• <tr id='9qXm1f'><strong id='9qXm1f'></strong><small id='9qXm1f'></small><button id='9qXm1f'></button><li id='9qXm1f'><noscript id='9qXm1f'><big id='9qXm1f'></big><dt id='9qXm1f'></dt></noscript></li></tr><ol id='9qXm1f'><option id='9qXm1f'><table id='9qXm1f'><blockquote id='9qXm1f'><tbody id='9qXm1f'></tbody></blockquote></table></option></ol><u id='9qXm1f'></u><kbd id='9qXm1f'><kbd id='9qXm1f'></kbd></kbd>

    <code id='9qXm1f'><strong id='9qXm1f'></strong></code>

    <fieldset id='9qXm1f'></fieldset>
          <span id='9qXm1f'></span>

              <ins id='9qXm1f'></ins>
              <acronym id='9qXm1f'><em id='9qXm1f'></em><td id='9qXm1f'><div id='9qXm1f'></div></td></acronym><address id='9qXm1f'><big id='9qXm1f'><big id='9qXm1f'></big><legend id='9qXm1f'></legend></big></address>

              <i id='9qXm1f'><div id='9qXm1f'><ins id='9qXm1f'></ins></div></i>
              <i id='9qXm1f'></i>
            1. <dl id='9qXm1f'></dl>
              1. <blockquote id='9qXm1f'><q id='9qXm1f'><noscript id='9qXm1f'></noscript><dt id='9qXm1f'></dt></q></blockquote><noframes id='9qXm1f'><i id='9qXm1f'></i>
                首页> 外文学位 >文献详情
                【6h】

                A new integrated web centric database-batch (IWCDB) coupled framework for object recognition and classification in security applications.

                机译:一个新的集成的以网络为中心的数据库批处理(IWCDB)耦合框架,用于安全应用程序中的对象识别和分类。

                获取原文
                获取原文并翻译 | 示例
                ¥30

                摘要

                This thesis describes an integrated framework that brings together the fields of Digital Image processing (DIP), web technology, Database system and batch processing to address object recognition and classification problems in computer vision. With the recent terrorist attacks across the globe places such as ports of entry have become important places to enforce stronger security measures. Security scanners work in tandem with image processing algorithms to prevent terrorist attacks by identifying threat objects like guns, chemical liquids in bottles and explosives in passenger baggage and cargo containers. In this thesis, an integrated web based object recognition and classification framework is proposed and demonstrated for automatic threat detection. The novel features of this framework include utilizing the Human Visual System model for segmentation, and a new ratio based edge detection algorithm that includes a new adaptive hysteresis thresholding method. The feature vectors of the baseline images are generated and stored in a relational database system using a batch window. The feature vectors of the segmented objects are generated using the Cell edge distribution (CED) method and are classified using a support vector machine (SVM) based classifier to identify threat objects. The framework leverages the strength of a database and batch processing system with web technology to facilitate the development of reusable, portable and scalable real time threat detection application. The experimental results demonstrate the proposed framework efficiency in reducing the classification time and provide accurate detection.
                机译:本文描述了一个集成的框架,该框架将数字图像处理(DIP),Web技术,数据库系统和批处理领域融合在一起,以解决计算机视觉中的对象识别和分类问题。随着最近在全球范围内发生的恐怖↘袭击,诸如入境口岸等地方已成为执行更强有力的安全措施的重要场所。安全扫描仪与图像处理算法协同工作,通过识别威胁对象(例如枪支,瓶中的化学液体以及旅客行李和货物集装箱中的爆炸物)来防止恐怖㊣袭击。本文提出了一种基于Web的集成目标识别和分类框架,并进行了自动威》胁检测。该框架的新颖功能包括利用人类视觉系统模型进行分割,以及基于比率的新型边缘检测算法,其中包括新的自适应滞后阈值化方法。使用批处理窗口生成基线图像的特征向量并将其存储在关系数据库系统中。分割对象的特征向量←是使用单元边缘分布(CED)方法◤生成的,并使用基于支持向量机(SVM)的分类器进行分类以识别威胁对象。该框架利用Web技术利用数据库和批处理系统的优势来⊙促进可重用,可移植和可伸缩的实时威胁检测应用程序的开发。实验结果证明了所提出〖的框架在减少分类时间╱和提供准确检测方面的效率。

                著录项

                • 作者单位

                  Tufts University.;

                • 授予单位 Tufts University.;
                • 学科 Engineering, Electronics and Electrical.
                • 学位 M.S.
                • 年度 2010
                • 页码 138 p.
                • 总页数 138
                • 原文格式 PDF
                • 正文语种 eng
                • 中图分类 ;
                • 原文服务方 国家工程技术数字图书馆
                • 关键词

                  ;

                相似文献

                • 外文文献
                站内服务
                • 论文查重
                • 论文收录引证报告
                • 文档翻译
                • 文档转换
                获取原文

                联系方式:18141920177 (微信同号)

                客服邮箱:kefu@zhangqiaokeyan.com

                京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司?版权所有
                • 客服微信

                • 服务号