检索结果-南通市图书馆

检索条件"导师=YU ZHULIANG"

共 3 条记录，以下是1-10 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

Research on Fast R-CNN for Visual Object Detection

Research on Fast R-CNN for Visual Object Detection

引用

作者： GHULAM ARBI 华南理工大学

学位级别：硕士

Over the last decade,object detection has emerged as an important area of research for computer *** this regard,some novel and impressing CNN architectures have been *** this work Fast R-CNN and VGGNet architectures have been incorporated for the given task of visual object *** proposed methods have been exercised and empirically it is realized that Fast R-CNN architecture exhibits considerable performance for the given *** particular for given problem,emphasis has been rendered on visual object Capra Falconeri（Markhor）,in different challenging environment and background *** most issue for object detection using deep learning techniques is to save the training and testing time without sacrificing its ***,existing Fast R-CNN technique provide good accuracy but it suffers from more training and testing time that limits the overall performance of *** this thesis,a new dataset is created named as Markhor-VI by various data augmentation methods such as translation,rotation,scaling and so *** we have modified existing Fast R-CNN technique by transfer learning by various methods from layer 1 to layer *** is found that convolutional object detection is still evolving as a technology,despite outranking other object detection *** is also found that transfer learning can save a lot of resources if fine tuning is performed in a particular way.

关键词：

来源：

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

Time Series Prediction Using Echo State Network

Time Series Prediction Using Echo State Network

引用

作者： MD TAUSIF ALAM 华南理工大学

学位级别：硕士

In the past few years,Time Series Analysis and Prediction based on the Echo State Network（ESN）has gently grief the attention of researchers at home and abroad,which become a central research issue with relevant theoretical and application-related *** State Networks is one of the most efficient types of Recurrent Neural Networks（RNN）.However,they are often present of cluster data,making the process of finding and efficient ESN for specific datasets quite *** our current study,preferring a novel Recurrent Neural Network – Echo State Network to predict the next Stock Market Closing Price between a defined time series based on prior Statistical *** the stock price movements,in this research also propose the Mackey-Glass system to determine how different global parameters optimize training in the echo state *** on the chaos in financial time series data has led us to another understanding of the distribution of relative stock prices over *** doing research found that Lorentzian and Voigt profile distribution is a good model that can be used to explain the representatives of substantial gains and losses thick tail but did not disclose this in a typical Gaussian *** distributions serve as an untrained stochastic model for the prediction based on historical price changes in the S＆P 500 index and the training of the Echo State Network of *** experimental results and output indicate that ESN with Mackey-Glass System demonstrates efficient results compared with different methodologies.

关键词：股票市场股价波动回声状态网络时间序列预测

来源：

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

Automated Complex Fabric Faults Detection Using a Hybrid Method

Automated Complex Fabric Faults Detection Using a Hybrid Met...

引用

作者： Abdullah-Al-Mamun 华南理工大学

学位级别：硕士

Fault detection or identification is highly important to fabric quality ***,faults are detected by human *** efficiency of this manual technique is low and the missed rate is high because of eye *** the best case,a quality control person cannot detect more than 60–70 % of the present *** this study a Hybrid technique is developed to identify complex faults of warp-knitted fabrics *** this system,we use smart cameras and a Human Machine Interface（HMI）*** hybrid detection or identification algorithm is comprised of numerous single techniques such as Gabor filter,PCNN,morphological filter,and another associated technique（Binarizing）those are running on the SOC processor of the smart camera for identifying distinctive types of complex faults on ***,Gabor filters are employed to enhance the contrast of images captured by a CMOS *** fault is not detected after Gabor filter it goes through ***,fault areas are segmented by PCNN with adaptive parameter ***,for making the PCNN output more clear and faults more obvious,the morphological filter has been used in this detection or identification ***,smart cameras will notice the controller to stop the warp-knitting machine once faults are found *** results demonstrate that the hybrid technique is superior to Gabor and PCNN techniques on fabric complex faults detection or identification *** this hybrid technique we get 97.08% fault detection or identification accuracy which is superior to Gabor and PCNN *** operations in a textile factory can be verified the effectiveness of the inspection system.

关键词：复杂织物故意诊断缺陷检测混合方法机器视觉

来源：

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

全选清除本页清除全部题录导出标记到“检索档案”

共1页 << < 1 > >>

回到顶部

执行限定条件

欢迎您,

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

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

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：