检索结果-内蒙古大学图书馆

Face detection using a first-order RCE classifier

EURASIP JOURNAL ON APPLIED SIGNAL PROCESSING 2003年第9期2003卷 878-889页

作者： Jeon, BH Lee, KM Lee, SU Seoul Natl Univ Sch Elect Engn Signal Proc Lab Seoul 151742 South Korea Samsung Elect Co Ltd Mechatron Ctr Inst Intelligent Syst Suwon 442742 Gyeonggi Do South Korea Hongik Univ Dept Elect & Elect Engn Seoul 121711 South Korea

We present a new face detection algorithm based on a first-order reduced Coulomb energy (RCE) classifier. The algorithm locates frontal views of human faces at any degree of rotation and scale in complex scenes. The face candidates and their orientations are first determined by computing the Hausdorff distance between simple face abstraction models and binary test windows in an image pyramid. Then, after normalizing the energy, each face candidate is verified by two subsequent classifiers: a binary image classifier and the first-order RCE classifier. While the binary image classifier is employed as a preclassifier to discard nonfaces with minimum computational complexity, the first-order RCE classifier is used as the main face classifier for final verification. An optimal training method to construct the representative face model database is also presented. Experimental results show that the proposed algorithm yields a high detection ratio while yielding no false alarm.

关键词： face detection face model Hausdorff distance clustering algorithm RCE classifier

来源：评论

学校读者我要写书评

暂无评论

VizOPTICS: Getting insights into OPTICS via interactive visual analysis

引用

COMPUTERS & ELECTRICAL ENGINEERING 2023年第1期107卷

作者： Wu, Caixia Chen, Yi Dong, Yu Zhou, Fangfang Zhao, Ying Liang, Christy Jie Beijing Technol & Business Univ Beijing Key Lab Big Data Technol Food Safety Beijing 100048 Peoples R China Univ Technol Sydney Sch Comp Sci Sydney NSW 2008 Australia Cent South Univ Sch Comp Sci & Engn Changsha 410083 Peoples R China

Ordering points to identify the clustering structure (OPTICS) is a density-based clustering algorithm that allows the exploration of the cluster structure in the dataset by outputting an ordered queue called cluster ordering. However, this nonexplicit output makes it greatly more difficult for practitioners to identify cluster patterns and obtain high-quality clusters. In this paper, we firstly investigate OPTICS in depth and identify the challenges facing users of OPTICS for cluster analysis through a pilot user study. Then, integrating human intelligence deeply with the machine intelligence of OPTICS, a visual analytics approach, VizOPTICS, is proposed to support practitioners in understanding and applying OPTICS to extract meaningful clustering results. It includes an ordered lattice plot for observing the generation process of cluster ordering, a density scatter plot for analyzing the cluster structure in datasets, and a dynamic reachability plot for optimizing clustering results, and also provides several interaction modes, such as selecting and highlighting, to help users analyze the cluster formation and algorithm operation processes interactively. Finally, we assess our approach through four case studies and a user evaluation study. The results demonstrate the effectiveness and efficiency of the system.

关键词： clustering algorithm OPTICS Cluster analysis Visualization Interaction Explainable machine learning

来源：评论

学校读者我要写书评

暂无评论

Ultrasonic signal enhancement for coarse grain materials by machine learning analysis

引用

ULTRASONICS 2021年 117卷 106550-106550页

作者： Xu, Wanli Li, Xiaohong Zhang, Jun Xue, Zhixiang Cao, Jiancheng Wuhan Univ Sch Power & Mech Engn Key Lab Hydraul Machinery Transients Minist Educ Wuhan 430072 Hubei Peoples R China Southeast Univ Sch Mat Sci & Engn Nanjing 211189 Jiangsu Peoples R China

This paper aims at dealing with the dilemma of examining the existence of a defect in ultrasonic detection of coarse grain materials. In such cases, defect echoes can be drowned in a strong noise background resulting from intricate coarse grain scattering, that is, grain noise. To this end, we develop an innovative signal reconstruction methodology from polluted measurements which combines basic statistical analysis with a series of machine learning algorithms. The proposed methodology analyzes abundant information from numerous raw signals to distinguish the desired signal from grain noise, avoiding the limitation of information provided only by a single signal. The technique is achieved by collecting similar signals together through a clustering algorithm and subsequently inputting these similar signals to a denoising autoencoder to suppress the grain noise. It is successfully employed to ultrasonic signals obtained from an as-cast stainless steel specimen with coarse equiaxed grains, a stainless steel specimen with relatively homogeneous dendrite fabricated by additive manufacturing and a stainless steel weld with heterogeneous columnar grains having variation of grain sizes in various locations. The influence of material microstructure and probe frequency on denoising performance is investigated in detail. Based on this, the proposed methodology is applied to defect detection. Desired A-scan results and B-scan imaging are achieved by the proposed method, where defects are well revealed. The experimental results demonstrate the developed methodology has stable excellent performance and superior denoising capabilities for defect detection with respect to conventional techniques, especially in the case where the noise is almost the same as the desired signal.

关键词： Grain noise Ultrasonic testing Signal reconstruction Machine learning clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

A novel method for quantitative fault diagnosis of photovoltaic systems based on data-driven

引用

ELECTRIC POWER SYSTEMS RESEARCH 2022年第0期210卷

作者： Guo, Hui Hu, Shan Wang, Fei Zhang, Lijun Shanghai Univ Shanghai Key Lab Power Stn Automat Technol Shanghai Peoples R China Univ Lisbon Inst Super Tecn Lisbon Portugal

Photovoltaic (PV) systems are usually exposed outdoors for a long time, easily lead to various faults. Apart from fault location and classification widely studied, accurate and quantitative fault severity diagnosis is vital for the decision making, given the cost of eliminating faults. However, there are scarce investigations into this problem. Therefore, a novel quantitative fault diagnosis method is proposed based on the deviation between the predicted power of fault states and the actual power of normal states. (1) Since the main characteristic of PV faults is the decline of power generation, the prediction model of PV power generation under the normal state is established via clustering algorithm and long short-term memory (LSTM) neural network, which has better processing ability for different weather conditions and long time-series data. (2) In the actual condition of limited fault data, the prediction model is innovatively used to quantitative fault diagnosis through transfer learning, which will provide reasonable and effective guidance for the intelligent operation and maintenance of PV system when the severity of predicted fault reaches the set threshold. Finally, the effectiveness and accuracy of the proposed quantitative fault diagnosis method are verified by experimental results.

关键词： Photovoltaic (PV) system Quantitative fault diagnosis Data-driven clustering algorithm Long short-term memory (LSTM) neural network Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Prospects and obstacles analysis of applying blockchain technology to power trading using a deeply improved model based on the DEMATEL approach

引用

SUSTAINABLE CITIES AND SOCIETY 2021年 70卷 102910-102910页

作者： Zhou, Jianli Wu, Yunna Liu, Fangtong Tao, Yao Gao, Jianwei North China Elect Power Univ Sch Econ & Management Beijing 102206 Peoples R China North China Elect Power Univ Beijing Key Lab New Energy & Low Carbon Dev Beijing 102206 Peoples R China

Blockchain technology is widely concerned, and its related applications can promote the process of smart cities and sustainable society. However, while mining the potential application scenarios in power trading, we must recognize the barriers, help it survive the hype stage, and promote its healthy development and technology landing. For the first time, hesitant fuzzy linguistic term set and K-mediods clustering algorithm are used to improve the decision-making trial and evaluation laboratory (DEMATEL) method, and the obstacle analysis model of the applied scene is constructed. Compared with the conventional DEMATEL method, the collection of evaluation information is more flexible and closer to reality. Besides, the classification of obstacle factors is more scientific and there can be more than two categories for effect degree. Firstly, thirteen barriers to its application in power trading are identified;at the same time, six specific application scenarios are summarized and analyzed. Then, a detailed discussion is conducted on each scenario: The quantification of the influence degree among obstacles, the classification and qualitative of the influence degree, and the causal mechanism analysis. The key obstacles identified can be used to guide practice. Finally, strategic solutions and policy recommendations are given to remove or alleviate these obstacles.

关键词： Power trading Sustainable consumption Obstacles analysis Blockchain technology DEMATEL method clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Research on Fabric Defect Detection algorithm Based on Lightweight YOLOv7-Tiny

引用

JOURNAL OF NATURAL FIBERS 2024年第1期21卷

作者： Li, Tang Shunqi, Mei Yishan, Shi Shi, Zhou Quan, Zheng Jiang, Hongkai Qiao, Xu Zhiming, Zhang Wuhan Text Univ Hubei Digital Text Equipment Key Lab Wuhan Peoples R China Jianhu Lab Adv Text Technol Innovat Ctr Shaoxing Peoples R China Zhejiang Tianxiong Ind Technol Co Ltd Shenzhen Zhejiang Peoples R China Chinese Univ Hongkong Hongkong Peoples R China Wuhan Text Univ Coll Mech Engn & Automat 1 Sunshine Ave Wuhan 430200 Hubei Peoples R China

The current advanced neural network models are expanding in size and complexity to achieve improved detection accuracy. This study designs a lightweight fabric defect detection algorithm based on YOLOv7-tiny, called YOLOv7-tiny-MGCK. Its objectives are to improve the performance of fabric defect detection against complex backgrounds and to find a balance between the algorithm's lightweight nature and its accuracy. The algorithm utilizes the Mish activation function, known for its superior nonlinear performance capability and smoother curve, enabling the neural network to manage more complex challenges. The Ghost convolution module is also incorporated to reduce computation and model parameters. The lightweight upsampling technique CARAFE facilitates the flexible extraction of deep features, coupled with their integration with shallow features. In addition, an improved K-Means clustering algorithm, KMMP, is employed to select appropriate anchor box for fabric defects. The experimental results show: a reduction in the number of parameters by 45.5% and computational volume by 41.0%, along with increases in precision by 3.9%, recall by 7.0%, and mAP by 3.0%. These results indicated that the improved algorithm achieves a more effective balance between detection performance and the requirement for a lightweight solution.

关键词： Neural network fabric defect detection activation function Ghost module upsampling clustering algorithm (sic)(sic)(sic)(sic) (sic)(sic)(sic)(sic)(sic)(sic) (sic)(sic)(sic)(sic) (sic)(sic)(sic) (sic)(sic)(sic) (sic)(sic)(sic)(sic)

来源：评论

学校读者我要写书评

暂无评论

Collaborative Evaluation Method of Multiple Team Membership: Case Study of Engineering Design Company

引用

JOURNAL OF MANAGEMENT IN ENGINEERING 2022年第6期38卷 05022014-05022014页

作者： Jiang, Qi Qiang, Maoshan Lin, Chen Tsinghua Univ State Key Lab Hydrosci & Engn Inst Project Management & Construct Technol Beijing 100084 Peoples R China Beijing Inst Water Dept Technol & Qual Management Landscape Architecture Beijing 100048 Peoples R China

Team members' project collaborative process evaluation is the key link to improve the management effectiveness of an organization. However, there is a lack of evaluation methods from the perspective of multiple team membership (MTM). Based on the data in the project management system (PMS) of engineering design enterprises, this article studies the project participation evaluation method of team members from the perspective of MTM. Using a clustering algorithm, this article developed and verified the participation evaluation index of single projects and multiple projects, a classification and statistics method, and the centrality evaluation method of team members. The research revealed the distribution law of MTM quantity and the influence of team members' knowledge and ability on it. We found that this index can evaluate objectively team members' project participation, which can divided into three levels: high, medium, and low. Collaborative activities that contribute significantly to the project account for 20% of the total. The reasonable range for team members to invest in high participation projects is from 2 to 4. For medium participation projects, the number is from 6 to 8. For low participation projects, the number is from 18 to 25, or even unlimited. This study found four positive and four negative effects of personal characteristics on work efficiency. When individuals have these characteristics, team members can better play to their abilities under the project management mechanism, which provides a reference for enterprises to identify and cultivate team members.

关键词： Multiple team membership (MTM) Project participation Collaborative evaluation method clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Optimization scheme of the orientation relationship from crystallographic statistics of variants and its application to lath martensite

引用

MATERIALS & DESIGN 2020年 195卷

作者： Sun, Duanjun Li, Chuanwei Xue, Xuan Liu, Yi Guo, Zhenghong Gu, Jianfeng Shanghai Jiao Tong Univ Sch Mat Sci & Engn Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Inst Mat Modificat & Modelling Shanghai 200240 Peoples R China

Without sacrificing any solution accuracy, an efficient optimization scheme of determining the orientation relationship only based on the daughter phase (i.e., lath martensite in steels in this paper) was established in this work, which involves two key steps. Firstly, an accurate clustering algorithm was developed to perform a statistic of all variants, which can give the most probable orientation and area fraction for each variant. Different from previous methods, the second step is to perform a newly-established optimization scheme, which is able to take variant fractions into consideration to determine the optimal orientation relationship and orientation of prior austenite simultaneously. In this case, the variant-level calculation, which can save much computational time compared with pixel/domain-level calculation, is beneficial for the efficiency promotion. Through inputting the equally-weighted variants into the optimization scheme, the influence of the random cross-sections on the solutions can be effectively eliminated. Furthermore, in-situ high temperature EBSD examinations were first performed to validate solution accuracy, the corresponding results clearly indicated that the orientation of austenite obtained from this approach exhibited a high accuracy and the error is less than 1 degrees. (c) 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).

关键词： clustering algorithm Optimization scheme Orientation relationship In-situ high temperature EBSD Martensitic transformation

来源：评论

学校读者我要写书评

暂无评论

Research on algorithm of Vehicle Track Data Mining Based on Cloud Computing

引用

AGRO FOOD INDUSTRY HI-TECH 2017年第1期28卷 1439-1443页

作者： Sun Hailong Li Fangsong NorthEast Forestry Univ Coll Informat & Comp Engn Haerbin 150040 Heilongjiang Peoples R China Comp Ctr Heilongjiang Prov Haerbin 150026 Heilongjiang Peoples R China

With the development of information technology and network, the face of today's large amount of data, data mining needs to be carried out in order to get useful information. The analysis and processing of these huge data requires computing powerful cloud computing. Based on the cloud computing technology, the vehicle trajectory was researched by data mining. Based on the basic principles of cloud computing and data mining, a vehicle trajectory data mining system based on cloud computing was constructed. The clustering algorithm was used to cluster the vehicle trajectory. Finally, the system was implemented and tested. The realization results show that the system can obtain the trajectory information of the vehicle effectively, which makes a certain contribution to the management of vehicles and the supervision of traffic.

关键词： Cloud computing data mining vehicle trajectory clustering algorithm

来源：评论

学校读者我要写书评

暂无评论

Electricity-consuming forecasting by using a self-tuned ANN-based adaptable predictor

引用

ELECTRIC POWER SYSTEMS RESEARCH 2022年第0期210卷

作者： Baba, Abdullatif Univ Turkish Aeronaut Assoc Comp Engn Dept Bahcekapi Quarter Okul St 11 TR-06790 Ankara Turkey

Accurate forecasting of power consumption is an essential and modern approach for planning smart infrastructural projects that are required to overcome the future challenges of power markets. In this context, a new design of a self-tuned ANN-based adaptable predictor is presented in this paper. At first, the main design of the original adaptable predictor is clarified including its new architecture that partially relies on the Hebbian law, its training process is also explained as well as the dataset which is used for training. Then, all the details of the self-tuning-based technique are explained with all the relevant results that prove its high capability to produce more accurate forecasting outcomes. The impact of our suggested approach is introduced by explaining two different practical examples that employ the K-means clustering algorithm, and the genetic algorithm when it is used to optimize the operating/outage schedule for a group of local solar units, respectively.

关键词： Self-tuning Adaptable predictor ANN model Smart grids clustering algorithm Evolutionary algorithms

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：