Processing Large-Scale Images on Ace16k Using Discrete Wavelet Transform

Öz

Abstract: Image processing has crucial effects in many fields like biomedical applications, traffic control, security, satellite systems and so on. Because of its improving importance, various methods are proposed for increasing computation speed and reliability. Cellular Neural Networks - Universal Machine (CNN-UM) is a promising hardware implementation for generating rapid results. In this study, we have implemented discrete wavelet transform (DWT) on input images in order to improve accuracy of edge detection applications on ACE16k which is one of the analog processors handling 128x128 images. Besides, DWT gave us an opportunity to process large-scale images. At the end of the study it is shown that DWT provides appreciable contribution to edge detection results. Keywords: CNNs, ACE16k, Discrete Wavelet Transform, Edge Detection, Iterative Annealing.

Anahtar Kelimeler

• -

Kaynakça

1. S. Madchakham, P. Thitimajshima, Y. Rangsanseri, “Edge Detection in Speckled SAR Images Using Wavelet Decomposition”, 22nd Asian Conference on Remote Sensing, pp. 1307 – 1310, 2001.
2. L. Y. Xue, J. J. Pan, “Edge detection combining wavelet transform and canny operator based on fusion rules”, International Conference on Wavelet Analysis and Pattern Recognition, ICWAPR 2009, pp. 324 – 328, 2009.
3. S. N. Yu, C. N. Lin, “An efficient paradigm for wavelet-based image processing using cellular neural networks“, International Journal Of Circuit Theory And Applications, Int. J. Circ. Theor. Appl., vol. 38, pp. 527 – 542, 2010.
4. L. O. Chua, L. Yang, “Cellular neural networks : Theory”, IEEE Trans. Circuits Syst., pp. 1257 – 1272, 1998.
5. L. O Chua, L. Yang, “Cellular neural networks: Applications”, IEEE Trans. Circuits Syst., pp. 1273 – 1290, 1998.
6. T. Roska, L. O. Chua, “The CNN Universal Machine: An Analogic Array Computer”, IEEE Transactions on Circuits and Systems- II: Analog and Digital Signal Processing, pp. 163 – 173, 1993.
7. S. Sevgen, E. Yucel, S. Arik, “Cellular Neural Networks Template Training System Using Iterative Annealing Optimization Technique on ACE16k Chip”, 16th International Conference on Neural Information Processing, ICONIP 2009, pp. 460 – 467, 2009.
8. D. Feiden, R. Tetzlaff, “Iterative annealing a new efficient optimization method for cellular neural networks”, Image Processing, pp. 549 – 552, 2001.

9. I. Daubechies, Ten Lectures on Wavelets, SIAM, Philadelphia, PA, 1992.
10. L. O. Chua, T. Roska, Cellular Neural Networks and Visiual Computing : Foundation and Applications, Cambridge University Press, 2002.
11. Bi-i Vision System : User Manual.
12. A. Zarandy, C. Rekeczky, “Bi-i: a standalone ultra high speed cellular vision system”, IEEE Circuit and Systems Magazine, pp. 36 – 45, 2005.
13. G. Linan, R. Domiguez-Castro, S. Espejo, A. RodrìguezVazquez, “ACE16k: A programmable focal plane vision processor with 128x128 resolution”, Eur. Conf. Circuit Theory and Design, pp. 345 – 348, 2001.