...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of computational and theoretical nanoscience >Content Based Image Retrieval Using Quad Tree Block Truncation Coding with Color Co-Occurrence Feature for the Big Data Platform
【24h】

Content Based Image Retrieval Using Quad Tree Block Truncation Coding with Color Co-Occurrence Feature for the Big Data Platform

机译：基于内容基于图像检索使用Quad Tree块截断与大数据平台的颜色共同发生功能进行编码

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Image retrieval systems are widely used as the requirement increases with the high visual and media data. It is highly important to understand and manage these big data since a lot of analysis needs to happen for image retrieval applications. It is a known fact that it is impossibleto handle these data in our traditional way and hence there is an urge to find the efficient way to manage big data with more accuracy and robustness. Hence we have proposed a new prototype in this paper called the “content Based Image Retrieval System.” Querying and searchingis an important process in image retrieval systems. This paper provides a new approach to alter the process of deriving the image features from the given image. This is done using the proposed algorithm called the Quad Tree Block Truncation Coding (QTBTC) compressed image. In this method theimage feature descriptor is done using QTBTC and after which the color quantizers and the bitmap images are derived. The CHF (color histogram feature) is calculated using the two color quantizers which represent the image contrast and color distribution of the image and the BPF (bit patternfeature) is constructed from the bitmap image and represents the edges and texture details of the image. These attributes are then represented using Vector quantization (VQ) to finally generate the image feature descriptor. The image features are now used to form the index of an image calledthe cooccurrence feature (CCF) and bit pattern features (BPF). Visual codebook is used to construct the index of the images. Support Vector Machine (SVM) is used for image classification by breaking the images into sub images.
机译：图像检索系统被广泛使用，因为需要使用高视觉和媒体数据增加。自我分析需要发生自我分析以进行图像检索应用，非常重要，非常重要。这是一种已知的事实，即ipossibleto以我们的传统方式处理这些数据，因此有一种促使能够以更准确和鲁棒性管理大数据的有效方法。因此，我们提出了本文的新原型称为“基于内容的图像检索系统”。查询和搜索图像检索系统中的一个重要过程。本文提供了一种改变来自给定图像的图像特征的过程的新方法。这是使用称为四边形块截断编码（QTBTC）压缩图像的所提出的算法完成。在此方法中，Mage特征描述符是使用QTBTC完成的，之后派生颜色量化器和位图图像。使用两种颜色量化器计算CHF（颜色直方图特征），该两种颜色量化器表示图像对比度和颜色分布，并且BPF（位图案比特）由位图图像构成，并且表示图像的边缘和纹理细节。然后使用矢量量化（VQ）表示这些属性以最终生成图像特征描述符。现在，图像特征用于形成称为Cooccurrence特征（CCF）和位模式特征（BPF）的图像的索引。 Visual Codebook用于构造图像的索引。支持向量机（SVM）通过将图像分解为子图像来用于图像分类。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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