本文将YOLOv3[1]目标检测算法与双目测距方法融合并应用在无人机上, 不仅能够检测出无人机前方是否有物体, 还能够对物体进行识别和分类, 同时测量与目标物体的距离。此算法首先通过训练好的YOLOv3tiny模型对前方的物体进行检测, 再用标定过的双目相机获取视角内的深度图, 对识别到的目标物体进行测距。为验证检测结果的有效性, 本文选用不同目标物体进行验证实验。实验结果表明, 该算法能够较为准确地识别目标物的类别, 从深度图中获取到的目标距离也较为精确。
Abstract
In the paper, the YOLOv3 target detection algorithm is integrated with binocular ranging method and applied to the UAV.It can not only detect whether there is an object in front of the UAV, but also identify and classify the object, and measure the distance of the target object.This algorithm first detects the object in front of it through the trained YOLOv3-tiny model, and then uses the calibrated binocular camera to obtain the depth map within the viewing angle and distance the identified target object.In order to verify the validity of the detection results, different target objects are selected for verification experiments.The experiment results show that the algorithm can identify the category of the target object more accurately, and the target distance obtained from the depth chart is also more accurate.
关键词
YOLOv3 /
目标检测识别 /
双目测距 /
无人机
Key words
YOLOv3 /
target detection and recognition /
binocular distance measurement /
UAV
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Redmon J, Divvala S, Girshick R, et al.You Only Look Once:Unified, Real-Time Object Detection[J].IEEE, 2016.
[2] Fang W, Wang L, Ren P.TinierYOLO:A Real-time Object Detection Method for Constrained Environments[J].IEEE Access, 2019, (99):1.
[3] Girshick R, Donahue J, Darrell T, et al.Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[J].IEEE Computer Society, 2013.
[4] Li Z, Zhou F.FSSD:Feature Fusion Single Shot Multibox Detector[J].2017.
[5] Duan K, Bai S, Xie L, et al.CenterNet:Keypoint Triplets for Object Detection[C]// International Conference on Computer Vision, 2019.
[6] Piao C, Wang X, Liu M.Pedestrian Detection Using Optimized YOLOv3 in UAV Scenario[C]//2019 International Conference on Intelligent Computing, Automation and Systems (ICICAS), 2019.
[7] Nguyen P, Arsalan M, Koo J, et al.LightDenseYOLO:A Fast and Accurate Marker Tracker for Autonomous UAV Landing by Visible Light Camera Sensor on Drone[J].Sensors, 2018, 18(6):1315.
[8] 倪文韬, 钟岱山, 唐向远.基于双目摄像头的测距系统的实现[J].无线互联科技, 2021, 18(2):3132.
[9] 李明东, 卢彪, 金传宇.基于双目视觉立体匹配技术的双目测距研究[J].廊坊师范学院学报(自然科学版), 2019, 19(2):1820.
