• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Future generation computer systems >Research on pedestrian detection technology based on the SVM classifier trained by HOG and LTP features
【24h】

Research on pedestrian detection technology based on the SVM classifier trained by HOG and LTP features

机译：基于SVM分类器的人行语检测技术研究由HOG和LTP特征训练

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

In pedestrian detection, it is often affected by factors such as changes in the position of moving targets, changes in illumination, and complex background interference. Therefore, the traditional pedestrian feature detection method is affected by the complex environment, which will cause a certain deviation in the detection accuracy. This study combines hog and LTP (Local Trinary Patterns) feature detection technology to improve the efficiency of pedestrian detection. Constructing a pedestrian detection system based on SVM (Support Vector Machine) classifier trained by hog and LTP features, and constructs a pedestrian detection system according to the actual needs. Moreover, this paper designs a pedestrian detection experiment based on HOG (Histogram of Oriented Gradient) and LTP feature training SVM classifier. In addition, this paper uses the weighted fusion method to fuse the color map features with the depth map features, and finally uses the classifier to detect pedestrians. Experiments show that the new pedestrian detection system has a certain effect.
机译：在行人检测中，它通常受到因素的影响，例如移动目标的位置，照明变化和复杂背景干扰的变化。因此，传统的行人特征检测方法受复杂环境的影响，这将导致检测精度的某种偏差。本研究结合了HOG和LTP（本地杂志）特征检测技术来提高行人检测的效率。构建基于SVM（支持向量机）分类器的行人检测系统，由HOG和LTP特征训练，并根据实际需要构建行人检测系统。此外，本文设计了一种基于猪的行人检测实验（取向梯度直方图）和LTP特征训练SVM分类器。此外，本文使用加权融合方法熔断彩色地图功能，深度映射功能，最后使用分类器来检测行人。实验表明，新的行人检测系统具有一定的效果。

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