[1]韩家明,杨忠,陈聪,等.无人机视觉导航着陆标识检测与分割方法[J].应用科技,2020,47(4):1-7,13.[doi:10.11991/yykj.202002013]
　HAN Jiaming,YANG Zhong,CHEN Cong,et al.Detection and segmentation method for the landmark based on UAV visual navigation[J].Applied science and technology,2020,47(4):1-7,13.[doi:10.11991/yykj.202002013]
点击复制

无人机视觉导航着陆标识检测与分割方法(/HTML)
分享到：

《应用科技》[ISSN:1009-671X/CN:23-1191/U]

卷:
第47卷
期数:
2020年4期
页码:
1-7,13
栏目:
智能科学与技术
出版日期:
2020-07-05

文章信息/Info

Title:
Detection and segmentation method for the landmark based on UAV visual navigation
作者:
韩家明1;  杨忠1;  陈聪1;  张秋雁2;  张驰1;  赖尚祥1;  李宏宸1;  方千慧1
1. 南京航空航天大学 自动化学院，江苏 南京 211106;
2. 贵州电网有限责任公司 电力科学研究院，贵州 贵阳 550002
Author(s):
HAN Jiaming1;  YANG Zhong1;  CHEN Cong1;  ZHANG Qiuyan2;  ZHANG Chi1;  LAI Shangxiang1;  LI Hongchen1;  FANG Qianhui1
1. College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China;
2. Electric Power Research Institute , Guizhou Power Grid Co., Ltd., Guiyang 550002, China
关键词:
无人机; 视觉导航; 着陆标识; 目标检测; YOLO; 特征增强; 网络裁剪; 图像分割
Keywords:
UAV; visual navigation; landmark; object detection; YOLO; feature enhancement; network slimming; image segmentation
分类号:
TP751
DOI:
10.11991/yykj.202002013
文献标志码:
A
摘要:
为实现基于视觉导航无人机自主着陆任务，提出一种无人机视觉导航着陆标识检测与分割方法。在Tiny-YOLO网络基础上融入自下而上的特征增强结构，得到Mark-YOLO网络。针对无人机硬件平台计算能力不足的问题，对网络模型进行裁剪，减少网络模型的参数量；对目标检测算法提取到的无人机着陆标识进行图像分割处理，获取降落标识的轮廓信息。实验结果表明：本文提出的Mark-YOLO算法具有更高的准确率；裁剪后的网络模型具有更少的参数量与更小的权重，且检测到的着陆标识通过图像分割方法处理后，可取得良好的分割效果。
Abstract:
To realize the autonomous landing task of UAV based on visual navigation, we propose a method for detection and segmentation of UAV visual navigation landmark. The Mark-YOLO network is obtained on the basis of Tiny-YOLO network and the bottom-up feature enhancement structure. Aiming at the problem of insufficient computing power of UAV platform, the network model is slimmed to reduce amount of the network parameters; the UAV landing mark extracted by the target detection algorithm is processed by image segmentation to obtain the contour information of the landmark. The experimental results show that the Mark-YOLO algorithm proposed in this paper has higher accuracy, the slimmed network model has less parameters and less weight, and the detected landmark could obtain outstanding segmentation results after being processed by image segmentation.

参考文献/References:

[1] 王东. 山地果园植保无人机自适应导航关键技术研究[D]. 咸阳: 西北农林科技大学, 2019.
[2] 丁为杰. 无人机自主电力巡线关键技术研究[D]. 济南: 山东大学, 2019.
[3] 黄莹, 尹文龙. 多旋翼无人机在救援领域的应用与发展[J]. 中国应急救援, 2019(5): 31–34
[4] 杨红琳. 无人机影像获取及其面向对象的土地利用分类应用[D]. 淮南: 安徽理工大学, 2019.
[5] FALANGA D, ZANCHETTIN A, SIMOVIC A, et al. Vision-based autonomous quadrotor landing on a moving platform[C]//2017 IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR). Shanghai, China, 2017: 200–207.
[6] 李靖, 马晓东, 陈怀民, 等. 无人机视觉导航着陆地标实时检测跟踪方法[J]. 西北工业大学学报, 2018, 36(2): 294–301
[7] JIN R, OWAIS H M, LIN D, et al. Ellipse proposal and convolutional neural network discriminant for autonomous landing marker detection[J]. Journal of field robotics, 2019, 36(1): 6–16.
[8] 刘全波, 侯永宏. 基于视觉的无人机自动着陆定位算法[J]. 电子产品世界, 2016, 23(7): 48–51
[9] S SARIPALLI, J F MONTGOMERY, G S SUKHATME. Visually guided landing of an unmanned aerial vehicle[J]. Transactions on robotics and automation, 2003, 19(3): 371–380.
[10] REDMON J, FARHADI A. YOLOv3: an incremental improvement[J]. arXiv: computer vision and pattern recognition, 2018.
[11] REZATOFIGHI H, TSOI N, GWAK J Y, et al. Generalized intersection over union: A metric and a loss for bounding box regression[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Long Beach, USA, 2019: 658–666.
[12] ZHANG Z, HE T, ZHANG H, et al. Bag of freebies for training object detection neural networks.[J]. arXiv: computer vision and pattern recognition, 2019.
[13] LIU Shu, QI Lu, QIN Haifang, et al. Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018: 8759–8768.
[14] RASHWAN A, KALRA A, POUPART P, et al. Matrix nets: a new deep architecture for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. California, USA, 2019.
[15] ZHANG P, ZHONG Y, LI X, et al. SlimYOLOv3: narrower, faster and better for real-Time UAV applications[J]. arXiv: computer vision and pattern recognition, 2019.
[16] NOBUYUKI OTSU. A threshold selection method from gray-level histograms[J]. IEEE transactions on systems man and cybernetics, 2007, 9(1): 62–66.
[17] CANNY J. A computational approach to edge detection[J]. IEEE transactions on pattern analysis and machine intelligence, 1986, 8(6): 679–698.
[18] 张秋雁, 杨忠, 陈聪, 等. 一种复杂环境下的电力线检测方法[J]. 应用科技, 2019, 46(5): 51–56
[19] 刘全波. 基于视觉的无人机自动着陆定位算法[D]. 天津: 天津大学, 2016.
[20] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Honolulu, USA, 2017: 7263–7271..

相似文献/References:

[1]杨成顺,杨忠,马俊,等.基于MS5540C的无人机高度传感器设计[J].应用科技,2010,37(08):26.[doi:10.3969/j.issn.1009-671X.2010.08.007]
　YANG Cheng-shun,YANG Zhong,MA Jun,et al.Design of UAV altimeter based on MS5540C[J].Applied science and technology,2010,37(4):26.[doi:10.3969/j.issn.1009-671X.2010.08.007]
[2]孙日明,赵龙.基于MapX的小型无人机地面站飞行控制软件[J].应用科技,2012,39(06):1.[doi:10.3969/j.issn.1009-671X.2012.201204005]
　SUN Riming,ZHAO Long.MapX-based ground control software for miniature UAV[J].Applied science and technology,2012,39(4):1.[doi:10.3969/j.issn.1009-671X.2012.201204005]
[3]刘鲁涛,王晓,李欣雨.基于视觉显著性的空中目标检测算法[J].应用科技,2019,46(01):88.[doi:10.11991/yykj.201811015]
　LIU Lutao,WANG Xiao,LI Xinyu.An aerial target detection algorithm based on visual significance[J].Applied science and technology,2019,46(4):88.[doi:10.11991/yykj.201811015]
[4]薛睿,韩璐.网络编码在无人机通信网络中的应用研究[J].应用科技,2019,46(06):25.[doi:10.11991/yykj.201812026]
　XUE Rui,HAN Lu.Application of network coding in UAV communication network[J].Applied science and technology,2019,46(4):25.[doi:10.11991/yykj.201812026]
[5]王赫男.基于激光测距无人机的输电线路覆冰厚度测量[J].应用科技,2019,46(06):46.[doi:10.11991/yykj.201812007]
　WANG Henan.Measurement for ice thickness of transmission line based on laser ranging drone[J].Applied science and technology,2019,46(4):46.[doi:10.11991/yykj.201812007]
[6]常乐,杨忠,张秋雁,等.悬挂负载空中机器人的抗摆控制[J].应用科技,2020,47(2):17.[doi:10.11991/yykj.201906003]
　CHANG Le,YANG Zhong,ZHANG Qiuyan,et al.Anti-swing control research of aerial robot with suspended load[J].Applied science and technology,2020,47(4):17.[doi:10.11991/yykj.201906003]
[7]赖尚祥,杨忠,韩家明,等.一种多信息级联聚类输电线检测方法[J].应用科技,2020,47(5):86.[doi:10.11991/yykj.202002010]
　LAI Shangxiang,YANG Zhong,HAN Jiaming,et al.A cascade clustering transmission line detection method based on multi-information[J].Applied science and technology,2020,47(4):86.[doi:10.11991/yykj.202002010]
[8]杨成顺,杨忠,薛八阳.基于人工标志的无人机视觉导航方法[J].应用科技,2015,42(05):34.[doi:10.11991/yykj.201505016]
　YANG Chengshun,YANG Zhong,XUE Bayang.An approach of vision navigation of UAV based on artificial marks[J].Applied science and technology,2015,42(4):34.[doi:10.11991/yykj.201505016]
[9]张天翼,杨忠,韩家明,等.基于连续自适应均值漂移和立体视觉的无人机目标跟踪方法[J].应用科技,2018,45(02):55.[doi:10.11991/yykj.201706012]
　ZHANG Tianyi,YANG Zhong,HAN Jiaming,et al.Approach of vision navigation of UAV based on continuously adaptive mean-shift and stereo vision[J].Applied science and technology,2018,45(4):55.[doi:10.11991/yykj.201706012]

备注/Memo

备注/Memo:
收稿日期:2020-02-29。
基金项目:国家自然科学基金项目(61473144)；中国南方电网有限责任公司科技项目(066600KK52170074)
作者简介:韩家明，男，博士研究生;杨忠，男，教授，博士生导师
通讯作者:杨忠，E-mail：Yangzhong@nuaa.edu.cn
更新日期/Last Update: 2020-11-27