Abstract
In this work we explore the idea that, in the presence of a small training set of images, it could be beneficial to use that set itself to obtain a transformed training set (by performing a random rotation on each sample), train a source network using the transformed data, then retrain the source network using the original data. Applying this transfer learning technique to three different types of character data, we achieve average relative improvements between 6 % and 16 % in the classification test error. Furthermore, we show that it is possible to achieve relative improvements between 8 % and 42 % in cases where the amount of original training samples is very limited (30 samples per class), by introducing not just one rotation but several random rotations per sample.
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
Preview
Unable to display preview. Download preview PDF.
References
Bengio, Y.: Learning deep architectures for AI. Foundations and Trends in Machine Learning 2(1), 1–127 (2009)
Bengio, Y., Courville, A., Vincent, P.: Representation learning: A review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence 35(8), 1798–1828 (2013)
Ciresan, D., Meier, U., Gambardella, L., Schmidhuber, J.: Deep, big, simple neural nets for handwritten digit recognition. Neural Computation 22(12), 3207–3220 (2010)
Ciresan, D., Meier, U., Schmidhuber, J.: Transfer learning for Latin and Chinese characters with deep neural networks. In: International Joint Conference on Neural Networks (IJCNN), pp. 1–6 (2012)
Deng, L., Yu, D.: Deep learning for signal and information processing. Microsoft Research Monograph (2013)
Glorot, X., Bordes, A., Bengio, Y.: Domain adaptation for large-scale sentiment classification: a deep learning approach. In: International Conference on Machine Learning (ICML), pp. 513–520 (2011)
Hinton, G., Osindero, S., Teh, Y.-W.: A fast learning algorithm for deep belief nets. Neural Computation 18(7), 1527–1554 (2006)
Larochelle, H., Erhan, D., Courville, A., Bergstra, J., Bengio, Y.: An empirical evaluation of deep architectures on problems with many factors of variation. In: International Conference on Machine Learning (ICML), pp. 473–480 (2007)
Pan, S., Yang, Q.: A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering 22(10), 1345–1359 (2010)
Simard, P., Steinkraus, D., Platt, J.C.: Best practices for convolutional neural networks applied to visual document analysis. In: International Conference on Document Analysis and Recognition (ICDAR), vol. 3, pp. 958–962 (2003)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Amaral, T., Silva, L.M., Alexandre, L.A., Kandaswamy, C., de Sá, J.M., Santos, J.M. (2014). Transfer Learning Using Rotated Image Data to Improve Deep Neural Network Performance. In: Campilho, A., Kamel, M. (eds) Image Analysis and Recognition. ICIAR 2014. Lecture Notes in Computer Science(), vol 8814. Springer, Cham. https://doi.org/10.1007/978-3-319-11758-4_32
Download citation
DOI: https://doi.org/10.1007/978-3-319-11758-4_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-11757-7
Online ISBN: 978-3-319-11758-4
eBook Packages: Computer ScienceComputer Science (R0)