基于YOLOv5的角度优化抓取检测算法研究

doi:10.3778/j.issn.1002-8331.2210-0499

摘要/Abstract

摘要： 针对目前机器人抓取检测方法抓取角度预测过于离散,抓取过程可能产生较大偏角，降低抓取检测精度，甚至导致抓取失败的问题，提出了一种基于YOLOv5神经网络模型改进的机器人实时抓取检测方法。以单阶段目标检测模型YOLOv5为基础,提取抓取框坐标及抓取角度。对抓取角度进行更细致的划分，同时引入环形平滑标签以适应角度的周期性，建立相邻角度之间的联系，将YOLOv5检测头进行解耦，并对损失函数进行优化，提高检测精度。在Cornell数据集上进行实验验证。实验结果表明，与经典的抓取检测方法相比，所提算法能够更好地预测抓取角度，提升抓取检测精度；在Cornell数据集上，此模型达到了97.5%的准确率以及71?FPS的检测速度。

关键词: 抓取检测, YOLOv5, 环形平滑标签, 检测头解耦

Abstract: Aiming at the problems that the current robot grasping detection method is too discrete in predicting the grasping angle and the grasping process may produce large off-angle, which reduces the grasping detection accuracy and even leads to grasping failure, an improved robot real-time grasping detection method based on the YOLOv5 neural network model is proposed. Firstly, the grasping frame coordinates and grasping angles are extracted based on the single-stage object detection model YOLOv5. Afterwards, the grasping angles are divided more carefully, while circular smoothing label is introduced to accommodate the periodicity of the angles, links between adjacent angles are established, the YOLOv5 detection head is decoupled, and the loss function is optimized to improve the detection accuracy. Finally, an experimental validation is performed on the Cornell dataset. The experimental results show that the proposed algorithm can better predict the grasping angle and improve the grasping detection accuracy compared with the classical grasping detection methods. The model achieves 97.5% accuracy and 71?FPS detection speed on the Cornell dataset.

Key words: grasp detection, YOLOv5, circular smooth label, decoupled head

陈春朝, 孙东红. 基于YOLOv5的角度优化抓取检测算法研究[J]. 计算机工程与应用, 2024, 60(6): 172-179.

CHEN Chunchao, SUN Donghong. Research on Angle-Optimised Grasp Detection Algorithm Based on YOLOv5[J]. Computer Engineering and Applications, 2024, 60(6): 172-179.

参考文献

[1] MALLICK A, DELPOBIL A P, CERVERA E. Deep learning based object recognition for robot picking task[C]//Proceedings of the 12th International Conference on Ubiquitous Information Management and Communication (IMCOM2018), New York, USA, 2018: 1-9.
[2] JIANG Y, MOSESON S, SAXENA A. Efficient grasping from RGBD images: learning using a new rectangle representation[C]//IEEE International Conference on Robotics and Automation, 2011: 9-13.
[3] 张志康, 魏赟. 基于语义分割的两阶段抓取检测算法[J/OL]. 计算机集成制造系统: 1-15[2022-12-11]. http://kns.cnki.net/kcms/detail/11.5946.TP.20220517.1009.008.html.
ZHANG Z K, WEI Y. Two-stage grasp detection algorithm based on semantic segmentation network[J/OL].Computer Integrated Manufacturing Systems:1-15[2022-12-11]. http://kns.cnki.net/kcms/detail/11.5946.TP.20220517.1009.008.html.
[4] 陈丹, 林清泉. 基于级联式Faster RCNN的三维目标最优抓取方法研究[J]. 仪器仪表学报, 2019, 40(4): 229-237.
CHEN D, LIN Q Q. Research on 3D object optimal grasping method based on cascaded Faster RCNN[J]. Chinese Journal of Scientific Instrument, 2019, 40(4): 229-237.
[5] 孙先涛, 程伟, 陈文杰, 等. 基于深度学习的视觉检测及抓取方法[J]. 北京航空航天大学学报, 2023, 49(10)：2635-2644.
SUN X T, CHENG W, CHEN W J, et al. A visual detection and grasping method based on deep learning[J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49(10)：2635-2644.
[6] CHU F J, XU R, PATRICIO V. Real-world multi-object, multi-grasp detection[J]. IEEE Robotics and Automation Letters, 2018, 3(4): 3355 -3362.
[7] 夏浩宇, 索双富, 王洋, 等. 基于Keypoint RCNN改进模型的物体抓取检测算法[J]. 仪器仪表学报, 2021, 42(4): 236-246.
XIA H Y, SUO S F, WANG Y, et al. Object grasping detection algorithm based on keypoint RCNN improved model[J]. Chinese Journal of Scientific Instrument, 2021, 42(4): 236-246.
[8] JOCHER G.YOLOv5[EB/OL].(2020?06?17)[2022?08?16]. https://github.com/ultralytics/YOLOv5.
[9] YANG X, YAN J. Arbitrary-oriented object detection with circular smooth label[C]//European Conference on Computer Vision. Cham: Springer, 2020: 677-694.
[10] REDMON J, DIVVAL S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2016: 779-788.
[11] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2017: 7263-7271.
[12] REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. (2018-04-08) [2022-08-16]. https://arxiv.org/abs/1804.02767 .
[13] WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[EB/OL]. (2022-07-06)[2022-08-16].https://arxiv.org/abs/2207.02696 .
[14] 熊军林, 赵铎. 基于RGB图像的二阶段机器人抓取位置检测方法[J]. 中国科学技术大学学报, 2020, 50(1): 1-10.
XIONG J L, ZHAO D. Two-stage grasping detection for robots based on RGB images[J]. Journal of University of Science and Technology of China, 2020, 50(1): 1-10.
[15] TEKIN B, SINHA S N, FUA P. Real-time seamless single shot 6d object pose prediction[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2018: 292-301.
[16] LENZ I, LEE H, SAXENA A. Deep learning for detecting robotic grasps[J]. The International Journal of Robotics Research, 2015, 34(4/5): 705-724.
[17] REDMON J, ANGELOVA A. Real-time grasp detection using convolutional neural networks[J]. arXiv:1412.3128, 2014.
[18] GE Z, LIU S, WANG F, et al. Yolox: exceeding yolo series in 2021[EB/OL]. (2021?08?16)[2022?08?16]. https://arxiv.org/abs/2107.08430.
[19] REZATOFIGHI H, TSOI N, GWAK J Y, et al. Generalized intersection over union: a metric and a loss for bounding box regression[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 658-666.
[20] XIONG R, YANG Y, HE D, et al. On layer normalization in the transformer architecture[C]//International Conference on Machine Learning, 2020: 10524-10533.
[21] LOSHCHILOV I, HUTTER F. SGDR: stochastic gradient descent with warm restarts[J]. arXiv:1608.03983, 2016.
[22] GUO D, SUN F, LIU H, et al. A hybrid deep architecture for robotic grasp detection[C]//2017 IEEE International Conference on Robotics and Automation, 2017: 1609-1614.