Appeal 2007-0337
Application 09/996,200
image distortion method to enhance analysis of the transformation (Answer
10-11, 19). Appellant argues that the prior art does not disclose or suggest
extracting a component of a distortion by calculating an affine transform
from a plurality of points, much less decomposing the affine transform into a
translation and a linear transform matrix (Br. 13-14).
We will sustain the Examiner’s rejection of claims 4, 19, and 36. We
first note that Thomas reasonably discloses extracting a component of a
distortion by calculating an affine transform from a plurality of points for the
reasons previously discussed.7 Reyzin discloses an affine transformation of
an image to be concurrently rotated, scaled, or otherwise transformed. To
this end, Reyzin provides a sequence of one-dimensional affine
transformations along different axes that more efficiently executes general
affine transforms (Reyzin, abstract; col. 2, ll. 5-26). As the Examiner
indicates, Reyzin’s affine transformation utilizes a transformation matrix, M,
and an offset (x0, y0). Such a teaching, in our view, would have amply
suggested resolving an affine transformation into a linear transform matrix
and a translation as claimed. Moreover, Appellant has simply not rebutted
the Examiner’s rationale for combining Reyzin with Thomas – a position
that we find reasonable. The Examiner’s obviousness rejection of claims 4,
19, and 36 is therefore sustained.
We next consider the Examiner’s rejection of claims 5 and 6 under 35
U.S.C. § 103(a) as unpatentable over Thomas in view of Foley. Regarding
claim 5, the Examiner finds that Thomas discloses all claimed subject matter
except for the extraction of magnification comprising calculating the
determinant of a linear transform matrix. The Examiner cites Foley as
7 See P. 8, supra, of this opinion.
11
Page: Previous 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Next
Last modified: September 9, 2013