Ex Parte Burnette et al

Patent Trial and Appeal Board

Sep 14, 2017

13427964 (P.T.A.B. Sep. 14, 2017)

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.
13/427,964 03/23/2012 Donald Jason Burnette XSDV 3.0-523 8837
146033
Waymo LLC
Lerner David
7590 09/18/2017 EXAMINER
NGUYEN, NGA X
600 South Avenue West
Westfield, NJ 07090 ART UNIT PAPER NUMBER
3662
NOTIFICATION DATE DELIVERY MODE
09/18/2017 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):
eofficeaction@lernerdavid.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte DONALD JASON BURNETTE and DAVID I. FERGUSON
Appeal 2015-004446
Application 13/427,964
Technology Center 3600
Before MICHAEL W. KIM, CHARLES N. GREENHUT, and
LISA M. GUIJT, Administrative Patent Judges.
GUIJT, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Appellants1 seek our review under 35 U.S.C. § 134 of the Examiner’s
decision2 rejecting claims 1, 3-14, and 16-26. We have jurisdiction under
35 U.S.C. § 6(b).
We REVERSE.
1 Appellants identify the real party in interest as Google Inc. Br. 1.
2 Appeal is taken from the Final Office Action dated April 7, 2014 (“Final
Act.”), as supplemented by the Advisory Action mailed July 2, 2014 (“Adv.
Act.”).
Appeal 2015-004446
Application 13/427,964
CLAIMED SUBJECT MATTER
Claims 1, 14, and 26 are the independent claims on appeal. Claim 1,
reproduced below, is illustrative of the subject matter on appeal.
1. A method comprising:
accessing scan data collected by a laser for a roadway, the
scan data including a plurality of data points having location and
intensity information for objects;
dividing the plurality of data points into sections;
for each section, identifying a threshold intensity;
generating, by a processor, a set of lane marker data points
from the plurality of data points by
(1) evaluating each particular data point of the plurality
by comparing the intensity value for the particular data point to
the threshold intensity value for the section of the particular data
point, and
(2) selecting data points of the plurality of data points
having locations within a threshold distance from a surface of the
roadway; and
storing the set of lane mark data points for later use.
REJECTIONS
I. Claims 1, 3-5, 8, and 10 stand rejected under 35 U.S.C.
§ 102(b) as anticipated by Saito (US 8,340,896 B2; issued Dec. 25, 2012).
II. Claims 6, 7, 9, 11-14, and 16-26 stand rejected under 35
U.S.C. § 103(a) as unpatentable over Saito and Zhang (US 2010/0121577
Al; published May 13, 2010).
2
Appeal 2015-004446
Application 13/427,964
ANALYSIS
Rejection I
Regarding independent claim 1, the Examiner finds, inter alia, that
Saito teaches capturing images using laser radar, such that the data includes
“distance data of an object based on the information of the reflected light.”
Final Act. 3 (citing Saito 9:25-50, 15:42-52). The Examiner determines
that this teaching in Saito discloses scanned data including a plurality of data
points having location and intensity information for objects, as claimed. Id.
The Examiner also finds that Saito discloses dividing the data points into
sections, as claimed, because Saito’s approximation line calculating means
10 divides distance data D into groups Gi and G2. Id. (citing Saito 16:60-
67). The Examiner further finds that Saito discloses identifying a threshold
intensity for each group, as claimed, because Saito’s approximation line
calculating means 10 “approximates lines Li and L2, standard deviations,
and variances.” Id. at 3—4 (citing Saito 9:53-10:30, 12:45-67, 15:45^16).
Appellants submit that
approximation line calculating means 10 is configured so as to
divide a plurality of the distance data D into a group Gi on the
near [side] of the vehicle and a group G2 on the far side of the
vehicle based on the distance in real space Z (parallax dp) of the
distance D detected by the distance and height detecting means
6.
Appeal Br. 7 (quoting Saito 9:53-58) (citing Saito Figs. 4A-8). Appellants
also submit that “[t]he height value of D, or Y, is then used by the
approximation line calculating means 10 to estimate lines . . . Li and L2. . .
‘which approximate the distance data D belonging to groups Gi and G2.’”
Id. at 7-8 (citing Saito 9:65-11:62). Appellants argue that “[t]hus, Saito is
clear that it is distance data which is divided into groups GI and G2 and
3
Appeal 2015-004446
Application 13/427,964
not scan data ‘including a plurality of data points having location and
intensity information’ as claimed.’” Id. at 8. Appellants argue that claim 1
requires “for each section, identifying a threshold intensity,” and that “it is
entirely unclear how Saito could teach or suggest identifying a threshold
intensity for the distance data D of groups Gi and G2,” because groups Gi
and G2 are disclosed as data comprised of measurements including only
distance and height values—not intensity values. Id.
The Examiner responds that Saito teaches that image processor 7 may
be “a stereo camera or a laser radar distance measuring device,” and that
distance and height detecting means 6 “may[ ]be a device which irradiates a
laser beam in front of the vehicle to measure the distance data D.” Ans. 3
(citing Saito 1:31-34, 7:30-31, 9:25-35) (emphasis omitted). The Examiner
also determines that Saito’s disclosure of “image data (distance data) having
pixels, wherein each pixel has coordinate (i, j), and luminance . . . meets the
scope of ‘a scan data including a plurality of data points having location and
intensity information for object[s].’” Id.
Indeed, Saito discloses that distance and height means 6 may include
an image processor or a laser device to detect points by measuring or
calculating distance data D (i.e., distance Z, height Y). See Saito 8:18-19
(“distance and height detecting means 6 include[s] an image processor 7,
and detects a plurality of distance data having at least information of
distance Z and height Y, in real space”); cf. id. at 9:30-33 (“distance and
height detecting means 6 may be, a device which irradiates a laser beam in
front of the vehicle to measure the distance data D of an object based on the
information of the reflected light”). However, to the extent the Examiner
relies on image processor 7 for analyzing pixels and luminance to result in
4
Appeal 2015-004446
Application 13/427,964
the claimed intensity information, intensity value, and threshold intensity
value, image processor 7 is exclusive (or an alternative to) to “accessing scan
data collected by a laser for a roadway,” as recited in claim 1. Thus, the
Examiner’s reliance on image processor 7 is in error.
Notwithstanding, the Specification discloses that “[e]ach data
point. . . may include an intensity value indicative of the reflectivity of the
object from which the light was received by the laser as well as a location
and elevation component: (x, y, z).” Spec. ^ 57. Thus, Saito’s use of a laser
to measure distance data D based on reflectivity of the object inherently
results in collecting and using “intensity information” for data points
scanned by the laser, and thus, a preponderance of the evidence supports the
Examiner’s finding that Saito discloses “scan data including a plurality of
data points having location and intensity information for the objects,” as
claimed.
However, regarding Appellants’ argument that Saito does not disclose
that the sectioned data in groups Gi and G2 include intensity values, such
that the group Gi and G2 data (and approximation line calculating means 10)
can be relied on by the Examiner for identifying a threshold intensity for
each section, we agree. Saito’s scan data is only disclosed for use as
distance data D by height and distance detecting means 6 “having at least
information of the distance Z and height Y in real space,” as discussed
supra. See also Saito 10:4 (“distance data D is represented with (Z, Y).”).
The Examiner has not provided, nor can we find independently, express
support in Saito that Saito’s distance data D includes, or is somehow an
equivalent of, an intensity value for each data point. Thus, when Saito’s
approximation line calculating means 10 divides distance data D into groups
5
Appeal 2015-004446
Application 13/427,964
Gi and G2, and calculates lines Li and L2 for approximating the data in each
group (see, e.g., Saito 9:53-64), the result is lines Li and L2 (i.e., linearly
fitting distance data D). The Examiner has failed to explain adequately how
lines Li and L2, which approximate distance data D in two-dimensions
(Z=distance, Y=height), also disclose identifying a threshold intensity for
each section of data points, as claimed, which would appear to only be in
one dimension.
Notably, the Examiner’s further reliance on Saito’s statistics
calculating means 11 does not cure the deficiencies in the Examiner’s
findings discussed supra. See Final Act. 3—4 (“. . . standard deviations, and
variances”). Saito discloses that the statistics calculated by Saito’s statistics
calculating means 11 include variances 012, 022 of distance data D of the
groups Gi and G2, as to the approximation lines Li and L2. Saito 12:60-67.
Thus, Saito fails to disclose that the statistics include, for each group,
identifying a threshold intensity, as claimed.
The Examiner alternatively finds that “one of ordinary skill in the art
would know that a near side and far side of the vehicle indicates a certain
short distance (or a certain strong intensity) and a certain long distance (or a
certain weak intensity), respectively, since the distance is a function of
signal strength (intensity),” and, therefore, because groups Gi and G2
identify near and far sides of the vehicle, groups Gi and G2 equivalently
identify threshold intensities. Adv. Act. 3; see also Ans. 3. Appellants
argue that the Examiner has not provided any evidence that distance and
intensity are interchangeable. Br. 8-9. We agree that the Examiner’s
conclusory determination lacks adequate evidentiary support.
6
Appeal 2015-004446
Application 13/427,964
The Examiner also alternatively finds that “Saito’s system also
teaches the parallax dp(n) is assigned in each pixel block PB(n), wherein the
parallax dp(n) is an averaging of the intensity (Ap(n) - Apth), wherein Apth
is a predetermined threshold,” such that parallax dp(n) is “equivalent to”
identifying threshold intensity for each section, as claimed. Adv. Act. 3
(citing Saito 19:49-20:7); see also Ans. 4 (“the distance and height data is
divided into pixel block dp (n), wherein each pixel block has a
predetermined threshold intensity.” (citing Saito 19:35-57) (emphasis
omitted)).
Appellants argue that “parallax dp(n) is ‘the distance in real space Z
of the distance D detected by the distance and height detecting means 6,”
and therefore, “it does not follow that this parallax dp(n) is an equivalent of
identifying for each of Gi and G2 a threshold intensity value.” Br. 10 (citing
Saito 9:56-58). Appellants also argue that “the Examiner has attempted to
combine completely different embodiments of Saito in order to achieve a
result,” in that “Saito’s reference pixels refer to camera images that were
captured by two cameras, and not a plurality of data points included in ‘scan
data collected by a laser for a roadway’ as in claim 1.” Id. The Examiner
responds that Saito’s disclosure of the “road marking detecting means” is in
addition to, but consistent with, previous embodiments. Ans. 4.
Appellants are correct—Saito’s road marking detecting means 13 is a
second embodiment that generates a road shape model “by detecting a
marking such as a lane line lateral to the vehicle,” rather than the road
surface generally. Saito 15:25, 35-36; see also id. at 16:20-27 (“road
marking detecting means 13 detects a marking on a road surface such as a
lane line or the like based on the luminance p that is the image data of each
7
Appeal 2015-004446
Application 13/427,964
pixel on the reference image T to be sequentially transmitted [to distance
and height detecting means 6].”). However, similar to the first embodiment,
the second embodiment may also use either a camera or a laser device to
obtain distance and height data (see, e.g., Saito 15:53-64), and the
Examiner’s reference to “an averaging of the intensity” relates to using
image capturing means 2 to detect the road marking, which is not collected
by a laser, as claimed.
Moreover, similar to distance data D discussed supra, Saito discloses
(in the context of the second embodiment) that “with regard to the great
number of road marking points cl and cr detected by the road marking
detecting means 13, the approximation line calculating means 10 calculates
the distance Z and height Y in real space,” and “as shown in the first
embodiment calculates the approximation lines Li and L2 for the groups Gi
and G2 respectively using the respective distance data D of the road marking
points cl and cr which have at least the distance Z and height Y in real
space.” Id. 16:49-59. Thus, for the same reasons discussed supra with
respect to Saito’s first embodiment, a preponderance of the evidence fails to
support the Examiner’s finding that Saito’s lines Li and L2, which
approximate distance data D in two-dimensions (Z=distance, Y=height),
disclose adequately identifying a threshold intensity for each section of data
points, as claimed, which only is in one.
Accordingly, we do not sustain the Examiner’s rejection of
independent claim 1 and claims 3-5, 8, and 10 depending therefrom.
8
Appeal 2015-004446
Application 13/427,964
Rejection II
Independent claim 14 and dependent claims 6, 7, 9, 11—13, and 16—25
Similar to independent claim 1, independent claim 14 requires, in
relevant part, accessing scan data collected by a laser for a roadway, the scan
data including a data points having location and intensity information for
objects, dividing the data points into sections, and for each section,
evaluating the data points to determine a respective average intensity, and
determining a threshold intensity based on the respective average intensity.
See Br. 44 (Claims App.). The Examiner relies on the same deficient
findings from Saito as relied upon for claim 1 in the Examiner’s rejection of
claim 14. Final Act. 8-9.
Additionally, the Examiner’s reliance on Zhang for more particularly
teaching the determination of a threshold intensity based on an average
intensity does not cure the deficiencies in the Examiner’s findings with
respect to Saito, because the Examiner’s reliance on Zhang lacks support.
Specifically, a preponderance of evidence supports Appellants’ argument
that “the ‘moving threshold’” disclosed in Zhang’s paragraph 114, and relied
on by the Examiner {see Final Act. 9), “is not a threshold intensity but
relates to location.” Br. 30; see Zhang ^ 106 (“Signal processing includes
refining the three-dimensional point cloud by identifying and eliminating
datapoints . . . , including any datapoint that has a large Z-axis component or
a Y-axis component that is a distance greater than a predetermined
threshold.”); id. 107 (“A sample dataset M is selected from the refined
three-dimensional point cloud.”).
Accordingly, for the same reasons stated supra, we do not sustain the
Examiner’s rejection of independent claim 14 and claims 16-25 depending
9
Appeal 2015-004446
Application 13/427,964
therefrom. Because we do not sustain the Examiner’s rejection of
independent claim 1, for the reasons stated supra, we also do not sustain the
Examiner’s rejection of claims 6, 7, 9, and 11-13, which depend from claim
1.
Independent claim 26
Regarding independent claim 26, the Examiner relies on the same
findings with respect to Saito as relied upon in the Examiner’s rejection of
independent claim 1. See Final Act. 7-8. The Examiner finds that the
claimed scan data is collected for a roadway, in Saito, by “cameras or laser
radar.” Final Act. 7 (citing Saito 9:25-50). Although claim 26 does not
require the accessed scan data to be collected by a laser (cf. independent
claims 1, 14), such that an image capturing device may be used, a
preponderance of evidence still fails to support the Examiner’s finding that
lines Li and L2, which approximate distance data D in two-dimensions
(Z=distance, Y=height), disclose identifying a threshold intensity for each
section (or group) of data points, as claimed.
Accordingly, we do not sustain the Examiner’s rejection of
independent claim 26.
DECISION3
We reverse the Examiner’s rejection of claim 1, 3-14, and 16-26.
REVERSED
3 No inference should be drawn from the failure of the Board to make a new
ground of rejection of any claims. As the Board’s function is primarily one
of review, we leave it to the Examiner to determine the patentability of the
claims subject to this appeal.
10