舰载无人机的自主降落问题越来越受到重视，图像处理方法以其高精度、低成本的优势受到青睐。首先研究在图像中无人机提取算法，用最大类间方差法对提取后图像进行处理，然后基于霍夫直线检测算法，通过对处理后图像进行霍夫直线变换将图像空间的点变换到霍夫空间，从而找出图像空间的直线，进一步标示无人机位置。最后，通过试飞无人机实地测试检验该算法的准确性及稳定性。结果表明，该检测方法准确度高，速度快，在2～10 m范围内均有良好的检测精度，验证了基于图像处理识别定位无人机的可行性，为舰载无人机精准降落奠定了技术基础。
More and more attention has been paid to the autonomous landing of carrier-borne UAVs. The image processing method is favored because of its advantages of high accuracy and low cost. First studied the unmanned aerial vehicle extraction algorithm in the image, with the most between-cluster variance method to deal with after extraction of the image, and then based on Hough linear detection algorithm, based on the processed image Hoff linear transform of the image space transformation to Hoff space, so as to find out the image space of straight line, further labeled unmanned aerial vehicle position. Finally, the accuracy and stability of the algorithm are verified by the field test of flying UAV. The experimental results show that the detection method has high accuracy, fast speed and good detection accuracy in the range of two meters to ten meters, which verifies the feasibility of machine vision positioning UAV and lay the technical foundation for the precise landing of carrier-based UAVs.
2023,45(16): 75-79 收稿日期：2022-9-12
DOI：10.3404/j.issn.1672-7649.2023.16.015
分类号：V279+.2
基金项目：海军潜艇学院青年基金科研项目（HJQTXY20220216）
作者简介：张佰顺(1992-),男,硕士,讲师,研究方向为控制理论与控制工程、自动控制技术等
参考文献：
[1] 杨喻茸. 基于视觉信息引导的舰载无人机精确着舰技术研究[D]. 南京: 南京航空航天大学, 2018.
[2] 梁磊, 肖静, 邓扬晨. 舰载无人机着舰技术现状及发展趋势[J]. 西安航空学院学报, 2020, 38(5): 23–28
LIANG Lei, XIAO Jing, DENG Yangchen. Status quo and development trend of shipborne uav landing technology[J]. Journal of Xi'an Institute of Aeronautics, 2020, 38(5): 23–28
[3] 赵君爱. 计算机视觉技术在无人机着舰控制中的应用[J]. 舰船科学技术, 2022, 44(3): 165–168
ZHAO Junai. Application of computer vision technology in UAV landing control[J]. Ship Science and Technology, 2022, 44(3): 165–168
[4] 侯青, 颜鹏, 孙静, 等. 基于最优阈值的叶类中药边缘检测算法研究[J]. 信息技术, 2014(2): 69–72
HOU Qing, YAN Peng, SUN Jing, et al. Research on edge detection algorithm of leaf traditional Chinese medicine based on optimal threshold[J]. Information Technology, 2014(2): 69–72
[5] 孙欢欢, 程耀瑜, 冀钰. 改进OTSU算法和边缘检测的图像分割算法研究[J]. 山西电子技术, 2014(2): 82–84
SUN Huanhuan, CHENG Yaoyu, JI Yu. Research on image segmentation algorithm improved OTSU algorithm and edge detection[J]. Shanxi Electronic Technology, 2014(2): 82–84
[6] 王艳红. 基于OpenCV的运动目标检测与跟踪算法的研究[D]. 杭州: 杭州电子科技大学, 2014.
[7] 马晓静. 无人机自主任务系统的研发与实现[D]. 济南: 山东大学, 2017.
