Appeal No. 2006-3032 14
Application No. 09/969,040
We note that after the illumination flicker frequency is determined by
Smith’s initial test, normal image capture is performed at the appropriate
frame rate [Smith, col. 7, lines 23-38]. Because AC line frequency varies
from 60 Hz to 50 Hz according to country, a single brief test for flicker
frequency suffices. Therefore, we agree with appellants that “motion noise”
is irrelevant to Smith because Smith tests only once for the flicker frequency
using image data from a single frame or multiple frames. Because “motion
noise” is irrelevant to Smith’s brief test for initial lighting conditions, we find
that the motivation suggested by the examiner is misplaced.
Furthermore, we note that Smith’s system describes a method of
identifying and eliminating the effects of flicker noise without the use of
frame-to-frame comparisons [Smith, col. 6, lines 41-46]. While the
examiner correctly points out that one embodiment of Smith’s invention
captures multiple successive frames, we note that Smith explicitly discloses:
“horizontally averaged image data can be ‘stitched’ together to form a
continuous series of horizontally averaged image data” [Smith, col. 6, lines
41-46]. We find that pixel-by-pixel differencing between successive frames
(as performed by Oyama’s subtractor, col. 4, lines 5-7) is clearly
distinguished from Smith’s use of a continuous series of horizontally
Page: Previous 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Next
Last modified: November 3, 2007