• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Journal of Advanced Robotic Systems >Computer vision-based techniques and path planning strategy in a slope monitoring system using unmanned aerial vehicle
【24h】

Computer vision-based techniques and path planning strategy in a slope monitoring system using unmanned aerial vehicle

机译：基于计算机视觉的技术和路径规划策略，使用无人驾驶系统利用无人驾驶系统

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Unmanned aerial vehicle is a typical field robot which can work in many unstructured environments like mines, forests, and even radiation areas. In our mine monitoring system built in a northeast province of China, special designed unmanned aerial vehicle is applied to take photos and perceive the environment. We select a series of image-based techniques to process aerial pictures to monitor the slope. The visual features are initially refined by histogram equalization. Then, the rocks and cracks can be detected by different digital image processing operators, like Canny, so as to assess displacements. Advanced semantic segmentation model, U-Net, is also selected to process the problem. Experimental results show that both Canny and U-Net can perceive the edges in pictures effectively, better than other operators. In addition, we model the inspection mission for mine slopes into a traveling salesman problem, then plan the path for unmanned aerial vehicle by swarm intelligence-based optimization.
机译：无人驾驶飞行器是一个典型的野外机器人，可以在许多非结构化环境中工作，如矿山，森林，甚至辐射区域。在我国矿山监控系统中建于中国东北省，特殊设计的无人机车辆应用于拍照并感知环境。我们选择一系列基于图像的技术来处理航空图片以监控斜率。可视化功能最初通过直方图均衡精制。然后，可以通过不同的数字图像处理操作员（如Canny）检测岩石和裂缝，以便评估位移。还选择高级语义分段模型，U-Net，以处理问题。实验结果表明，Canny和U-Net都可以有效地将边缘感知，而不是其他操作员。此外，我们模拟了矿山山坡的检验使命，进入旅行推销员问题，然后通过基于群体的智能的优化计划无人机车辆的路径。

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