Self-Learning of Feature Regions for Image Recognition

Satoru Yokota^1, , Jiang Li¹, Yuichi Ogishima¹, Hiromasa Kubo¹, Hakaru Tamukoh^2, and Masatoshi Sekine¹

¹Graduate School of Engineering, Tokyo University of Agriculture and Technology, Tokyo, Japan

²Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, Kitakyushu, Japan

Cite this paper:
Satoru Yokota, Jiang Li, Yuichi Ogishima, Hiromasa Kubo, Hakaru Tamukoh and Masatoshi Sekine. Self-Learning of Feature Regions for Image Recognition. Journal of Computer Sciences and Applications. 2015; 3(1):1-10. doi: 10.12691/jcsa-3-1-1

Abstract

Mobile systems are used in various environments. Thus, it is practical for image recognition systems to autonomously learn template images that are adaptive to objects in their various environments. However, learning the features of such objects requires large-scale computation and complex control. Hence, we propose an image recognition system that selects and learns regions that have a given object's features. This system is designed as a hardware/software (hw/sw) complex system with the multi-dimensional field programmable gate array (FPGA) “Vocalise.” This study discusses the possibility of dynamically building image databases and of real-time learning using the proposed image recognition system. Results indicate that the learning speed of the proposed method is estimated to be 1.4 × 10³ faster than that obtained with a conventional software method. This suggests the possibility of real-time learning.

Keywords:
autonomous learning feature region hw/sw complex system image recognition vocalise

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Figures

References:

[1]	P. Viola and M. Jones, “Robust real-time face detection,” in IEEE International Conference on Computer Vision, vol.2, p.747, 2001.
[2]	Y. Freund and R. E. Schapire, “A decision theoretic generalization of on-line learning and an application to boosting,” Journal of Computer and System Sciences, No. 1, Vol. 55, pp. 119-139, 1997.
[3]	J. P. Harvey, “Gpu acceleration of object classification algorithms using nvidia cuda,” Master's thesis, Rochester Institute of Technology, Rochester, NY, Sept. 2009.
[4]	Dalal, N.; Triggs, B., “Histograms of oriented gradients for human detection,” in IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol.1, no., pp.886, 2005.
[5]	Ma, X.; Najjar, W.A.; Roy-Chowdhury, A.K., “Evaluation and Acceleration of High-Throughput Fixed-Point Object Detection on FPGAs,” IEEE Transactions on Circuits and Systems for Video Technology, vol.PP, no.99, pp.1, 1, 2014.
[6]	K. Kudo, Y. Myokan, W. C. Than, S. Akimoto, T. Kanamaru, and M. Sekine, “Hardware object model and its application to the image processing,” IEICE Trans. Fund.
[7]	M. Yokokawa, I. Sudo, T. Yuno, M. Sekine, “Face detection with the union of hardware and software,” IEICE Tech. Rep, Vol.106, No.453, pp.13-18, Jan 2007.
[8]	Y. Usami, H. Kotaki, K. Takahashi, M. Sekine, “The voice recognition circuit by using hardware and software complex”, IEICE Tech. Rep, EA2007-112, pp.1-6, 2008.
[9]	Y. Ogishima, J. Li, S. Yokota, H. Kubo, M. Sekine, “Voice Recognition System using hw/sw Complex,” IEICE Tech. Rep, RECONF2014-43, vol.114, no.331, pp.51-56, Nov 2014.
[10]	H. Kubo, J. Li, S. Yokota, Y. Ogishima, M. Sekine, “Mobile robot system based on hw/sw Complex System using 3D FPGA-Array System “Vocalise””, IEICE Tech. Rep, RECONF2014-37, vol.114, no.331, pp.19-24, Nov 2014.
[11]	M. Sekine, T. Kanamaru, H. Ito, “Multi-level Matching for Detecting Faces”, J. IEICE, Vol.J86-A, No.9, pp.969-973, Sep 2003.
[12]	T. Yuno, I. Sudo, M. Yokokawa, R. Sato, K. Kudo, M. Sekine, “Self-Organizing Map Algorithm that used Base Vector”, IEICE Tech. Rep, Vol.106, No.428, pp.1-6, Dec 2006.
[13]	M. Ariizumi, B. Ogasawara, H. Tamukoh, M. Sekine, “An Image Recognition System with Hierarchical Feature Learning Function”, IEICE Tech. Rep, VLD2011-95, vol.111, no.397, pp.25-30, Jan 2012.
[14]	B. Ogasawara, S. Yokota, H. Tamukoh, M. Sekine, “Implementation of an Image Recognition System with Hierarchical Feature Learning Function,” IEICE Tech. Rep, RECONF2012-60, vol.112, no.325, pp.77-82, Nov 2012.
[15]	S. Yokota, B. Ogasawara, M. Sekine, “A Method for Learning Multi-resolutional Feature Regions”, Workshop on Circuits and Systems 26, pp.524-529, Jul 2013.
[16]	P.J. Phillips, H. Wechsler, J. Huang, P. Rauss, “The FERET database and evaluation procedure for face recognition algorithms, ” Image and Vision Computing J, Vol. 16, No. 5, pp. 295-306, 1998.
[17]	P.J. Phillips, H. Moon, S.A. Rizvi, P.J. Rauss, “The FERET Evaluation Methodology for Face Recognition Algorithms,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, pp. 1090-1104, 2000.
[18]	Hakaru Tamukoh, Kentaro Hanai, Ryosuke Kurogi, Soichiro Matsushita, Masashi Watanabe, Yuichi Kobayashi, and Masatoshi Sekine, “Internet Booster: A Networked Hw/Sw Complex System and Its Application to Hi-Performance WEB Application,” Proc. of World Automation Congress (WAC2010), 7th International Forum on Multimedia and Image Processing, 6 pages in CD-ROM, Sep., 2010. Kobe.
[19]	Xilinx, “SDAccel Development Environment,” Available: http://www.xilinx.c tools/sdx/sdaccel.html. [Accessed Dec. 26, 2014].
[20]	Hefenbrock, D.; Oberg, J.; Nhat Thanh; Kastner, R.; Baden, S.B., “Accelerating Viola-Jones Face Detection to FPGA-Level Using GPUs, ” 18th IEEE Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM), vol., no., pp.11,18, 2-4 May 2010.
[21]	Y. Atsumari, J. Li, H. Kubo, H. Tamukoh, M. Sekine, “A 3D FPGA-Array HPC System “Vocalise” and its Performance Evaluation, ” IEICE Tech. Rep, Vol.112, No.321, pp.201-206, Nov 2012.