煤炭工程 ›› 2021, Vol. 53 ›› Issue (2): 99-105.doi: 10.11799/ce202102020

• 研究探讨 • 上一篇    下一篇

基于无人机倾斜摄影的抛掷爆破爆堆形态测量方法

孙健东,陈需,周宇,王韬,林钰淇,李方,于鑫鑫,刘鑫   

  1. 华北科技学院
  • 收稿日期:2020-09-04 修回日期:2020-10-09 出版日期:2021-02-20 发布日期:2021-05-10
  • 通讯作者: 孙健东 E-mail:sjd_xx@126.com

Method For Mapping Cast Blasting Stockpile Morphology Based On Drone Oblique Photography

  • Received:2020-09-04 Revised:2020-10-09 Online:2021-02-20 Published:2021-05-10

摘要: 露天矿抛掷爆破的爆堆形态对于爆破效果的评价以及后续剥离工艺的设计极为重要。当前现场采用的激光扫描手段存在设备笨重、程序复杂、耗时较长、架设位置受限、作业存在安全风险、存在扫描盲区等诸多问题。针对上述问题,研究经过现场反复试验,提出了基于无人机倾斜摄影的抛掷爆破爆堆形态测量方法。对于外业采集方案进行了详细的设计,并对内业数据处理的关键算法:基于sift算法的图像匹配、光束法区域网空三测量、基于PMVS的影像密集匹配进行了分析。研究结果表明,模型误差完全满足比例尺1∶500测绘成果I级精度要求,且无人机测绘技术与传统技术相比,具有测绘作业效率高、设备系统成本低、安全系数高、作业人员少、测绘结果精准等特点。

关键词: 露天矿山, 抛掷爆破, 爆堆, 无人机, 倾斜摄影

Abstract: The shape of cast blasting pile in open-pit mine is extremely important for the evaluation of the blasting effect and the design of the subsequent stripping process. The laser scanning methods currently used in the field have many problems such as the equipment is heavy, the procedures is complicated, long time-consuming, restricted installation positions, safety risks in operations, and scan blind spots. In response to the above problems, after repeated field trials, the research puts forward a method for measuring the shape of the blasting pile based on the tilt photography of the drone. A detailed design of the field acquisition program is carried out, and the key algorithms of the internal data processing are analyzed: image matching based on the sift algorithm, three measurement of the beam method in the area cyberspace, and dense image matching based on PMVS. The research results show that the model error fully meets the level I accuracy requirements of the scale 1:500 surveying and mapping results. Compared with the traditional technology, the UAV surveying and mapping technology has high surveying and mapping operation efficiency, low equipment system cost, high safety factor, and fewer operators. Features such as accurate surveying and mapping results.