus, we propose to pre-train a multi-speaker encoder-decoder TTS model and transfer knowledge from the TTS model to a VC model by (1) adopting the TTS acoustic decoder as the VC acoustic decoder, and (2) forcing the VC speech encoder to learn the same speaker-agnostic linguistic features from the TTS text encoder so as to achieve speaker disentanglement in the VC encoder output. We further control the conversion of the pitch contour from source speech to target speech, and condition the VC decoder on the converted pitch contour during inference. Subjective evaluation shows that our proposed model is able to handle VC between any speaker pairs in the training speech corpus of over 200 speakers with high naturalness and speaker similarity.