Ex Parte Sebastian et al

Patent Trial and Appeal Board

Dec 19, 2016

13673052 (P.T.A.B. Dec. 19, 2016)

United States Patent and Trademark Office
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
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
13/673,052 11/09/2012 Thomas Baby Sebastian 261548/71966-0138 3857
91753 7590 12/21/2016
General Electric Company
Global Patent Operation-Aviation
901 Main Avenue
3rd Floor
Norwalk, CT 06851
EXAMINER
NIU, FENG
ART UNIT PAPER NUMBER
2669
NOTIFICATION DATE DELIVERY MODE
12/21/2016 ELECTRONIC
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.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
gpo.mail@ge.com
lori.E.rooney@ge.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte THOMAS BABY SEBASTIAN, ERIC DANIEL BUEHLER,
BENJAMIN THOMAS OCCHIPINTI, and
KONRAD ROBERT KUCZYNSKI
Appeal 2015-007042
Application 13/673,052
Technology Center 2600
Before THU A. DANG, NATHAN A. ENGELS, and ALEX S. YAP,
Administrative Patent Judges.
DANG, Administrative Patent Judge.
DECISION ON APPEAL
I. STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134(a) from a non-final rejection
of claims 1, 2, and 4—10. Claim 3 has been canceled. We have jurisdiction
under 35 U.S.C. § 6(b).
We AFFIRM.
A. INVENTION
According to Appellants, the invention relates to “a method of
removing stationary objects from at least one hyperspectral image” (Spec.
13).
Appeal 2015-007042
Application 13/673,052
B. ILLUSTRATIVE CLAIM
Claim 1 is exemplary:
1. A method of compressing hyperspectral imagery, the
method comprising:
collecting a series of hyperspectral images of a target
scene;
determining and storing a first hyperspectral image to
represent a background of the target scene from the series of
hyperspectral images of the target scene;
collecting a series of current hyperspectral images of the
target scene;
determining a second hyperspectral image having
moving or new objects in the target scene from the series of
current hyperspectral images of the target scene;
determining a value of each spectral band for each pixel
of the first and second hyperspectral images to establish a
signature for each pixel;
iterating through each corresponding pair of pixels while
comparing one signature to another;
determining a dissimilarity measure value for each
comparison;
comparing the dissimilarity measure value to a
predetermined threshold, and if the dissimilarity measure value
comparison is within the predetermined threshold then setting
the value of the pixel in the second hyperspectral image to zero
to create a compressed background-subtracted hyperspectral
image; and
storing the compressed background-subtracted
hyperspectral image wherein the stored compressed
background-subtracted hyperspectral image comprises only
stored pixels of the moving or new objects in the target scene
while preserving the values of all spectral bands in the stored
pixels.
2
Appeal 2015-007042
Application 13/673,052
C. REJECTIONS
1. Claims 4 and 9 stand rejected under 35U.S.C. § 112, first paragraph,
as failing to comply with the written description requirement.
2. Claims 1, 2, 4, 7, and 10 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over Auty (US 5,809,161, iss Sept. 15, 1998), Oyaizu
(US 2011/0243451 Al, pub. Oct. 6, 2011), and Banerjee (US 2011/0322480,
pub. Dec. 23, 2010).1
3. Claim 5 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Auty, Oyaizu, Banerjee, and Tsuchikawa (US 5,748,775,
iss. May 5, 1998).
4. Claim 6 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Auty, Oyaizu, Banerjee, and Jain (US 5,745,126, iss. Apr.
28, 1998).
5. Claim 8 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Auty, Oyaizu, Banerjee, and Jin (US 5,937,102, iss. Aug.
10, 1999).
6. Claim 9 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Auty, Oyaizu, Banerjee, and Brill (US 6,937,651, iss.
Aug. 30, 2005).
1 Page 9 of the the Non-Final Action includes a heading identifying claims
1,2, and 7 as being rejection over Auty, Oyaizu, and Banerjee, but the
heading does not list claims 4 and 10 (Non-Final Act. 9). The Non-Final
Action nevertheless includes substantive rejections of claims 4 and 10 with
the rejections of claims 1, 2, and 7 and based on the same references (Non-
Final Act. 14—15). Accordingly, claims 4 and 10 also stand rejected over
Auty, Oyaizu, and Banerjee.
3
Appeal 2015-007042
Application 13/673,052
II. ISSUE
The principal issue before us is whether the Examiner erred in finding
that the combination of Auty, Oyaizu, and Banerjee teaches or suggests a
method of compressing “hyperspectral” imagery, comprising “determining
and storing a first hyperspectral image to represent a background of the
target scene;” “determining a second hyperspectral image having moving or
new objects in the target scene;” “determining a value for each spectral band
for each pixel... to establish a signature for each pixel;” and “iterating
through each corresponding pair of pixels while comparing one signature to
another” (claim 1).
III. FINDINGS OF FACT
The following Findings of Fact (FF) are shown by a preponderance of
the evidence.
Auty
1. Auty discloses monitoring movement of an object (Abst.), which
detects a moving object as different from a still background (col. 4,11. 58—
66), wherein the background image representing a static background is
determined and stored (col. 9,11. 35—37). In particular, a background
difference board subtracts a background image from the current or live
image to produce a preliminary difference image (col. 9,11. 24—29).
Oyaizu
2. Oyaizu discloses storing a reference background image, detecting an
object from an input image and estimating the rough position and shape of
the detected object, generating a background difference image between the
input image and the reference background image, and calculating a
4
Appeal 2015-007042
Application 13/673,052
relationship equation of pixel values between pixels corresponding to the
background difference image and the reference background image (Abstr.).
Banerjee
3. Banerjee discloses tagging and tracking of objects using
hyperspectral video sensors to exploit the objects’ unique signatures, which
includes examining image pixels and computing how closely a pixel’s
spectrum matches a known object spectral signature (Abstr.). In particular,
Banerjee detects and discriminates an object from the surrounding objects
and background by examining every image pixel and computing how closely
each pixel’s spectrum matches a known object spectral signature (19).
IV. ANALYSIS
35 U.S.C. § 112, first paragraph
The Examiner rejects claims 4 and 9 under 35 U.S.C. § 112, first
paragraph as failing to comply with the written description requirement
(Non-Final Act. 7). In particular, the Examiner finds the following newly
added limitation of claim 4 lacks supporting description in the Specification:
“if the absolute difference between an averaged value of the hyperspectral
signatures of the first hyperspectral image and an averaged value of the
hyperspectral signatures of the second hyperspectral image is less than a
predetermined threshold value, the value of the pixel in the second
hyperspectral image is set to zero” (id.). According to the Examiner, “there
is no disclosure in the specification about multiple signatures at each pixel,”
or “setting the value of the pixel in the second image to zero” (Non-Final
Act. 7—8). The Examiner similarly finds the following newly added
limitation of claim 9 lacks supporting description in the Specification:
5
Appeal 2015-007042
Application 13/673,052
“determining the second hyperspectral image is done by comparing the first
hyperspectral image to the series of current hyperspectral images” (Non-
Final Act. 8).
In response, Appellants contend that “Claim 4 finds support in
[0024] and [0027]” of the Specification (App. Br. 11), wherein “the
processor may average the multiple hyperspectral images ... to create a
single background image” and “the processor may perform the subtraction
... by setting the resulting pixel value to zero if the absolute difference
between the signature of the background image pixel and the signature of the
corresponding pixel of the hyperspectral image is less than a predetermined
threshold value” (id.). According to Appellants, “one of ordinary skill
would read H [0024] and [0027] with an understanding that Appellants had
possession of and were the inventors of the invention” (id.). Appellants then
contend “Claim 9 finds support | [0025]” of the Specification “which
discusses aspects of determining the background image” (id.).
We do not find Appellants’ contention persuasive. Here, although
Appellants point to “support in || [0024] and [0027]” of the Specification
(App. Br. 11) for claim 4, we agree with the Examiner that the claimed
“averaged value of the hyperspectral signatures’ ’ is not supported in |
[0024], which merely describes an “average [of the] multiple hyperspectral
images’ ’ (Ans. 16 (emphasis added)). Similarly, we agree with the Examiner
that the claim recites “setting ‘the value of the pixel in the second
hyperspectral image ’ to zero rather than ‘setting the resulting pixel value to
zero,” as stated in | [0027] of the Specification (Ans. 16 (emphasis added)).
We also agree with the Examiner that there is no support in | [0025] of the
Specification for the amended features of “determining the second
6
Appeal 2015-007042
Application 13/673,052
hyperspectral image” of claim 9, since the paragraph merely “refers to the
process of selecting the first hyperspectral image” (Ans. 16—17).
For the above reasons, we find that the preponderance of evidence
supports the Examiner's determination that the claims are not adequately
supported by a written description under 35 U.S.C. § 112, first paragraph.
35 U.S.C. § 103(a)
As for independent claim 1, Appellants contend that “[t]he question
here is whether Auty is analogous . . . art” since “processing monochrome
images is fundamentally different than processing hyperspectral images”
(App. Br. 14). That is, Appellants contend “Auty is not from the same field
of endeavor” (id. at 15) andis not reasonably pertinent to the problem faced
by the inventors” (id. at 16). Similarly, although Appellants concede Oyaizu
discloses color imagery instead of monochrome as in Auty (id. at 20),
Appellants also contend “Oyaizu is non analogous art” (id. at 19). Although
Appellants concede Banerjee discloses “pixel-by-pixel comparison of
hyperspectral signatures” (id. at 20), Appellants contend there is no support
for the Examiner’s conclusion “that combining Bannerjee [sic] with Auty
will yield predictable results” (id. at 21).
We have considered all of Appellants’ arguments and evidence
presented. However, we disagree with Appellants’ contentions regarding the
Examiner’s rejections of the claims. We agree with the Examiner’s findings,
and find no error with the Examiner’s combination of the references and the
conclusion that the claims would have been obvious over the combined
teachings.
Although Appellants contend Auty and Oyaizu are nonanalogous, the
nonanalogous art test considers the threshold question whether a prior art
7
Appeal 2015-007042
Application 13/673,052
reference is “too remote to be treated as prior art.” In re Clay, 966 F.2d 656,
658 (Fed. Cir. 1992) (quoting In re Sovish, 769 F.2d 738, 741 (Fed. Cir.
1985)). A reference is analogous art to the claimed invention if: (1) “the
reference is from the same field of endeavor as the claimed invention (even
if it addresses a different problem)”; or (2) “the reference is reasonably
pertinent to the problem faced by the inventor (even if it is not in the same
field of endeavor as the claimed invention)” (In re Bigio, 381 F.3d 1320,
1325 (Fed. Cir. 2004)).
We agree with the Examiner that Auty and Oyaizu are both “in the
field of applicant’s endeavor (image processing)” and also use “the very
technique Appellants] [are] using, i.e., background subtraction” (Ans. 19-
20). In particular, like Appellants’ invention, Auty discloses monitoring
movement of an object by detecting a moving object as different from the
background (FF 1). Similarly, Oyaizu also detects an object from the
background (FF 2). Thus, we agree that both Auty and Oyaizu are from the
same field of endeavor as the claimed invention (image processing) and are
reasonably pertinent to the problem faced by the inventor (detecting a
moving object from the background). See Bigio, 381 F.3dat 1325.
Thus, we find no error with the Examiner’s finding that Auty and
Oyaizu as analogous art, as the references are not “too remote to be treated
as prior art.” See Clay, 966 F.2d at 658 (quoting In re Sovish, 769 F.2d 738,
741 (Fed. Cir. 1985)).
Auty discloses monitoring movement of an object using background
subtraction, which includes detecting a moving object as different from a
stored static background, and then subtracting a background image from the
current or live image to produce a preliminary difference image (FF 1). We
8
Appeal 2015-007042
Application 13/673,052
agree with the Examiner’s reliance on Auty to disclose the steps of
determining and storing an image to represent a background of the target
scene, and then determining an image having moving or new objects in the
target scene (Non-Final Act. 9—12).
In the same field of endeavor of image processing using background
subtraction, Oyaizu discloses storing a reference background image,
detecting an object from an input image, and generating a background
difference image (FF 2). That is, similar to Auty, Oyaizu also discloses
determining and storing an image to represent a background of the target
scene and then determining an image having moving or new objects in the
target scene (id.).
Furthermore, we agree with the Examiner that Auty discloses pixel-
by-pixel subtraction (Ans. 6; FF 1). Similarly, Oyaizu discloses calculating
a relationship equation of pixel values between pixels corresponding to the
background difference image and the reference background image (FF 2).
We agree with the Examiner that such pixel-by-pixel subtraction comprises
“comparing” the pixels, wherein the “difference value at each pixel [can be]
interpreted as the ‘dissimilarity measure value for each comparison’” (Ans.
6, emphasis omitted). Thus, we find no error with the Examiner’s reliance
on Auty and Oyaizu for disclosing and suggesting determining a value for
each pixel and iterating through each corresponding pair of pixels while
comparing one to another (Ans. 7—8). Accordingly, we agree with the
Examiner that Auty in view of Oyaizu discloses and suggests all of the
claimed limitations except that the determined image having the
moving/new objects is not a “hyperspectral” image, wherein the value for
9
Appeal 2015-007042
Application 13/673,052
each pixel is not a value “for each spectral band” to establish a signature for
each pixel (id.).
However, Banerjee discloses tracking/monitoring of objects using
hyperspectral video sensors to exploit the objects’ unique signatures (FF 3).
In particular, Banerjee detects and discriminates the object from the
surrounding objects and background by examining every image pixel and
computing how closely each pixel’s spectrum matches a known object
spectral signature (id.). Accordingly, we find no error with the Examiner’s
reliance on Banerjee for teaching and suggesting determining a value for
each spectral band for each pixel and iterating through each pixel while
comparing the signatures (id.). That is, we find no error with the Examiner’s
reliance on Banerjee for at least suggesting a hyperspectral image may be
determined as an image having the moving/new objects, wherein the value
for each spectral band for each pixel is determined to establish a signature
for each pixel. Accordingly, we agree with the Examiner that the
combination of Auty, Oyaizu, and Banerjee teaches and suggests the
contested limitations of claim 1.
We are guided by the Supreme Court’s guidance regarding
modification of prior art teachings by a practitioner in the art: an improved
product in the art is obvious if that “product [is] not [one] of innovation but
of ordinary skill and common sense” (KSR Int’l Co. v. Teleflex Inc., 550
U.S. 398, 421 (2007)). The skilled artisan is “[a] person of ordinary
creativity, not an automaton” (id.).
Here, we agree with the Examiner that Banerjee “provides a solution
(hyperspectral) to the very ‘variability’ problem that ‘compromises the
performance of many vision systems, including background subtraction
10
Appeal 2015-007042
Application 13/673,052
methods’” (Ans. 22). We conclude that a person of ordinary skill in the art,
upon reading Banerjee’s teaching that a hyperspectral image may be
determined as an image having the moving/new objects, would have found it
obvious to modify Auty and Oyaizu’s image processing to include
“hyperspectral” images, wherein the value for each pixel is a value for each
spectral band to establish a signature for each pixel. As the Supreme Court
guides, “[w]hen a work is available in one field of endeavor, design
incentives and other market forces can prompt variations of it, either in the
same field or a different one” (KSR, 550 U.S. at 417). When considering
obviousness of a combination of known elements, the operative question is
thus “whether the improvement is more than the predictable use of prior art
elements according to their established functions” (Id.).
Appellants have not provided any evidence that such modification to
apply Auty’s and Oyaizu’s method with hyperspectral images would have
been “uniquely challenging or difficult for one of ordinary skill in the art”
(Leapfrog Enters., Inc. v. Fisher-Price, Inc., 485 F.3d 1157, 1162 (Fed. Cir.
2007)), or would have yielded unexpected results. Here, we agree with the
Examiner that Appellants’ invention is simply a modification of familiar
prior art teachings (as taught or suggested by the cited references) that would
have realized a predictable result (KSR, 550 U.S. at 421).
On this record, we find no error in the Examiner’s rejection of
claim 1, and claims 2, 7, and 10 depending therefrom over Auty, Oyaizu,
and Banerjee. Appellants do not provide substantive arguments for the
dependent claims separate from claim 1 from which they depend (App. Br.
21—22). Thus, we also affirm the rejection of claim 5 over Auty, Oyaizu,
and Banerjee, in further view of Tsuchikawa; of claim 6 over Auty, Oyaizu,
11
Appeal 2015-007042
Application 13/673,052
and Banerjee, in further view of Jain; of claim 8 over Auty, Oyaizu, and
Banerjee, in further view of Jin; of claim 9 over Auty, Oyaizu, and Banerjee,
in further view of Brill.
V. CONCLUSION AND DECISION
We affirm the Examiner’s rejections of claims 1, 2, and 4—10 under 35
U.S.C. § 103(a) and of claims 4 and 9 under 35 U.S.C. § 112, first
paragraph.
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)(iv).
AFFIRMED
12