Pixel-and-column pipeline architecture for FFT-based image processor

Fast Fourier Transform based image-processing algorithms, such as Phase-Only Correlation, and Fourier-Mellin Transform are known to be time-consuming, but they are becoming more popular by the benefit of advances in the area of semiconductor technology. These algorithms need a series of pre-processing and postprocessing to the main processing of FFT. A pixel-and-column pipeline architecture, that has two-layered pipeline for a pixel and a column, is proposed. An LSI that provides those algorithms at the real-time response is designed based on the architecture. The processing time of 256/spl times/256 POC is 10.06 ms, and 2.57 times faster than the previous implementation.

[1]  John V. McCanny,et al.  A 64-point Fourier transform chip for video motion compensation using phase correlation , 1996, IEEE J. Solid State Circuits.

[2]  Tom Chen,et al.  COBRA: a 100-MOPS single-chip programmable and expandable FFT , 1999, IEEE Trans. Very Large Scale Integr. Syst..

[3]  Bevan M. Baas,et al.  A low-power, high-performance, 1024-point FFT processor , 1999, IEEE J. Solid State Circuits.

[4]  K. Kobayashi,et al.  An image processor implementing algorithms using characteristics of phase spectrum of two-dimensional Fourier transformation , 1999, ISIE '99. Proceedings of the IEEE International Symposium on Industrial Electronics (Cat. No.99TH8465).

[5]  George Wolberg,et al.  Robust image registration using log-polar transform , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[6]  Michel Defrise,et al.  Symmetric Phase-Only Matched Filtering of Fourier-Mellin Transforms for Image Registration and Recognition , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Thierry Pun,et al.  Rotation, scale and translation invariant digital image watermarking , 1997, Proceedings of International Conference on Image Processing.

[8]  Lewis Johnson,et al.  Conflict free memory addressing for dedicated FFT hardware , 1992 .