Abstract
In the context of video processing, transmission to a remote server is not always possible nor suitable. Video processing on the edge could offer a solution. However, lower processing capacities constraint the number of techniques available for devices, in this work we report the performance of two techniques for classification from video on a minicomputer. The implementation of a real-time vehicle counting and classification system is evaluated through Support Vector Machine (SVM) and the Single Shot Detector Framework (SSD) in a minicomputer. We compare two SVM bases techniques, IPHOG and a MBF using Scale Invariant Features. The obtained results show that with a video resolution of 1280 \(\times \) 720 pixels and using SVM, precision and recognition rates of 86% and 94% are obtained respectively, while with SSD 93% and 67% rates are reached with times of processing higher than SVM.
Thanks to the Ecuadorian Corporation for the Development of Research and Academia, CEDIA, for the financing provided to research, through the CEPRA projects, especially the CEPRA project - XII -2018; Clasificador-video para actores de la movilidad como alternativa a conteos volumetricos manuales.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Coco - common objects in context. http://cocodataset.org/
Imagenet. http://www.image-net.org/
Opencv people counter, August 2018. https://www.pyimagesearch.com/2018/08/13/opencv-people-counter/
Chen, Z., Ellis, T., Velastin, S.A.: Vehicle type categorization: a comparison of classification schemes. In: 2011 14th International IEEE Conference on Intelligent Transportation Systems (ITSC), pp. 74–79, October 2011. https://doi.org/10.1109/ITSC.2011.6083075
Chen, Z., Ellis, T., Velastin, S.A.: Vehicle detection, tracking and classification in urban traffic. In: 2012 15th International IEEE Conference on Intelligent Transportation Systems, pp. 951–956, September 2012. https://doi.org/10.1109/ITSC.2012.6338852
chuanqi305: chuanqi305/mobilenet-ssd, October 2018. https://github.com/chuanqi305/MobileNet-SSD
Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 20(3), 273–297 (1995). https://doi.org/10.1007/BF00994018
Howard, A.G., et al.: Mobilenets: efficient convolutional neural networks for mobile vision applications. In: Computer Vision and Pattern Recognition, pp. 1–7, August 2017. https://doi.org/10.1109/CAIPT.2017.8555657
LeCun, Y., Cortes, C., Burges, C.: The mnist database. http://yann.lecun.com/exdb/mnist/
Lu, Z., Chan, K.S., Porta, T.L.: A computing platform for video crowdprocessing using deep learning. In: IEEE INFOCOM 2018 - IEEE Conference on Computer Communications, pp. 1430–1438, April 2018. https://doi.org/10.1109/INFOCOM.2018.8486406
Yadav, N., Binay, U.: Comparative study of object detection algorithms. Int. Res. J. Eng. Technol. (IRJET), 586–591 (2017). https://doi.org/10.23919/MVA.2017.7986913
Oh, S., Kim, M., Kim, D., Jeong, M., Lee, M.: Investigation on performance and energy efficiency of CNN-based object detection on embedded device. In: 2017 4th International Conference on Computer Applications and Information Processing Technology (CAIPT), pp. 1–4, August 2017. https://doi.org/10.1109/CAIPT.2017.8320657
Peña-González, R.H., Nuño-Maganda, M.A.: Computer vision based real-time vehicle tracking and classification system. In: 2014 IEEE 57th International Midwest Symposium on Circuits and Systems (MWSCAS), pp. 679–682, August 2014. https://doi.org/10.1109/MWSCAS.2014.6908506
Ren, X., Du, S., Zheng, Y.: Parallel RCNN: a deep learning method for people detection using RGB-D images. In: 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), pp. 1–6, October 2017. https://doi.org/10.1109/CISP-BMEI.2017.8302069
Roh, M., Lee, J.: Refining faster-RCNN for accurate object detection. In: 2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA), pp. 514–517, May 2017. https://doi.org/10.23919/MVA.2017.7986913
Ryus, P., et al.: NCHRP report 797. Guidebook on pedestrian and bicycle volume data collection. Transportation Research Board, Washington, DC (2014). ISSN: 0077–5614
Song, G.Y., Lee, K.Y., Lee, J.W.: Vehicle detection by edge-based candidate generation and appearance-based classification. In: 2008 IEEE Intelligent Vehicles Symposium, pp. 428–433, June 2008. https://doi.org/10.1109/IVS.2008.4621139
Torres-Espinoza, F., Barros-Gavilanes, G., Barros, M.J.: Computer vision classifier and platform for automatic counting: more than cars. In: Ecuador Technical Chapters Meeting 2017; 2nd ETCM Conference. pp. 1–6, October 2017. https://doi.org/10.1109/ETCM.2017.8247454, https://ieeexplore.ieee.org/document/8247454
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Heredia, A., Barros-Gavilanes, G. (2019). SVM and SSD for Classification of Mobility Actors on the Edge. In: Orjuela-Cañón, A., Figueroa-García, J., Arias-Londoño, J. (eds) Applications of Computational Intelligence. ColCACI 2019. Communications in Computer and Information Science, vol 1096. Springer, Cham. https://doi.org/10.1007/978-3-030-36211-9_1
Download citation
DOI: https://doi.org/10.1007/978-3-030-36211-9_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-36210-2
Online ISBN: 978-3-030-36211-9
eBook Packages: Computer ScienceComputer Science (R0)