首页> 外文会议>International Symposium on Intelligence Information Processing and Trusted Computing >An Automatic Incident of Freeway Detection Algorithm Based on Support Vector Machine
【24h】

An Automatic Incident of Freeway Detection Algorithm Based on Support Vector Machine

机译:基于支持向量机的高速公路检测算法自动事件

获取原文

摘要

Aimed at the research on freeway detection algorithm has great significance for improving efficiency and effectiveness of freeway traffic management, this paper based on the freeway traffic flow’s characteristics, in accordance with the incident detection’s basic principle, researches on freeway incident detection based on Support Vector Machine (SVM). This paper designs four different simulation experiments based on linearly non-separable SVM, Gauss kernel function and hyperbolic tangent function respectively. Experiments above verify the effectiveness and portability of algorithms. This paper adopts parameters optimization module of Libsvm tool box provided by the associate professor Chih-Jen Lin, after optimal parameters achieved, simulates the above experiments and compared with California algorithm, the simulation results show that choosing appropriate SVM model and kernel function, we can achieve better performances than California algorithm according to different experiments.
机译:针对高速公路检测算法的研究具有重要意义,可提高高速公路交通管理的效率和有效性,本文根据自高速公路交通流量的特点,按照事件检测的基本原理,基于支持向量机的高速公路事件检测研究(SVM)。本文根据线性不可分离的SVM,高斯内核功能和双曲线切线功能设计了四种不同的仿真实验。以上实验验证了算法的有效性和可移植性。本文采用Chih-jen Lin的副教授提供的Libsvm工具盒参数优化模块,在实现最佳参数之后,模拟了上述实验并与加州算法相比,仿真结果表明,选择合适的SVM模型和内核功能,我们可以根据不同的实验,实现比加利福尼亚州更好的表演。

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