检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:顾兆军[1] 韩强[1,2] 王家亮 陈辉[1,2] 董楷 GU Zhaojun;HAN Qiang;WANG Jialiang;CHEN Hui;DONG Kai(Information Security Assessment Center,Civil Aviation University of China,Tianjin 300300,China;School of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China)
机构地区:[1]中国民航大学信息安全测评中心,天津300300 [2]中国民航大学计算机科学与技术学院,天津300300
出 处:《小型微型计算机系统》2024年第1期199-206,共8页Journal of Chinese Computer Systems
基 金:天津市教委科研计划项目(2020KJ026)资助;中国民航大学信息安全测评中心开放基金项目(ISECCA-202007)资助;民航安全能力建设基金项目(PESA2020100,PESA2021007,PESA2021009)资助.
摘 要:为解决四轴飞行器在执行目标检测与跟踪任务时无法判断机身与障碍物或跟踪目标之间安全距离问题,提出一种四轴飞行器单目视觉测距算法,算法融合基于帧间差分技术的水平测距模型与基于小孔成像的垂直测距模型.首先,算法可利用基于YOLOV4-Tiny的目标检测算法识别视频帧中的物体类别,并可在视频帧中框选出被测物体,依据现场环境选择最佳的测距方法进行测距;其次,系统终端执行测距方法,基于被测物体的视频帧成像位置与设定参数之间的差值控制四轴飞行器飞行,并记录飞行数据;最后,通过测距模型对飞行数据进行处理,计算出四轴飞行器在初始位置和最终位置与被测物体之间的距离.通过在特洛(Tello)飞行器平台上的实际飞行验证,实验结果表明所提出的四轴飞行器单目视觉测距算法可以有效分析出视频帧中各物体的景深,测距误差百分比均值为4.07%.在四轴飞行器与障碍物或跟踪目标之间距离较近,却无法评估此距离是否处于安全距离时,所提出的算法可为搭载单目摄像头的四轴飞行器提供可靠的测距技术支持.In order to solve the carrying of monocular camera small four axis in the process of the aircraft in flight could not judge the fuselage with obstacles or tracking problem of the safe distance between the target,put forward a kind of small four axis aircraft monocular vision ranging algorithm,this algorithm fusion technology based on frame difference between four axis craft level ranging model and four axis aircraft vertical distance measurement model based on camera.Firstly,objects in video frames are identified and selected by yoloV4-TINY algorithm.Secondly,the measured object is selected in the frame of the video,and the best ranging method is selected according to the field environment.Thirdly,the system terminal performs ranging method,controls the flight of the quadcopter based on the difference between the imaging position of the video frame of the measured object and the set parameters,and records the flight data;Finally,the distance between the initial position and the final position of the quadcopter and the measured object is calculated by processing the flight data through the ranging model.The experimental results show that the proposed monocular vision ranging algorithm based on template method can effectively analyze the depth of field of each object in the video frame,and the average range error percentage is 3.77%.When the distance between quad-rotor helicopter and obstacle or tracking target is relatively close,but it is impossible to evaluate whether the distance is in a safe distance,the proposed algorithm can provide reliable ranging technology support for quad-rotor helicopter with monocular camera.
分 类 号:TP391[自动化与计算机技术—计算机应用技术]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:3.145.68.176