Intelligent Tennis Robot Based on a Deep Neural Network
Source abstract
Abstract
In this paper, an improved you only look once (YOLOv3) algorithm is proposed to make the detection effect better and improve the performance of a tennis ball detection robot. The depth-separable convolution network is combined with the original YOLOv3 and the residual block is added to extract the features of the object. The feature map output by the residual block is merged with the target detection layer through the shortcut layer to improve the network structure of YOLOv3. Both the original model and the improved model are trained by the same tennis ball data set. The results show that the recall is improved from 67.70% to 75.41% and the precision is 88.33%, which outperforms the original 77.18%. The recognition speed of the model is increased by half and the weight is reduced by half after training. All these features provide a great convenience for the application of the deep neural network in embedded devices. Our goal is that the robot is capable of picking up more tennis balls as soon as possible. Inspired by the maximum clique problem (MCP), the pointer network (Ptr-Net) and backtracking algorithm (BA) are utilized to make the robot find the place with the highest concentration of tennis balls. According to the training results, when the number of tennis balls is less than 45, the accuracy of determining the concentration of tennis balls can be as high as 80%.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Neural Network Applications · Anomaly Detection Techniques and Applications · Video Analysis and Summarization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, China
Department of Electronic and Electrical Engineering, The University of Sheffield, Sheffield S10 2TN, UK
Center for Artificial Intelligence, University of Technology, Sydney 2007, Australia
Funding
National Natural Science Foundation of China
References (24)
Show source-provided references
1. Gu, S., Chen, X., Zeng, W., and Wang, X. (2018, January 9–12). A Deep Learning Tennis Ball Collection Robot and the Implementation on NVIDIA Jetson TX1 Board. Proceedings of the 2018 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM), Auckland, New Zealand.10.1109/AIM.2018.8452263 ↗
2. Face Recognition using Histograms of Oriented Gradients10.1016/j.patrec.2011.01.004 ↗
3. Facial Feature Detection using Haar Classifiers
4. Krizhevsky, A., Sutskever, I., and Hinton, G. (2012, January 3–6). ImageNet Classification with Deep Convolutional Neural Networks. Proceedings of the 25th International Conference on Neural Information Processing Systems, Lake Tahoe, NV, USA.
5. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23–28). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.81 ↗
6. Purkait, P., Zhao, C., and Zach, C. (2018, January 3–6). SPP-Net: Deep Absolute Pose Regression with Synthetic Views. Proceedings of the British Machine Vision Conference (BMVC 2018), Newcastle, UK.
7. Girshick, R. (2015). Fast R-CNN. arXiv.10.1109/ICCV.2015.169 ↗
8. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks10.1109/TPAMI.2016.2577031 ↗
9. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2015). You Only Look Once: Unified, Real-Time Object Detection. arXiv.10.1109/CVPR.2016.91 ↗
10. Redmon, J., and Farhadi, A. (2017, January 21–26). YOLO9000: Better, Faster, Stronger. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.690 ↗
11. Redmon, J., and Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv.
12. Xu, J.X., Lin, T.C., Yu, T.C., Tai, T., and Chang, P. (2018, January 10–12). Acoustic Scene Classification Using Reduced MobileNet Architecture. Proceedings of the 2018 IEEE International Symposium on Multimedia (ISM), Taichung, Taiwan.10.1109/ISM.2018.00038 ↗
13. Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). Mobilenets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv.
14. Simonyan, K., and Zisserman, A. (2014). Very Deep Convolutional Networks for Large-scale Image Recognition. arXiv.
15. Szegedy, C., Ioffe, S., Vanhoucke, V., and Alemi, A. (2016). Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning. arXiv.10.1609/aaai.v31i1.11231 ↗
16. An Exact Algorithm for the Maximum Clique Problem10.1016/0167-6377(90)90057-C ↗
17. Bello, I., Pham, H., Le, Q.V., Norouzi, M., and Bengio, S. (2016). Neural Combinatorial Optimization with Reinforcement Learning. arXiv.
18. Vinyals, O., Fortunato, M., and Jaitly, N. (2015). Pointer Networks. arXiv.
19. Gu, S., and Hao, T. (2018, January 29–31). A Pointer Network Based Deep Learning Algorithm for 0-1 Knapsack Problem. Proceedings of the Tenth International Conference on Advanced Computational Intelligence (ICACI), Xiamen, China.10.1109/ICACI.2018.8377505 ↗
20. Gu, S., Hao, T., and Yang, S. (2018, January 25–28). The Implementation of a Pointer Network Model for Traveling Salesman Problem on a Xilinx PYNQ Board. Proceedings of the International Symposium on Neural Networks, Minsk, Belarus.10.1007/978-3-319-92537-0_16 ↗
21. Backtracking Algorithms
22. Attention Solves Your TSP, Approximately
23. Keneshloo, Y., Tian, S., Reddy, C.K., and Ramakrishnan, N. (2018). Deep Reinforcement Learning For Sequence to Sequence Models. arXiv.10.1109/TNNLS.2019.2929141 ↗
24. Bay, A., and Sengupta, B. (2017). StackSeq2Seq: Dual Encoder Seq2seq Recurrent Networks. arXiv.