Scholarlink

<p>To realize the automatic adjustment of car seat based on feature matching algorithm, a detection and control system is designed with laser scanning. A human point cloud feature matching algorithm is proposed with stereo vision technology. According to feature analysis, an automatic seat adjustment model is established based on the human point cloud feature matching algorithm. In the experiment, the point cloud data of testers is compared and analyzed. The position test accuracy improves from 0.81&#xa0;cm to 0.62&#xa0;cm by the algorithm. In the convergence test, when the amount of data increases, the convergence time of traditional algorithms is 12&#xa0;s, and the convergence time of this algorithm is 8.61&#xa0;s. It increased by 3.93&#xa0;s. It proves that the algorithm is better solution accuracy and convergence speed in the process of matching and collecting human point cloud feature.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Human point cloud feature matching algorithm based on stereo vision

  • 期刊论文

  • 发表日期：

    2026年2月5日

查看全文
  • Xiayan Si,
  • Chunyan Liu,
  • Hui Xian,
  • Jianwei Wang,
  • Taiyuan Qu,
  • Jiapeng Li,
  • Xiaokun Ren,
  • Yuanyi Cha

摘要

To realize the automatic adjustment of car seat based on feature matching algorithm, a detection and control system is designed with laser scanning. A human point cloud feature matching algorithm is proposed with stereo vision technology. According to feature analysis, an automatic seat adjustment model is established based on the human point cloud feature matching algorithm. In the experiment, the point cloud data of testers is compared and analyzed. The position test accuracy improves from 0.81 cm to 0.62 cm by the algorithm. In the convergence test, when the amount of data increases, the convergence time of traditional algorithms is 12 s, and the convergence time of this algorithm is 8.61 s. It increased by 3.93 s. It proves that the algorithm is better solution accuracy and convergence speed in the process of matching and collecting human point cloud feature.