Appeal No. 1997-1981
Application No. 08/298,547
Appellants argue that Hartnett does not suggest multiplying the input elements by
the scaled filter coefficients and then summing the multiplied inputs to get the output
elements. (See brief at page 6.) The examiner argues that appellants are arguing the
references individually rather than the combined teachings of the references.
(See answer at pages 5 and 6.) We agree with the examiner. The examiner relies on
Niehaus for this teaching. Niehaus clearly discloses that the Infinite Impulse Response (IIR)
filter and the Finite Impulse Response (FIR) filters use the weighted sums of present and
previous inputs. (See Niehaus at col. 2.) Since the filters of Hartnett and Nakamura are
FIR filters, the use thereof would similarly be a sum of weighted values. (See answer at
pages 6-7.)
Appellants argue that there is no suggestion in Hartnett or Nakamura that the scaled
coefficients are based on the close to ideal frequency response estimate. (See brief at
page 6.) The examiner relies on the disclosure of Hartnett at page 1768 to disclose the
use of the ideal filter frequency response. (See answer at page 6.) Also, Niehaus
discloses the use of various filters depending on the desired characteristics. Clearly these
filter characteristics must be derived from ideal or close to ideal frequency responses.
Furthermore, depending on the use of the filters, the skilled artisan would have been
motivated to use the ideal or close to ideal filter frequency response with image scaling in
5
Page: Previous 1 2 3 4 5 6 7 8 9 10 Next
Last modified: November 3, 2007