...
首页> 外文期刊>International Journal of Innovative Computing Information and Control >HUMAN DETECTION USING COLOR CONTRAST-BASED HISTOGRAMS OF ORIENTED GRADIENTS
【24h】

HUMAN DETECTION USING COLOR CONTRAST-BASED HISTOGRAMS OF ORIENTED GRADIENTS

机译:使用基于颜色的面向梯度的基于颜色对比的直方图的人体检测

获取原文
获取原文并翻译 | 示例
           

摘要

In this paper, we propose a method for human detection using color contrast-based Histograms of Oriented Gradients (HOG). The proposed method calculates the color similarities between a pixel of interest and the pixels in its neighborhood. By applying the same gradient calculations as HOG on these color similarities, gradient orientation histogram can be made for color contrast. It can capture edge information derived from color contrast even when the luminance contrast is small. We evaluate the proposed method using the following three types of classifiers in the INRIA and NICTA datasets: Real Adaboost, Support Vector Machine (SVM), and random forest. As a result, the proposed method exhibits higher performance than HOG.
机译:在本文中,我们提出了一种使用基于颜色的面向梯度(HOG)的彩色对比度直方图的人体检测方法。该方法计算感兴趣像素与其邻域的像素之间的颜色相似度。通过将相同的梯度计算应用于这些颜色相似性上,可以对颜色对比度进行梯度方向直方图。即使亮度对比度小,它也可以捕获从颜色对比的边缘信息。我们使用inria和nicta数据集中的以下三种类型的分类器评估所提出的方法:真正的Adaboost,支持向量机(SVM)和随机森林。结果,该方法表现出比HOG更高的性能。

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