基于Apriltag的室外抗干扰识别定位系统研究

doi:10.16180/j.cnki.issn1007-7820.2023.11.008

摘要/Abstract

摘要：

图像识别定位技术是实现工业自动化、智能化的关键。实际应用中,在室外干扰环境下目标经常出现被部分遮挡等情况,导致图像识别定位失效。针对室外干扰环境下图像识别定位失效问题,文中提出一种基于Apriltag的室外抗干扰识别定位系统,用以替代常规图像识别定位系统。将Apriltag特性与单目视觉原理结合,构建基于Apriltag的图像识别定位系统。利用Apriltag检测算法和基于SURF特征检测的遮挡目标识别算法实现全天候、抗干扰的室外目标识别和定位。采用基于Apriltag的定位靶标、工业相机和工控机等搭建实验系统,实验结果表明该系统在室外干扰环境下XY平面的定位误差分别为1.44 mm和1.36 mm,识别时间为90 ms。在重复实验下误差波动幅度小于0.5 mm,表明该识别定位系统具有高精度、快速性和稳定性。

关键词: 图像识别, Apriltag, 遮挡目标, 识别定位, SURF算法, 特征检测, 干扰环境, 单目视觉

Abstract:

Image recognition and positioning technology is the key to realize industrial automation and intelligence. However,in practical applications, the target is often partially occluded in the outdoor interference environment, which results in the invalidation of the image recognition and positioning system. To solve the problem of invalidation of image recognition and positioning system in outdoor interference environment, an outdoor anti-jamming recognition and positioning system based on Apriltag is proposed to replace the conventional system. Combining Apriltag feature with monocular vision principle, an image recognition and positioning system based on Apriltag is established. The Apriltag detection algorithm and the occluded target recognition algorithm based on SURF feature detection are used to achieve all-weather and anti-interference outdoor target recognition and positioning. The experimental system is established with the positioning target based on Apriltag, industrial camera and industrial computer. The experimental results show that the positioning errors of the XY plane are 1.44 mm and 1.36 mm, respectively, the recognition time is 90 ms, The error fluctuation under repeated experiments is less than 0.5 mm, which indicates that the recognition positioning system has high accuracy, rapidity and stability.

Key words: image recognition, Apriltag, occlusion targets, identify the location, SURF algorithm, feature detection, disturbing the environment, monocular vision

中图分类号:

TP249

代进洪,向振文,喻会福. 基于Apriltag的室外抗干扰识别定位系统研究[J]. 电子科技, 2023, 36(11): 47-55.

DAI Jinhong,XIANG Zhenwen,YU Huifu. Study on Outdoor Anti-Jamming Identification and Location System Based on Apriltag[J]. Electronic Science and Technology, 2023, 36(11): 47-55.

图/表 21

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

图15

图16

图17

图18

图19

表1

表2

参考文献 26

[1]	Mack J, Lenz C, Teutrine J, et al. High-precision 3D detection and reconstruction of grapes from laser rangedata for efficient phenotyping based on supervised learning[J]. Computers & Electronics in Agriculture, 2017, 32(135):300-311.
[2]	Xie Q, Remil O, Guo Y, et al. Object detection and tracking under occlusion for object-level RGB-D video segmentation[J]. IEEE Transactions on Multimedia, 2018, 20(3):580-592. doi: 10.1109/TMM.6046
[3]	Wang J, Olson E. Apriltag 2:Efficient and robust fiducial detection[C]. Daejeon: IEEE/RSJ International Conference on Intelligent Robots & Systems, 2016:4193-4198.
[4]	Jia P Y, Peng X D, Shen F F, et al. UAV's moving target recognition and tracking based on improved April tags algorithm[J]. Electronic Design Engineering, 2017, 25(17):31-35.
[5]	Chen J, Gao Y, Li S. Real-time April tag inertial fusion localization for large indoor navigation[C]. Shanghai: Chinese Automation Congress, 2020:6912-6916.
[6]	Farinha C, Santos T, Cunha-Vaz J, et al. OCT-Leakage mapping:A new automated method of OCT data analysis to identify and locate abnormal fluid in retinal edema[J]. Ophthalmology Retina, 2017, 1(6):486-496. doi: 10.1016/j.oret.2017.03.004
[7]	Tang G, Liu Z, Xiong J. Distinctive image features from illumination and scale invariant key points[J]. Multimedia Tools and Applications, 2019, 24(78):23415-23442.
[8]	Lightbody P, Krajnbo T, Hanheide M. A versatile high-performance visual fiducial marker detection system with scalable identity encoding[C]. Marrakech: The Symposium, 2017:276-282.
[9]	Lei J, Wu L, Li Y, et al. A novel FPGA-based architect-ture for fast automatic target detection in hyperspectral images[J]. Remote Sensing, 2019, 11(2):146-146. doi: 10.3390/rs11020146
[10]	Liang X, Chen G D, Zhao S R, et al. UAV moving target tracking method based on April tags[J]. Journal of Shenyang Aerospace University, 2019, 36(4):62-68.
[11]	Peng Y, Wang Y. Real-time forest smoke detection using hand-designed features and deep learning[J]. Computers and Electronics in Agriculture, 2019, 16(7):105029-105029.
[12]	Eilbrecht J, Stursberg O. Cooperative driving using a hierarchy of mixed-integer programming and trackingcontrol[C]. Redondo Beach: Intelligent Vehicles Sympos, 2017:673-678.
[13]	Wang H, Feng J, Bu Q, et al. Breast mass detection in digital mammogram based on gestalt psychology[J]. Journal of Healthcare Engineering, 2018, 5(2):1-13. doi: 10.1260/2040-2295.5.1.1
[14]	Schmitter D, Unser M. Landmark-based shape encodingand sparse-dictionary learning in the continuous domain[J]. IEEE Transactions on Image Processing, 2018, 27(1):365-378. doi: 10.1109/TIP.2017.2762582 pmid: 29028193
[15]	Wang L, Liu F, Xu J. A semantic segmentation and edge detection model based on edge information constraint training[J]. Journal of Physics Conference Series, 2020, 1518(1):12046-12046. doi: 10.1088/1742-6596/1518/1/012046
[16]	Tafti Ahamod P, Boghaie A, Holz Jessica D, et al. A comparative study on the application of SIFT,SURF,BRIEF and ORB for 3D surface reconstruction of electron microscopy images[J]. Computer Methods in Biomechanics & Biomedical Engineering Imaging & Visualization, 2018, 6(1):17-30.
[17]	Prinka, Wasson V. An efficient content based image re-trieval based on speeded up robust features with optimization technique[C]. Bangalore: IEEE International Conference on Recent Trends in Electronics, 2017:730-735.
[18]	Puyda V. Special processor for feature detection based on the SURF algorithm[J]. Computer Systems and Network, 2017, 88(1):129-134.
[19]	Vinay A. An efficient ORB based face recognition frame work for human-robot interaction-science direct[J]. Procedia Computer Science, 2018, 133(10):913-923. doi: 10.1016/j.procs.2018.07.095
[20]	杨云辉. 基于单目视觉的工件定位技术研究[J]. 电子科技, 2019, 32(12):72-75.
	Yang Yunhui. Research on workpiece location technology based on monocular vision[J]. Electronic Science and Technology, 2019, 32(12):72-75.
[21]	赵腾飞, 辛大欣, 华瑾. 改进SURF算法的特征提取与匹配方法研究[J]. 机械与电子, 2017, 35(9):77-80.
	Zhao Tengfei, Xin Daxin, Hua Jin. Research on feature extraction and matching method of improved SURF algorithm[J]. Machinery & Electronics, 2017, 35(9):77-80.
[22]	任云. 基于区域卷积神经网络的光学遥感图像部分遮挡目标检测识别技术研究[D]. 长沙: 国防科技大学, 2018:90-93.
	Ren Yun. Research on methods for partially occluded object detection in optical remote sensing images viar egions with CNN features[D]. Changsha: National University of Defense Technology, 2018:90-93.
[23]	陈敏, 汤晓安. SIFT与SURF特征提取算法在图像匹配中的应用对比研究[J]. 现代电子技术, 2018, 41(7):41-44.
	Chen Min, Tang Xiaoan. Comparison study on application of SIFT and SURF feature extraction algorithms in image matching[J]. Modern Electronics Technique, 2018, 41(7):41-44.
[24]	邱鹏, 赵和鹏, 朱长仁. 部分遮挡目标的稳健局部特征点提取方法[J]. 现代电子技术, 2013, 36(22):76-80.
	Qiu Peng, Zhao Hepeng, Zhu Changren. Extraction method of robust local feature points of partially-occluded object[J]. Modern Electronics Technique, 2013, 36(22):76-80.
[25]	顾伟, 李菲菲, 陈虬. 基于多特征融合的行人检测方法[J]. 电子科技, 2021, 34(5):29-34.
	Gu Wei, Li Feifei, Chen Qiu. Pedestrian detection algor-ithm based on multiple feature fusion[J]. Electronic S-cience and Technology, 2021, 34(5):29-34.
[26]	余俊鹏, 林洁鸿, 詹松辉, 等. 近景影像特征点匹配方法比较研究[J]. 广东工业大学学报, 2018, 35(4):56-60.
	Yu Junpeng, Lin Jiehong, Zhan Songhui, et al. A comparative ctudy of close-range image feature points matching methods[J]. Journal of Guangdong University of Technology, 2018, 35(4):56-60.

工况	X方向误差	Y方向误差	X最大波动	Y最大波动
工况	均值/mm	均值/mm	幅度/mm	幅度/mm
光照	1.80	1.15	0.44	0.39
淋雨	1.67	1.36	0.50	0.43
遮挡	2.11	2.70	0.48	0.48
多靶标	0.65	0.58	0.46	0.42
杂乱背景	0.41	0.55	0.26	0.27
反光背景	0.51	0.53	0.31	0.35