Title :
A new temporal domain optical flow measurement technique for focal plane VLSI implementation
Author :
Etienne-Cummings, R. ; Fernando, S. ; Takahashi, N. ; Shtonov, V. ; Van der Spiegel, J. ; Mueller, P.
Author_Institution :
Dept. of Electr. Eng., Univ. of Pennsylvania, Philadelphia, PA, USA
Abstract :
A new temporal domain technique for optical flow measurement is presented. This approach, which has been developed primarily for VLSI implementation, requires only the sign of spatiotemporal derivatives, 1-b Boolean multiplication and integer arithmetic to compute image velocity. Hence, it can be easily and efficiently implemented in hardware and software. It is composed of a hybrid of the Reichardt and Ullman-Marr motion models and measures both speed and direction at every pixel. For image sequences, it measures the number of frames required for an edge to translate over a pixel, while in analog hardware, it measures the time. Direction is given by correlating the disappearance of an edge at one pixel with its reappearance at a neighboring pixel. A test chip of this scheme has been implemented in 2 μm VLSI and was found to measure 2-D velocity over three orders of magnitude. A total of 0.1 mW of power was consumed per pixel in room light. Therefore, this technique offers simple, computationally efficient, and direct means for measuring wide spatiotemporal bandwidth image motion in both hardware and software
Keywords :
image sequences; 0.1 mW; 1-b Boolean multiplication; 2 micron; Reichardt motion model; Ullman-Marr motion models; analog hardware; direction; edge; focal plane VLSI implementation; frames; hardware; image sequences; image velocity; integer arithmetic; pixel; room light; software; speed; temporal domain optical flow measurement technique; test chip; wide spatiotemporal bandwidth image motion; Fluid flow measurement; Hardware; Image motion analysis; Measurement techniques; Motion measurement; Semiconductor device measurement; Spatiotemporal phenomena; Time measurement; Velocity measurement; Very large scale integration;
Conference_Titel :
Computer Architectures for Machine Perception, 1993. Proceedings
Conference_Location :
New Orleans, LA
Print_ISBN :
0-8186-5420-1
DOI :
10.1109/CAMP.1993.622478