Abstract
This paper presents a new convolutional neural network (CNN) architecture which improve performance of occluded-object classification by adding an assistant unit. The classification architecture is OCC-VGG19. The OCC-VGG19 outputs two parts classification information. First information is occluded state of target, and second information is the objectness information of target. To access the performance of the proposed architecture, we generate a new dataset that referred to as OCC-CIFAR10 based on CIFAR-10. The OCC-CIFAR10 include 40,000 original images and 10,000 generated image that are occluded by noise, and the OCC-CIFAR10 samples are RGB color images with size 32 × 32. The OCC-CIFAR10 is used in both of training and testing step. Experimental results show that the proposed assistant unit enhance network robustness in occluded-objects classification task.
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
Hearst, M.A.: Support vector machines. IEEE Intell. Syst. 13, 18–28 (1998)
LeCun, Y.: LeNet-5, convolutional neural networks (2015). http://yann.lecun.com/exdb/lenet
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556, September 2014
Szegedyet, C., et al.: Going deeper with convolutions. In: Proceedings of IEEE Conference Computer Vision Pattern Recognition, pp. 1–9 (2015)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Proceedings of Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)
Russakovsky, O., et al.: ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115(3), 211–252 (2015)
Ciresan, D.C., Meier, U., Masci, J., Schmidhuber, J.: Multi-column deep neural network for traffic sign classification. Neural Netw. 32, 333–338 (2012)
Azulay, A., Weiss, Y.: Why do deep convolutional networks generalize so poorly to small image transformations? arXiv preprint arXiv:1805.12177, abs/1805.12177 (2018)
Krizhevsky, A.: Learning Multiple layers of features from tiny images (2009)
Wang, J., Luis, P.: The effectiveness of data augmentation in image classification using deep learning. Stanford University research report (2017)
Acknowledgement
This research was supported by the MSIT (Ministry of Science and ICT), Korea, under the ICT Consilience Creative program (IITP-2019-2016-0-00318) supervised by the IITP (Institute for Information & communications Technology Planning & Evaluation).
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
Tang, Q., Lee, Y., Jo, K. (2019). Occluded Object Classification with Assistant Unit. In: Huang, DS., Huang, ZK., Hussain, A. (eds) Intelligent Computing Methodologies. ICIC 2019. Lecture Notes in Computer Science(), vol 11645. Springer, Cham. https://doi.org/10.1007/978-3-030-26766-7_69
Download citation
DOI: https://doi.org/10.1007/978-3-030-26766-7_69
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-26765-0
Online ISBN: 978-3-030-26766-7
eBook Packages: Computer ScienceComputer Science (R0)