Ex Parte PRAKASH et al

Patent Trial and Appeal Board

May 17, 2016

13401694 (P.T.A.B. May. 17, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
13/401,694 02/21/2012 Adityo PRAKASH
15610 7590 05/17/2016
Okamoto & Benedicto LLP
PO Box 641330
San Jose, CA 95164-1330
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www .uspto.gov
ATTORNEY DOCKET NO. CONFIRMATION NO.
10006.000621
(A01433C1Dl)
EXAMINER
5761
ROSARIO, DENNIS
ART UNIT PAPER NUMBER
2667
MAILDATE DELIVERY MODE
05/1712016 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Exparte ADITYO PRAKASH, EDWARD RATNER,
and DIMITRIOS ANTSOS
Appeal 2014-004119
Application 13/401,694
Technology Center 2600
Before JEAN R. HOMERE, JOHN A. EV ANS, and
DANIEL J. GALLIGAN, Administrative Patent Judges.
EV ANS, Administrative Patent Judge.
DECISION ON APPEAL 1
Appellants2 seek our review3 under 35 U.S.C. § 134(a) from the final
rejection of claims 4, 43, 44, 47-52, 55---60 and 62 as obvious. App. Br. 6.
Claims 1-3, 5--42, 45, 46, 53, 54, and 61 have been canceled. App. Br. 2.
We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.
1 Our Decision refers to Appellants' Appeal Brief filed Sept. 27, 2013
("App. Br."); Appellants' Reply Brief filed Feb. 11, 2014 ("Reply Br."); the
Examiner's Answer mailed Dec. 13, 2013 ("Ans."); and the Final Office
Action mailed June 28, 2013 ("Final Act.").
2 The real party in interest identified by Appellants is Altera Corporation.
App. Br. 2.
3 We have considered in this decision only those arguments the Appellants
actually raised in the Briefs. Any other arguments which the Appellants
could have made but chose not to make in the Briefs are deemed to be
waived. See 37 C.F.R. § 41.37(c)(l)(iv).
Appeal 2014-004119
Application 13/401,694
STATEMENT OF THE CASE
An understanding of the invention can be derived from a reading of
exemplary Claim 4, which is reproduced below with disputed limitations
italicized and some formatting added:
4. A transform coder device for processing a multi-dimensional
signal, the transform coder being configured to perform at least the
steps of:
obtaining the multi-dimensional signal comprising image data
for a digital image frame;
applying an inverse transform to the multi-dimensional signal,
the inverse transform interpolating parent data on a coarser scale to
predict child data on a finer scale;
using an interpolation filter with data-dependent filter
coefficients in performing said interpolating, wherein each filter
coefficient comprises a coefficient that is scaled by an inverse of a
gradient -valite, and vvherein the gradient -valite is compitted from a
center to a value in the parent data; and
adapting the interpolation filter to a pattern in the parent data.
References and Rejections
The Examiner relies upon the prior art as follows:
G. L. Anderson et al., Image Restoration Based on a Subjective
Criterion, IEEE Transactions on Sys's., Man, and Cybernetics Vol.
SMC-6, No. 12, 845 (1976). ("Anderson")
Jeong H. Shin et al., Image Fusion-Based Adaptive Regularization for
Image Expansion, Proc. SPIE Vol. 3974 1040 (2000). ("Shin II")
2
Appeal 2014-004119
Application 13/401,694
Jeong H. Shin et al., Adaptive Image Sequence Resolution
Enhancement Using Multiscale Decomposition Based Image Fusion,
Proc. SPIE Vol. 4067 1589 (2000). ("Shin I")
The claims stand rejected as follows:
1. Claims 4, 43, 44, 47-52, 55---60, and 62 stand rejected under 35 U.S.C.
§ 103(a) as being unpatentable over Shin I, Anderson, and Shin II. Final
Act. 3-9.
ANALYSIS
We have reviewed the Examiner's rejections in light of Appellants'
arguments that the Examiner has erred. We disagree with Appellants'
arguments.
Appellants argue Shin I does not teach "applying an inverse transform
to the multi-dimensional signal, the inverse transform interpolating parent
data on a coarser scale to predict child data on a finer scale," as recited in
Claim 4. App. Br. 7. Appellants contend the claimed inverse transform
requires "interpolating parent data on a coarser scale to predict child data on
a finer scale," as recited in the claim. Reply Br. 2. Appellants next argue
the claimed phrase, "to predict child data on a finer scale," limits the claim,
not indicating intended use. Id. at 2-3.
The Examiner finds Shin I teaches a regularized iterative image
interpolation algorithm that corresponds to the "inverse transform," as
recited in the claim. Ans. 2, 3, 5 (citing Shin I § 3). The Examiner further
finds Shin I teaches using low resolution image, "yi," as parent data on a
coarser scale to predict child data on a finer scale in the regularized iterative
image interpolation algorithm in equations (7) and (8). Ans. 3, 5, 7 (citing
Shin I § 3). Although the Examiner explains that the limitation, "to predict
3
Appeal 2014-004119
Application 13/401,694
child data on a finer scale," recited in the claim is intended use and not
required (Ans. 3, 6), the Examiner nevertheless finds Shin I teaches this
limitation. Final Act. 4 (citing Fig. 2 ). We agree with the Examiner that the
high resolution image N corresponding toxin equations (7) and (8) at least
suggests the "child data on a finer scale," recited in the claim. Ans. 3 (citing
Shin I § 3); Final Act. 4 (citing Shin I Fig. 2 (image NxN, x(m1,m2))).
Appellants contend Shin I teaches a "process which starts at the NxN
high resolution image and ends at the N/2xN/2 low resolution image." App.
Br. at 8 (citing Shin I Fig. 2). Specifically, Appellants then assert the NxN
high resolution image is not "child data on a finer scale," as recited in the
claim. Id.
The Examiner finds Shin I teaches a low resolution image processed from a
high resolution image from which the high resolution image is restored from
the processed low resolution image. Ans. 2-3, 7-8 (citing Shin I pp. 1591-
92, Fig. 2); see Shin I p.1590 ("This paper proposes an adaptive image
sequence resolution enhancement algorithm using multiscale decomposition
(MSD) based image fusion. Restored high resolution image frames with
high frequency components along the direction of edges can be obtained
from low resolution image frames every frames."). Specifically, the
Examiner finds Shin I teaches yin equation ( 4) as a low resolution image
indicated as y(n1,n2) in Figure 2. Ans. 3, 7 (citing Shin Ip. 1591, Fig. 2); see
Shin Ip. 1591 ("y ... represent[s] the lexicographically ordered low
resolution image."). The Examiner finds Shin I uses a regularized iterative
image interpolation algorithm represented by equations (7) and (8)
interpolating parent data on a coarser scale designated by y to calculate the
interpolated image x. Ans. 3 (citing Shin Ip. 1592). Shin I teaches
4
Appeal 2014-004119
Application 13/401,694
equations (7) and (8) are used in a first iteration; successive approximation
equation (12) is used for subsequent iterations. Ans. 7 (citing Shin Ip.
1592). The Examiner agrees with Appellants that Figure 2 in Shin I depicts
creating a low resolution image (N/2xN/2, y(n1,n2)) from a high resolution
image (NxN, x(m1,m2)). Ans. 8 (citing Shin I Fig. 2). However, we agree
with the Examiner that Shin I teaches both creating a low resolution image
from a high resolution image and restoring the high resolution image from
the low resolution image. Ans. 8 (citing Shin I§§ 3, 4.2). Shin I teaches the
interpolated image x as the high resolution image restored from
lexicographically ordered low resolution image y. Shin I§§ 2-3. Contrary
to Appellants' assertion, the Examiner finds, and we agree, the NxN high
resolution image correspond to "child data on a finer scale," as recited in the
claim. Ans. 3 (citing Shin I§ 3); Final Act. 4 (citing Shin I Fig. 2 (image
NxN, x(m1,m2))).
Appellants additionally argue Shin I teaches fusing two low-resolution
frames to restore image details. App. Br. 8 (citing Shin I p. 1590).
Appellants assert fusing two low-resolution frames is not an "inverse
transform," as recited in the claim. App. Br. 8. The Examiner finds the
multi-scale decomposition (MSD)-based image fusion in Shin I is combining
images as an inverse transform or inverse action of decomposing or
separating an image. Ans. 9. (citing Shin I Fig. 3); Final Act. 3--4 (citing
Shin I §§ 3, 4.2). We agree with the Examiner that Shin I teaches the
adaptive regularized image resolution enhancement algorithm to restore a
high resolution image as an inverse transform of a transform to degrade,
decompose, or separate an image into a low resolution image. Final Act.
5
Appeal 2014-004119
Application 13/401,694
3--4 (citing Shin I Fig. 3). Shin I teaches that the adaptive regularized image
resolution enhancement algorithm reverses the processes of DWT, DWP,
and Steerable Pyramid that degrade, decompose, or separate an image into a
low resolution image to result in a high resolution image. Shin I Fig. 3.
Claim 4 further recites:
using an interpolation filter with data-dependent filter coefficients in
performing said interpolating, wherein each filter coefficient
comprises a coefficient that is scaled by an inverse of a gradient value,
and wherein the gradient value is computed from a center to a value in
the parent data.
Appellants argue that Shin II does not teach this limitation. App. Br. 9.
Specifically, Appellants assert that Shin II teaches that yTy is not a gradient
value, but a diagonal matrix. App. Br. 9. Appellants further point out
scaling by a value is multiplying by a scalar value. Appellants additionally
assert multiplying by a matrix is distinctly different from multiplying by a
scalar value. Id.
The Examiner finds yTy is a gradient value because it derived from
the inverse of a function of the gradient magnitude, providing a detailed
explanation tracing through the equations in Shin II. Ans. 4--5, 11-12. In
particular, the Examiner finds Shin II teaches that yTy is replaced by v2(iJ).
Ans. 4, 10-11 (citing Shin II p.1045 (see equation 18, "where yTy
represents N2 x N2 diagonal matrix whose diagonal elements takes the value
of v2 (i,j).")). The Examiner finds Shin II teaches that v(iJ) is represented in
equation (17) as an inverse of a function of the gradient magnitude. Id.; see
Shin II (equation 17 showing inverse of function of "I grad f(i,j) I," which
"represents the gradient magnitude in (14)."). This finding in Shin II at least
suggests "a coefficient that is scaled by an inverse of a gradient value," as
6
Appeal 2014-004119
Application 13/401,694
recited in the claim. We agree with the Examiner that Shin II evaluates
VTV, a diagonal matrix, applicable to the data as a scalar value v2(iJ) for
each point (iJ) in the data in accordance with equation (17). Ans. 4, 12
(citing Shin II 1045). This teaching in Shin II at least suggests "wherein the
gradient value is computed from a center to a value in the parent data," as
recited in the claim.
In view of the foregoing, we are not persuaded that the Examiner has
erred in the rejection of Claim 4.
Appellants contend Claims 43, 44, and 47--49 are patentable in view
of their dependence from Claim 4. App. Br. 9. Appellants further assert
Claims 50-52 and 55-57 are patentable for the same reasons as Claim 4. Id.
Appellants additionally argue Claims 5 8---60 and 62 are patentable for the
same reasons as Claim 4. App. Br. 1 O; see also Reply Br. 3. In view of the
foregoing, we are not persuaded that the Examiner has erred in rejecting
Claims 43, 44, 47-52, 55-60, and 62.
DECISION
The rejection of Claims 4, 43, 44, 47-52, 55---60, and 62 is
AFFIRMED.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l). See 37 C.F.R.
§ 1.136(a)(l)(iv) (2013).
AFFIRMED
7