Appeal 2007-1733
Application 09/978,275
resolution level. (Finding 6.) We note that to produce each detail image,
Vuylsteke iteratively uses a relative coefficient of the original image as well
as a corresponding function based on the resolution of the original image.
(Finding 7.) We further note that Vuylsteke’s pyramidal decomposition
captures the original signal in its entirety by preserving all the pixels with
their respective resolutions, as well as a residual image. (Id.) In light of
these findings, we agree with the Examiner that Vuylsteke’s production of
the detail images with their respective resolutions teaches the calculation of
a characteristic of the image signal that is based upon an exposure dose with
which the image had been produced.
Second, we note that the above independent claims also recite the
smoothing unit adapting a second characteristic of the smoothing filter to the
input image based on the first characteristic of the input image. (Br. Claim
Appendix.) We find that Vuylsteke does not reasonably teach this
limitation.
As detailed in the Findings of Fact section above, Vuylsteke teaches
that each of the detail images is fed into the noise filter unit to iteratively
attenuate them as a function of a locally estimated amount of relevant signal
content and in accordance with an estimated noise level. (Finding 7.) We
find that Vuylsteke’s teaching is limited to filtering the detail images.
Vuylsteke, however, does not explicitly teach using the detail images to
adjust the filter by adapting a characteristic of the filter to a value of the
calculated detail image.
It follows that the Examiner erred in rejecting independent claims 1,
7, 8, 9, 10, 11, 20, and 21 as being anticipated by Vuylsteke. We find for
9
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 Next
Last modified: September 9, 2013