This Article 
 Bibliographic References 
 Add to: 
A QVGA-Size Pixel-Parallel Image Processor for 1,000-fps Vision
November/December 2009 (vol. 29 no. 6)
pp. 58-67
Takashi Komuro, University of Tokyo
Atsushi Iwashita, University of Tokyo
Masatoshi Ishikawa, University of Tokyo

Massively parallel image processing with as many processing elements as pixels can achieve real-time vision with rates as high as 1,000 frames per second. The authors implemented a QVGA-size pixel-parallel image processor for object identification and pose estimation on a single chip. Bit-serial operation and dynamic logic reduce the circuit area, and pipelining enables high processing speed.

1. A. Namiki, T. Komuro, and M. Ishikawa, "High Speed Sensory-Motor Fusion Based on Dynamics Matching," Proc. IEEE, vol. 90, no. 7, 2002, pp. 1178-1187.
2. N. Furukawa et al., "Dynamic Regrasping Using a High-Speed Multifingered Hand and a High-Speed Vision System," Proc. IEEE Int'l Conf. Robotics and Automation, IEEE Press, 2006, pp. 181-187.
3. Y. Watanabe et al., "Real-Time Visual Measurements Using High-Speed Vision," Proc. SPIE, vol. 5603, 2004, pp. 234-242.
4. H. Oku et al., "Two-Dimensional Tracking of a Motile Microorganism Allowing High-Resolution Observation with Various Imaging Techniques," Rev. Scientific Instruments, vol. 76, no. 3, 2005, article no. 034301.
5. M. Hu, "Visual Pattern Recognition by Moment Invariants," IRE Trans. Information Theory, vol. 8, no. 2, 1962, pp. 179-187.
6. J. Flusser, J. Boldys, and B. Zitova, "Moment Forms Invariant to Rotation and Blur in Arbitrary Number of Dimensions," IEEE Trans. Pattern Analysis and Machine Intelligence (PAMI), vol. 25, no. 2, 2003, pp. 234-246.
7. M. Elad, P. Milanfar, and G. Golub, "Shape from Moments—An Estimation Theory Perspective," IEEE Trans. Signal Processing, vol. 52, no. 7, 2004, pp. 1814-1829.
8. R. Mukundan, N. Malik, and K. Ramakrishnan, "Attitude Estimation Using Moment Invariants," Pattern Recognition Letters, vol. 14, no. 3, 1993, pp. 199-205.
9. O. Tahri and F. Chaumette, "Application of Moment Invariants to Visual Servoing," Proc. IEEE Int'l Conf. Robotics and Automation, IEEE Press, 2003, pp. 4276-4281.

Index Terms:
high-speed vision, massively parallel processor, smart image sensor, vision chip
Takashi Komuro, Atsushi Iwashita, Masatoshi Ishikawa, "A QVGA-Size Pixel-Parallel Image Processor for 1,000-fps Vision," IEEE Micro, vol. 29, no. 6, pp. 58-67, Nov.-Dec. 2009, doi:10.1109/MM.2009.89
Usage of this product signifies your acceptance of the Terms of Use.