欢迎访问机床与液压官方网站!

咨询热线:020-32385312 32385313 RSS EMAIL-ALERT
基于SURF+RANSAC的双目立体视觉目标识别与抓取
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:


Binocular Stereo Vision Target Recognition and Capture Based on SURF+RANSAC
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对单目视觉处理简单、无法获取获取三维维空间数据的问题,提出一种双目立体视觉目标识别和抓取方法。结合SUR算法和RANSAC算法对图像进行匹配,以滤除图像中的“外点”;利用仿射变换获取关键点信息及形心坐标,根据立体重建原理实现目标测距与定位,最终完成抓取任务。实验结果表明:采用距离辅助坐标定位法可以准确识别目标物体并获得距离信息,实现准确定位。

    Abstract:

    Binocular stereo vision system can realize the target recognition, matching, positioning and ranging. The system included three modules: image matching, camera calibration and stereo ranging. SURF+RANSAC algorithm was used to match image and effectively filter “Outside point”. Then the key points and centroid coordinates were obtained by affine transformation. Finally, the targets distance and location were realized by the principle of stereo reconstruction and the task was finished. Experimental results show the method of distance aided coordinate positioning can be used to accurately identify the target object and to get the distance information and the accurate positioning is gotten.

    参考文献
    相似文献
    引证文献
引用本文

严亮,舒志兵,李照.基于SURF+RANSAC的双目立体视觉目标识别与抓取[J].机床与液压,2019,47(9):76-80.
. Binocular Stereo Vision Target Recognition and Capture Based on SURF+RANSAC[J]. Machine Tool & Hydraulics,2019,47(9):76-80

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2020-01-19
  • 出版日期: