Hyperparameters Tuning of Faster R-CNN Deep Learning Transfer for Persistent Object Detection in Radar Images



Modelo de detección de objetos, hiperparámetros, sobreajuste, Faster R-CNN, objetos persistentes


In previous work, a methodology was proposed to obtain a sea surface object detection model based on the FasterR-CNN architecture using Sperry Marine commercial navigation radar images. Unfortunately, the percentage of recall using the validation dataset was 75.76% with a minimum score for true positives of 7% due to a network overfitting problem. In this research, the overfitting problem is solved by comparing three experiments. Each experiment consists of the combinations of different hyperparameters within the Faster RCNN architecture. The main hyperparameters modified to improve the performance of the model were weights initialization and the optimizer. The results finally achieved, show a significant improvement in relation to the previous work. The improved persistent object detection model shows a recall of 93.94% with a minimum score for true positives of 98%.


Download data is not yet available.

Author Biographies

Rosa Gonzales-Martínez, Universidad de Piura

Rosa Gonzales-Martínez is a final year Ph.D. student in Engineering with mention in Automation,
Control and Process Optimization. Currently, she is a researcher in deep learning algorithms at the the Department of Mechanical-Electrical Engineering of the University of Piura, Peru. Has a Master’s Degree in Computer Science and Technology (2009) from the Carlos III University in Spain. She is a
Computer and Systems Engineer. 

Javier Machacuay, Universidad de Piura

Javier Machacuay is a last year Mechanical-Electrical Engineer student at the Universidad de
Piura. Currently, he is a researcher and developer of deep learning algorithms at the Electronics Laboratory of the university. He has experience in image processing and neural network programming aimed to solve agricultural and radar tasks.


Pedro Rotta, Universidad de Piura

Pedro Rotta is a researcher in artificial intelligence from the University of Piura. He obtained his degree of Electrical Mechanical Engineer at the University of Piura (2020). He has experience in artificial intelligence research and automatic control systems. His interests are: Deep Learning, Image Recognition,
Image detection, Robotics and automation.

César Chinguel, Universidad de Piura

César Chinguel has a doctorate in Industrial Engineering (1994) and a Master’s degree in MMF
(2006) from the University of Navarra in Spain. He is an Industrial Engineer from the University of Piura. He graduated in Automatic Control Systems from the Higher School of Industrial Engineers of San Sebastián, Spain (1987). He is currently a member of the Directorate of the Master and Doctorate Program in Engineering of the IME, Principal Professor of the Faculty of Engineering and founding member of the Institute of Family Sciences of the University of Piura, Peru. In 2015 he received the Medal of Naval Merit of Honor in the Degree of Knight of the Peruvian Navy for his contributions to naval technology for defense.


U. Kanjir, H. Greidanus, K. Oštir, "Vessel detection and classification from spaceborne optical images: A literature survey,Remote Sensing of Environment, Volume 207, 2018, pp. 1-26, ISSN 0034-4257, https://doi.org/10.1016/j.rse.2017.12.033.

F. Meyer, S. Hinz, A. Laika, D. Weihing, R. Bamler, "Performance analysis of the TerraSAR-X Traffic monitoring concept, ÏSPRS Journal of Photogrammetry and Remote Sensing, Volume 61, Issues 3–4, 2006, pp. 225-242, ISSN 0924-2716, https://doi.org/10.1016/j.isprsjprs.2006.08.002.

M. Petit, J. Stretta, H. Farrugio and A. Wadsworth, "Synthetic aperture radar imaging of sea surface life and fishing activities,ïn IEEE Transactions on Geoscience and Remote Sensing, vol. 30, no. 5, pp. 1085-1089, Sept. 1992, doi: 10.1109/36.175346.

A.K. Mazur, A.K. Wåhlin, A. Kr˛e˙zel, .An object-based SAR image iceberg detection algorithm applied to the Amundsen Sea,Remote Sensing of Environment, Volume 189, 2017, pp. 67-83, ISSN 0034-4257, https://doi.org/10.1016/j.rse.2016.11.013.

C. N. Koyama, H. Gokon, M. Jimbo, S. Koshimura, M. Sato, "Disaster debris estimation using high-resolution polarimetric stereo-SAR,ÏSPRS Journal of Photogrammetry and Remote Sensing, Volume 120, 2016, pp. 84-98, ISSN 0924-2716, https://doi.org/10.1016/j.isprsjprs.2016.08.003

S. Kingsley and S. Quegan, Ünderstanding Radar Systems,ïn SciTech Publishing, Inc., ed. 1992 originally, USA: McGraw-Hill, 1992, pp. 38- 42.

D. Purizaga-Céspedes, .Analysis of a New Two-Parameter Filter for Contact Detection in Marine Radar Images,"M.S. thesis, Department of Mechanical-Electrical Engineering, University of Piura, Perú, 2018.

S. H. Javadi, A. Farina, Radar networks: A review of features and challenges,Ïnformation Fusion, Volume 61, 2020, pp. 48-55, https://doi.org/10.1016/j.inffus.2020.03.005

D. Callaghan, J. Burger and A. Mishra, .A machine learning approach to radar sea clutter suppression,Çonference: 2017 IEEE Radar Conference (RadarConf17), pp. 1222-1227. DOI: 10.1109/RADAR.2017.7944391.

T. Zhang, X. Zhang, J. Shi, S. Wei, "HyperLi-Net: A hyper-light deep learning network for high-accurate and high-speed ship detection from synthetic aperture radar imagery,ÏSPRS Journal of Photogrammetry and Remote Sensing, Volume 167, 2020, pp. 123-153, September 2020.

N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), 2005, pp. 886-893 vol. 1. DOI: 10.1109/CVPR.2005.177.

T. Mita, T. Kaneko and O. Hori, "Joint Haar-like features for face detection,"Tenth IEEE International Conference on Computer Vision (ICCV’05) Volume 1, 2005, pp. 1619-1626 Vol. 2. DOI: 10.1109/ICCV.2005.129.

S. Yi-Kang and C. Ching-Te, "Local binary pattern orientation based face recognition,ÏCASSP 2015 - 2015 IEEE International Conference on Acoustics, Speech and Signal Processing, April 2015. DOI:10.1109/ICASSP.2015.7178138.

David G. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints", Computer Science Department, University of British Columbia, Vancouver, B.C., Canada International Journal of Computer Vision 60(2), 91–110, 2004.

W. An, C. Xie and X. Yuan, "An Improved Iterative Censoring Scheme for CFAR Ship Detection With SAR Imagery" in IEEE Transactions on Geoscience and Remote Sensing, vol. 52, no. 8, pp. 4585-4595, Aug. 2014. DOI: 10.1109/TGRS.2013.2282820.

D. Crisp. "The state-of-the-art in ship detection in Synthetic Aperture Radar imagery", Defence Science and Technology Organisation Salisbury (Australia) Info Sciences Lab, Edinburgh, Australia, Research Rep., Mayo, 2004. [Online]. Available: https://apps.dtic.mil/sti/pdfs/ADA426096.pdf

M. Kang, X. Leng, Z. Lin and K. Ji, "A modified faster R-CNN based on CFAR algorithm for SAR ship detection", International Workshop

on Remote Sensing with Intelligent Processing (RSIP), 2017, pp. 1-4. DOI: 10.1109/RSIP.2017.7958815.

Y. Lecun, L. Bottou, Y. Bengio and P. Haffner, "Gradient-based learning applied to document recognition", ïn Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, Nov. 1998. DOI: 10.1109/5.726791.

Y. LeCun, B. Yoshua, H. Geoffrey, "Deep learning" Nature 521, 436–444, 2015. DOI: https://doi.org/10.1038/nature14539.

Jayanth Koushik, " Understanding Convolutional Neural Networks" arXiv preprint, arXiv:1605.09081, 2016, https://arxiv.org/abs/1605.09081.

Y. Gui, X. Li and L. Xue, "A Multilayer Fusion Light-Head Detector for SAR Ship Detection", Sensors, 19(5), 1124, 2019. DOI:10.3390/s19051124

Y. Wang, Y. Zhang, H. Qu and Q. Tian, "Target Detection and Recognition Based on Convolutional Neural Network for SAR Image," 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2018, pp. 1-5.DOI: 10.1109/CISP-BMEI.2018.8633151.

J. Jiao, Y. Zhang, H. Sun, X. Yang, X. Gao, W. Hong, K. Fu, X. Sun, "A Densely Connected End-to-End Neural Network for Multiscale and

Multiscene SAR Ship Detection", IEEE Access, vol. 6, pp. 20881-20892, 2018. DOI: 10.1109/ACCESS.2018.2825376.

J. Redmon, S. Divvala, R. Girshick, A. Farhadi, "You Only Look Once: Unified, Real-Time Object Detection", 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779-788. DOI: 10.1109/CVPR.2016.91.

W. Liu, D. Anguelov, D. Erhan, "SSD: Single Shot MultiBox Detector", Proceedings of the Computer. Vision—ECCV 2016, Amsterdam, The Netherlands, 8–16 October 2016; pp. 21–37.

R. G. Martínez, J. M. Vera and C. C. Arrese, "Real-Time Detection Method of Persistent Objects in Radar Imagery with Deep Learning", 2020 IEEE Engineering International Research Conference (EIRCON), Lima, Peru, 2020, pp. 1-4. DOI: 10.1109/EIRCON51178.2020.9254021.

M. Dong, Y. Cui, X. Jing, X. Liu and J. Li, "End-to-End Target Detection and Classification with Data Augmentation in SAR Images," 2019 IEEE International Conference on Computational Electromagnetics (ICCEM), 2019, pp. 1-3. DOI: 10.1109/COMPEM.2019.8779096.

C. Shorten, T.M. Khoshgoftaar, .A survey on Image Data Augmentation for Deep Learning,"J Big Data 6, 60 (2019). https://doi.org/10.1186/s40537-019-0197-0

L. Taylor and G. Nitschke, "Improving Deep Learning with Generic Data Augmentation," 2018 IEEE Symposium Series on Computational Intelligence (SSCI), 2018, pp. 1542-1547. DOI: 10.1109/SSCI.2018.8628742.

Y. Wang, H. Zhang, G. Zhang, "cPSO-CNN: An efficient PSO-based algorithm for fine-tuning hyper-parameters of convolutional neural networks", Swarm and Evolutionary Computation. Volume 49.2019, pp.114-123, ISSN 2210-6502, https://doi.org/10.1016/j.swevo.2019.06.002.

Woo-Young Lee, Seung-Min Park, Kwee-Bo Sim, "Optimal hyperparameter tuning of convolutional neural networks based on the parametersetting-free harmony search algorithm", Optik, Volume 172, 2018, pp. 359-367, ISSN 0030-4026, https://doi.org/10.1016/j.ijleo.2018.07.044.

Muhammad Danial Siddiqi, Boyuan Jiang, Reza Asadi, Amelia Regan, "Hyperparameter Tuning to Optimize Implementations of Denoising Autoencoders for Imputation of Missing Spatio-temporal Data," Procedia Computer Science, Volume 184, 2021, pp. 107-114, ISSN 1877-0509, https://doi.org/10.1016/j.procs.2021.04.001.

R. Shaoqing, H. Kaiming, R. Girshick, J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks", IEEE

Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 1 June 2017. DOI: 10.1109/TPAMI.2016.2577031.

J. Zhang, M. Xing and G. Sun, "A Fast Target Detection Method for SAR Image Based on Electromagnetic Characteristics", 2018

China International SAR Symposium (CISS), 2018, pp. 1-3. DOI: 10.1109/SARS.2018.8552037.

Ross Girshick, "Fast R-CNN", Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2015, pp. 1440-1448.

M. Pham and S. Lefèvre, "Buried Object Detection from B-Scan Ground Penetrating Radar Data Using Faster-RCNN", IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, 2018, pp. 6804-6807. DOI: 10.1109/IGARSS.2018.8517683.

Bose, S. Rubin; Kumar, V. Sathiesh, "Efficient inception V2 based deep convolutional neural network for real-time hand action recognition", IET Image Processing, 2020, 14, (4), p. 688-696, DOI: 10.1049/ietipr. 2019.0985.

Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun, "Deep Residual Learning for Image Recognition", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770-778.

Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich, "Going deeper with convolutions," 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 1-9. DOI: 10.1109/CVPR.2015.7298594.

Frederik E.T. Schöller, Martin K. Plenge-Feidenhans’l, Jonathan D. Stets, Mogens Blanke, "Assessing Deep-learning Methods for Object Detection at Sea from LWIR Images", 2019 IFAC-PapersOnLine, Volume 52, Issue 21, pp. 64-71. DOI: https://doi.org/10.1016/j.ifacol.2019.12.284.

Tianwen Zhang, Xiaoling Zhang, "High-Speed Ship Detection in SAR Images Based on a Grid Convolutional Neural Network", Remote Sensing ( IF 4.509 ) Pub Date : 2019-05-21. DOI: 10.3390/rs11101206.

Feifei Hou, Wentai Lei, Shuai Li, Jingchun Xi, Mengdi Xu, Jiabin Luo, "Improved Mask R-CNN with distance guided intersection over union for GPR signature detection and segmentation, Automation in Construction", Volume 121, 2021, 103414, ISSN 0926-5805, https://doi.org/10.1016/j.autcon.2020.103414.

Yao, Yuan Rosasco, Lorenzo Caponnetto, Andrea, "On Early Stopping in Gradient Descent Learning. Constructive Approximation", 2017, 26. 289-315. 10.1007/s00365-006-0663-2.

You, Y.; Gitman, I.; Ginsburg, B, "Scaling SGD Batch Size to 32K for ImageNet Training", 2017. Available online: https://arxiv.org/abs/1708.03888v1 (accessed on 1 April 2019).

Chandra Kusuma Dewa, Afiahayati, "Suitable CNN Weight Initialization and Activation Function for Javanese Vowels Classification", Procedia Computer Science. Volume 144, 2018, Pages 124-132, ISSN 1877-0509, https://doi.org/10.1016/j.procs.2018.10.512.

Diederik P. Kingma, Jimmy Ba, "Adam: A Method for Stochastic Optimization", arXiv 2014, arXiv:1412.6980.

R. Padilla, W. L. Passos, T. L. B. Dias, S. L. Netto and E. A. B. da Silva, "A Comparative Analysis of Object Detection Metrics with a Companion Open-Source Toolkit", Electronics 2021, Vol 10(3), pp. 279. DOI: 10.3390/electronics10030279



How to Cite

Gonzales-Martínez, R., Machacuay, J., Rotta, P., & Chinguel, C. (2021). Hyperparameters Tuning of Faster R-CNN Deep Learning Transfer for Persistent Object Detection in Radar Images. IEEE Latin America Transactions, 20(4), 677–685. Retrieved from https://latamt.ieeer9.org/index.php/transactions/article/view/5892