检索结果-内蒙古大学图书馆

Bandwidth selection for nonparametric modal regression

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION 2019年第4期48卷 968-984页

作者： Zhou, Haiming Huang, Xianzheng Northern Illinois Univ Div Stat De Kalb IL USA Univ South Carolina Dept Stat Columbia SC 29208 USA

In the context of estimating local modes of a conditional density based on kernel density estimators, we show that existing bandwidth selection methods developed for kernel density estimation are unsuitable for mode estimation. We propose two methods to select bandwidths tailored for mode estimation in the regression setting . Numerical studies using synthetic data and a real-life dataset are carried out to demonstrate the performance of the proposed methods in comparison with several well-received bandwidth selection methods for density estimation.

关键词： Bootstrap Cross-validation Hausdorff distance mean shift algorithm

来源：评论

学校读者我要写书评

暂无评论

Mixture model modal clustering

引用

ADVANCES IN DATA ANALYSIS AND CLASSIFICATION 2019年第2期13卷 379-404页

作者： Chacon, Jose E. Univ Extremadura Dept Matemat Badajoz 06006 Spain

The two most extended density-based approaches to clustering are surely mixture model clustering and modal clustering. In the mixture model approach, the density is represented as a mixture and clusters are associated to the different mixture components. In modal clustering, clusters are understood as regions of high density separated from each other by zones of lower density, so that they are closely related to certain regions around the density modes. If the true density is indeed in the assumed class of mixture densities, then mixture model clustering allows to scrutinize more subtle situations than modal clustering. However, when mixture modeling is used in a nonparametric way, taking advantage of the denseness of the sieve of mixture densities to approximate any density, then the correspondence between clusters and mixture components may become questionable. In this paper we introduce two methods to adopt a modal clustering point of view after a mixture model fit. Examples are provided to illustrate that mixture modeling can also be used for clustering in a nonparametric sense, as long as clusters are understood as the domains of attraction of the density modes. Finally, a simulation study reveals that the new methods are extremely efficient from a computational point of view, while at the same time they retain a high level of accuracy.

关键词： Mixture modeling Modal clustering Component merging mean shift algorithm

来源：评论

学校读者我要写书评

暂无评论

Fast multi-feature image segmentation

引用

APPLIED MATHEMATICAL MODELLING 2021年 90卷 742-757页

作者： Cuevas, Erik Becerra, Hector Luque, Alberto Abd Elaziz, Mohamed Univ Guadalajara Dept Elect CUCEI Av Revoluc 1500 Guadalajara 44430 Jalisco Mexico Zagazig Univ Fac Sci Dept Math Zagazig Egypt Tomsk Polytech Univ Sch Comp Sci & Robot Tomsk Russia

Multi-feature segmentation has demonstrated its superiority against one-dimensional feature approaches based on only grayscale information. mean shift (MS) is an algorithm that has been used commonly for multi-feature segmentation. In spite of its interesting results, MS maintains a computational cost that is prohibitive for segmentation scenarios where the feature map consists of multi-dimensional features. In this paper, a new competitive segmentation algorithm for grayscale images is introduced. The proposed approach considers a two-dimensional feature map that includes the grayscale value and the local variance for each pixel in the image. To reduce the computational cost, the mean shift (MS) algorithm is modified to operate with a very limited number of points from all available data. Under such conditions, two sets of elements are differentiated: involved data (the reduced dataset considered in the MS operation) and not involved data (the rest of the available data). Different from the classical MS, which employs Gaussian functions, in our approach, the process of estimating the feature map is carried out using a more accurate approach such as the Epanechnikov kernel function. Once the MS results are obtained, they are generalized to include the not involved data. Therefore, each unused element is assigned to the same cluster of the closest used data. Finally, clusters with the fewest elements are fused with other neighboring clusters. The proposed segmentation method has been compared with other state-of-art algorithms considering the full number of images from the Berkeley dataset. Experimental results confirm that the proposed scheme produces segmented images with a 50% better quality of visual perception approximately two times (approximate to 1.8 - 2) faster than its competitors. (C) 2020 Elsevier Inc. All rights reserved.

关键词： Image segmentation mean shift algorithm Kernel density estimator (KDE)

来源：评论

学校读者我要写书评

暂无评论

Human tracking using joint color-texture features and foreground-weighted histogram

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2018年第11期77卷 13947-13981页

作者： Laaroussi, Khadija Saaidi, Abderrahim Masrar, Mohamed Satori, Khalid Sidi Mohamed Ben Abdellah Univ Dept Math & Comp Sci Fac Sci LIIAN BP 1796 Fes Atlas Morocco Sidi Mohamed Ben Abdellah Univ Polydisciplinary Fac Taza Dept Math Phys & Comp Sci LSI BP 1223 Taza Morocco

This paper proposes a new appearance model for human tracking based on mean shift framework. The proposed method uses a novel target representation by using joint Color-Texture features and Foreground-Weighted Histogram (CTFWH) for a more distinctive and effective target representation. Our contribution is threefold: firstly, to exploit the texture information of the target, we have used joint color-texture histogram to represent the target. Local Binary Pattern (LBP) technique is employed to identify texture features in the target region. Secondly, we have proposed a representation model of the foreground region named Foreground-Weighted Histogram (FWH), in order to exploit the significant features of the foreground region and to use it for selecting only the salient parts from the target model. Thirdly, we propose a simple method to update the foreground model due to the important foreground changes over the tracking process. Hence, by combining these concepts we generate new features for target representation and human tracking. The proposed method is designed for human tracking in complex scenarios and tested for comparative results with existing state-of-the-art algorithms. Experimental results on numerous challenging video sequences verify the significance of the proposed approach in terms of robustness and performance to complex background, illumination and appearance changes, similar target and background appearance, presence of distractors, target and camera motion, occlusions and large background variation.

关键词： Human tracking mean shift algorithm Color-texture histogram Foreground-weighted histogram

来源：评论

学校读者我要写书评

暂无评论

A Two-Stage Charging Facilities Planning Method for Electric Vehicle Sharing Systems

引用

IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS 2021年第1期57卷 149-157页

作者： Wang, Ning Wang, Chengfu Niu, Yuanfang Yang, Ming Yu, Yixiao Shandong Univ Sch Elect Engn Jinan 250100 Peoples R China Shandong Elect Power Engn Consulting Inst Co Ltd Jinan 250013 Peoples R China

The large-scale sharing electric vehicle (SEV) need adequate charging infrastructure to maintain the optimal operation. Considering the convenience of SEV users' charging during driving, a two-stage planning method combining charging stations and scattered charging piles for SEV is proposed to balance the economy of SEV operators and the convenience of SEVs users. In the first stage, according to the urban area division and trip characteristics of SEV, the charging power demand of SEV in each SEV placement point can be estimated exactly. Then, aiming for the minimum annual total cost, the mean clustering algorithm and queuing theory are used to determine the location and capacity of charging stations. In the second stage, an improved S-shaped function is used to assess the unsatisfied charging demand points of SEV users, and the compensated charging piles without in the stations are planned based on the mean shift algorithm. Based on the re-division of the service scope, the capacity of the charging stations is updated. Finally, the simulation results verifying the proposed model and algorithm is effective and feasible.

关键词： Charging stations Planning Electric vehicles Clustering algorithms Capacity planning Urban areas Space vehicles Charging facility mean clustering mean shift algorithm queuing theory sharing electric vehicle (SEV) two-stage planning

来源：评论

学校读者我要写书评

暂无评论

Design of a Real-Time Monitoring and Early Warning System for Engineering Safety Hazards Using Image Analysis Technology

引用

TRAITEMENT DU SIGNAL 2024年第5期41卷 2381-2390页

作者： Xing, Haoran Univ New South Wales Art Design & Architecture Sydney 2033 Australia

As the scale of engineering projects continues to grow, safety management on construction sites faces significant challenges. Traditional methods such as manual inspections and periodic checks struggle to achieve real-time and effective monitoring of potential hazards, which can lead to accidents. In recent years, image analysis technology has increasingly been applied to the monitoring of engineering safety hazards due to its automation, Areal-time capabilities, and high efficiency. However, existing image analysis algorithms still encounter issues such as insufficient tracking accuracy and delayed warning responses in complex engineering environments. To address these problems, this study proposes aArealtime hazard tracking and identification method based on an improved mean shift algorithm, combined with a support vector machine (SVM) for critical state early warning of engineering safety hazards. The system improves recognition accuracy and early warning response speed in complex environments through algorithm optimization, offering higher practicality and reliability. This provides a technical safeguard for safety management at construction sites.

关键词： engineering safety hazards image analysis mean shift algorithm support vector machine (SVM) Areal-time monitoring early warning system

来源：评论

学校读者我要写书评

暂无评论

Data-driven density derivative estimation, with applications to nonparametric clustering and bump hunting

引用

ELECTRONIC JOURNAL OF STATISTICS 2013年第none期7卷 499-532页

作者： Chacon, Jose E. Tarn Duong Univ Extremadura Dept Matemat E-06006 Badajoz Spain Univ Paris 06 Theoret & Appl Stat Lab LSTA F-75005 Paris France Grp Hosp Pitie Salpetriere Inst Translat Neurosci IHU A ICM F-75005 Paris France

Important information concerning a multivariate data set, such as clusters and modal regions, is contained in the derivatives of the probability density function. Despite this importance, nonparametric estimation of higher order derivatives of the density functions have received only relatively scant attention. Kernel estimators of density functions are widely used as they exhibit excellent theoretical and practical properties, though their generalization to density derivatives has progressed more slowly due to the mathematical intractabilities encountered in the crucial problem of bandwidth (or smoothing parameter) selection. This paper presents the first fully automatic, data-based bandwidth selectors for multivariate kernel density derivative estimators. This is achieved by synthesizing recent advances in matrix analytic theory which allow mathematically and computationally tractable representations of higher order derivatives of multivariate vector valued functions. The theoretical asymptotic properties as well as the finite sample behaviour of the proposed selectors are studied. In addition, we explore in detail the applications of the new data-driven methods for two other statistical problems: clustering and bump hunting. The introduced techniques are combined with the mean shift algorithm to develop novel automatic, nonparametric clustering procedures which are shown to outperform mixture-model cluster analysis and other recent nonparametric approaches in practice. Furthermore, the advantage of the use of smoothing parameters designed for density derivative estimation for feature significance analysis for bump hunting is illustrated with a real data example.

关键词： Adjusted Rand index cross validation feature significance nonparametric kernel method mean integrated squared error mean shift algorithm plug-in choice

来源：评论

学校读者我要写书评

暂无评论

A fast approach for person detection and tracking

引用

INTERNATIONAL JOURNAL OF COMPUTER APPLICATIONS IN TECHNOLOGY 2012年第3期44卷 210-216页

作者： Ali, Ahmed Terada, Kenji Univ Tokushima Grad Sch Adv Technol & Sci 2-1 Minami Josanjima Tokushima 7708506 Japan

Object detection and tracking is one of the most popular areas of video processing and the essential requirement of any surveillance system. A fast approach for person detection and tracking is presented. This work proposes to do target tracking with Kalman filter. In case of a misdetection, which would lead a wrong update of the filter, a fast mean shift iteration based on integral computation is performed to propose a more accurate detection. Tracking results are demonstrated for complex scenes and evaluation of the proposed tracking approach is presented.

关键词： Kalman filter mean shift algorithm object tracking

来源：评论

学校读者我要写书评

暂无评论

Object Tracking with the Multi-Templates Regression Model Based MS algorithm

引用

JOURNAL OF INFORMATION PROCESSING SYSTEMS 2018年第6期14卷 1307-1317页

作者： Zhang, Hua Wang, Lijia Shijiazhuang Univ Appl Technol Coll Elect & Elect Engn Shijiazhuang Hebei Peoples R China Hebei Coll Ind & Technol Dept Intelligent Manufacture Shijiazhuang Hebei Peoples R China

To deal with the problems of occlusion, pose variations and illumination changes in the object tracking system, a regression model weighted multi-templates mean-shift (MS) algorithm is proposed in this paper. Target templates and occlusion templates are extracted to compose a multi-templates set. Then, the MS algorithm is applied to the multi-templates set for obtaining the candidate areas. Moreover, a regression model is trained to estimate the Bhattacharyya coefficients between the templates and candidate areas. Finally, the geometric center of the tracked areas is considered as the object's position. The proposed algorithm is evaluated on several classical videos. The experimental results show that the regression model weighted multi-templates MS algorithm can track an object accurately in terms of occlusion, illumination changes and pose variations.

关键词： mean shift algorithm Multi-Templates Object tracking Regression Model

来源：评论

学校读者我要写书评

暂无评论

Calculation of local skin doses with ICRP adult mesh-type reference computational phantoms

引用

JOURNAL OF THE KOREAN PHYSICAL SOCIETY 2018年第1期72卷 177-182页

作者： Yeom, Yeon Soo Han, Haegin Choi, Chansoo Thang Tat Nguyen Lee, Hanjin Shin, Bangho Kim, Chan Hyeong Han, Min Cheol Hanyang Univ Dept Nucl Engn Seoul 04763 South Korea Ist Nazl Fis Nucl Sez Genova I-16146 Genoa Italy

Recently, Task Group 103 of the International Commission on Radiological Protection (ICRP) developed new mesh-type reference computational phantoms (MRCPs) for adult males and females in order to address the limitations of the current voxel-type reference phantoms described in ICRP Publication 110 due to their limited voxel resolutions and the nature of the voxel geometry. One of the substantial advantages of the MRCPs over the ICRP-110 reference phantoms is the inclusion of a 50-mu m-thick radiosensitive skin basal-cell layer;however, a methodology for calculating the local skin dose (LSD), i.e., the maximum dose to the basal layer averaged over a 1-cm(2) area, has yet to be developed. In the present study, a dedicated program for the LSD calculation with the MRCPs was developed based on the mean shift algorithm and the Geant4 Monte Carlo code. The developed program was used to calculate local skin dose coefficients (LSDCs) for electrons and alpha particles, which were then compared with the values given in ICRP Publication 116 that were produced with a simple tissue-equivalent cube model. The results of the present study show that the LSDCs of the MRCPs are generally in good agreement with the ICRP-116 values for alpha particles, but for electrons, significant differences are found at energies higher than 0.15 MeV. The LSDCs of the MRCPs are greater than the ICRP-116 values by as much as similar to 2.7 times at 10 MeV, which is due mainly to the different curvature between realistic MRCPs (i.e., curved) and the simple cube model (i.e., flat).

关键词： ICRP Mesh-type reference phantom Local skin dose mean shift algorithm Geant4

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：