`
`By:
`
`Steven Ross
`Christopher P. O’Hagan
`Ross IP Group PLLC
`1700 Pacific Ave., Suite 3750
`Dallas, TX 75201
`Tel: (972) 661-9400
`Fax: (972) 661-9401
`E-mail:
`sross@rossipg.com
`cpohagan@rossipg.com
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`____________________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`____________________________
`
`ROTHSCHILD MOBILE IMAGING INNOVATIONS, LLC
`
`Petitioner
`
`V.
`
`MITEK SYSTEMS, INC.
`
`Patent Owner
`
`____________________________
`
`Patent No. 8,379,914
`
`____________________________
`
`DECLARATION OF DR. FANG QIU
`
`Page 1 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`TABLE OF CONTENTS
`
`I.
`
`Introduction ......................................................................................................5
`
`II. Qualifications....................................................................................................5
`
`III. Summary of Opinions.......................................................................................6
`
`IV. The ’914 Patent.................................................................................................8
`
`A. Overview of the ’914 Patent ......................................................................... 8
`
`B. Level of Ordinary Skill in the Art ............................................................... 10
`
`V. Claim Construction.........................................................................................10
`
`VI. Claimed Subject Matter of the ’914 Patent Disclosed by the Prior art.............11
`
`A. Brief Summary of Prior art: ........................................................................ 11
`
`1. Hoyos (Ex. 1001): ................................................................................... 11
`
`2. Pandian (Ex. 1003): ................................................................................. 13
`
`3. Baker (Ex. 1004): .................................................................................... 15
`
`4. Sipe (Ex. 1005):....................................................................................... 15
`
`5. Schwalb (Ex. 1006): ................................................................................ 16
`
`6. Du (Ex. 1007):......................................................................................... 16
`
`2
`
`Page 2 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`7. Hung (Ex. 1008):..................................................................................... 16
`
`8. Bressan (Ex. 1009): ................................................................................. 17
`
`B. Hoyos discloses Claims 1 and 10................................................................ 18
`
`Claim 1.......................................................................................................... 18
`
`Claim 10........................................................................................................ 27
`
`C. Combination of Hoyos and Pandian discloses Claims 1, 4 and 10 .............. 29
`
`Claim 1.......................................................................................................... 29
`
`Claim 4.......................................................................................................... 33
`
`Claim 10........................................................................................................ 34
`
`D. Combination of Hoyos and Hung discloses Claim 2 ................................... 35
`
`E. Combination of Hoyos, Pandian and Hung discloses Claim 2..................... 36
`
`F. Combination of Hoyos and Bressan discloses Claim 3 ............................... 36
`
`G. Combination of Hoyos, Pandian and Bressan discloses Claim 3................. 38
`
`H. Combination of Hoyos and Baker discloses Claim 5 .................................. 38
`
`I. Combination of Hoyos, Pandian and Baker discloses Claim 5.................... 40
`
`J. Combination of Hoyos and Sipe discloses Claim 6..................................... 40
`
`K. Combination of Hoyos, Pandian and Sipe discloses Claim 6 ...................... 41
`
`3
`
`Page 3 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`L. Combination of Hoyos, Sipe and Baker discloses Claim 7.......................... 41
`
`M. Combination of Hoyos, Pandian, Sipe and Baker discloses Claim 7 ........ 42
`
`N. Combination of Hoyos, Sipe, Baker and Schwalb discloses Claim 8 .......... 42
`
`O. Combination of Hoyos, Pandian, Sipe, Baker and Schwalb discloses Claim 8
`
`43
`
`P. Combination of Hoyos and Du discloses Claim 9....................................... 44
`
`Q. Combination of Hoyos, Pandian and Du discloses Claim 9 ........................ 46
`
`VII. Conclusion ..................................................... Error! Bookmark not defined.
`
`4
`
`Page 4 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`I, Dr. Fang Qiu, declare as follows:
`
`I.
`
`Introduction
`
`a)
`
`My name is Fang Qiu. I reside at Texas. My findings, as set forth
`
`herein, are based on my education and background in the fields discussed below.
`
`b)
`
`I have been retained on behalf of Petitioner Rothschild Mobile
`
`Imaging Innovations, LLC to provide this Declaration concerning technical subject
`
`matter relevant to the inter partes review petition (“Petition”) pertaining to U.S
`
`Patent No. 8,379,914 (“the ̕ 914 Patent”) (filed herewith as Exhibit 1002.
`
`c)
`
`I am over 18 years of age. I have personal knowledge of the facts
`
`stated in this Declaration and could testify competently to them if asked to do so.
`
`II. Qualifications
`
`1.
`
`I am a Professor and Department Head of Geospatial Information
`
`Sciences and Remote Sensing at the University of Texas at Dallas. I teach graduate
`
`classes in Computer Programming for GIS, Advanced Remote Sensing, and
`
`Advanced Geospatial Information Sciences. I received a Bachelor of Science degree
`
`in Geography from East China Normal University in Shanghai, China, in 1990. I
`
`received a Master of Science degree in Geographic Information Systems from the
`
`Chinese Academy of Sciences in Beijing, China, in 1993.
`
`I received a Ph.D. in
`
`Geography with a concentration in geographic information processing from
`
`University of South Carolina in Columbia, South Carolina, in 2000.
`
`5
`
`Page 5 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`2.
`
`My research and instructional areas are remote sensing digital image
`
`processing, spatial analysis and modeling, GIS application software development,
`
`and web-based mapping and information processing. I have extensive expertise in
`
`image segmentation, image matching, address matching and geocoding, which are
`
`pertaining to this declaration.
`
`3.
`
`My publications have appeared in peer-reviewed journals such as
`
`International Journal of Remote Sensing, Photogrammetric Engineering and
`
`Remote Sensing, Computers and Geosciences, GIScience and Remote Sensing,
`
`GeoCarto International, Forest Ecology and Management, Journal of Geophysical
`
`Research and other academic journals.
`
`4.
`
`I was the Winner of 2013 ERDAS Award for Best Scientific Paper in
`
`Remote Sensing by American Society of Photogrammetry and Remote Sensing, and
`
`Winners of Remote Sensing Special Group Award for 2011 and 2013 by American
`
`Association of Geographers.
`
`5.
`
`A true and correct copy of my resume is attached hereto.
`
`I am being compensated at my standard flat-fee rate of $2000 for my work in
`
`this matter. However, the compensation in no way affects my statements in this
`
`declaration.
`
`III. Summary of Opinions
`
`6
`
`Page 6 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`The analysis and opinions that I provide in this Declaration are based on my
`
`experience in this field, my review of the ’914 Patent, as well as the following
`
`documents as disclosed herein:
`
`• The prosecution history of the ’914 Patent
`
`• U.S. Patent Publication No. 2002/0037097 to Hoyos , Hector ; et al.
`
`(“Hoyos” ) (“Ex. 1001” )
`
`• U.S. Patent Publication No. 2005/0289182 to Pandian, Suresh S. ; et al.
`
`(“Pandian”) (“Ex. 1003”)
`
`• U.S. Patent No. 5,862,243 to Baker, et al. (“Baker”) (“Ex. 1004”)
`
`• U.S. Patent Publication No. 2004/0093222 to Sipe , Wayne et al. (“Sipe”)
`
`(“Ex. 1005”)
`
`• U.S. Patent No. 7,877,403 to Schwalb (“Schwalb”) (“Ex. 1006”)
`
`• U.S. Patent No. 7,734,729 to Du, et al. (“Du”) (“Ex. 1007”)
`
`• U.S. Patent No. 6,749,120 to Hung et al. (“Hung”) (“Ex. 1008”)
`
`• U.S. Patent No. 7,826,665 to Bressan et al. (“Bressan”) (“Ex. 1009”)
`
`• U.S. Provisional Patent Application No. 61/561,772 (“Ex. 1010”)
`
`• U.S. Patent Application No. 12/906,036 (“Ex. 1011”)
`
`• U.S. Patent Application No. 12/778,943 (“Ex. 1012”)
`
`• U.S. Patent Application No. 12/346,026 (“Ex. 1013”)
`
`• U.S. Provisional Patent Application No. 61/022,279 (“Ex. 1014”)
`7
`
`Page 7 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`• U.S. Patent Application No. 13/622,329 (issued as the ’914 Patent) (“Ex.
`
`1015”)
`
`IV. The ’914 Patent
`
`A. Overview of the ’914 Patent
`
`Generally, I understand that the ’914 Patent is directed to automated
`
`processing of a financial document such as a coupon. Ex. 1002, col. 1, ll. 26-32.
`
`The ’914 Patent teaches that an image of the remittance coupon is captured by a
`
`mobile device, and at least one aspect of the image is corrected to produce a
`
`corrected image. See id., FIGs. 1 and 2. The corrected image is subjected to a first
`
`content recognition pass to extract content from the remittance coupon. In my
`
`opinion, capturing an image using a mobile device, and correcting a digital image
`
`using different correction techniques, and extracting data from a corrected digital
`
`image are very well-known steps in the art at the time of the alleged invention.
`
`The ’914 Patent further teaches, as is known in the art, that an address of a
`
`biller on the remittance coupon is identified by comparing address content in the
`
`extracted content with an address database. The ’914 Patent further teaches a
`
`conventional method of determining the identity of a biller by comparing one
`
`identified information of the biller and comparing the identified information
`
`against a database of biller profile information. The identified information
`
`according to the claims of the ’914 Patent is the address information. The ’914
`
`8
`
`Page 8 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Patent also teaches a conventional step of using a set of billing information, which
`
`includes the extracted content and the identity of the biller for processing a
`
`payment of the bill. Further, the ’914 Patent teaches using the biller profile
`
`information of the biller to extract content from the remittance coupon. The biller
`
`profile information according to the ’914 Patent includes at least one of a
`
`remittance coupon format, a remittance coupon mask, a location of at least one
`
`field on the remittance coupon and a format of at least one field. See id., col.3, ll.
`
`4-10. In my opinion, to improve the quality of data extraction, it is known to use a
`
`coupon mask/format or location information of the fields to extract required fields
`
`from a coupon image based on the coupon mask.
`
`Thus, I am of the opinion that the techniques and processes recited in claims
`
`1-10 of the ’914 patent are nothing more than mere description of prior art.
`
`I have reviewed the earlier filed patent application nos. 61/561,772 (Ex.
`
`1010), 12/906,036 (Ex. 1011), 12/778,943 (Ex. 1012), 12/346,026 (Ex. 1013) and
`
`61/022,279 (Ex. 1014), from which the ’914 Patent claims priority, and none of
`
`these applications, in my opinion, disclose “identifying an address of a biller on the
`
`remittance coupon by comparing address content in the extracted content with an
`
`address database” and “determining biller profile information of the biller,
`
`including an identity of the biller on the remittance coupon, by comparing the
`
`identified address of the biller with a database of biller profile information.” The
`
`disclosure of these features, in my opinion, is first found in U.S. Patent Application
`9
`
`Page 9 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`No. 13/622,329 (issued as the ‘914 patent, Ex. 1015) at Figs. 1and 2, and ¶¶ [0145]
`
`and [0146].
`
`B. Level of Ordinary Skill in the Art
`
`Based on my experience in this area, a person of ordinary skill in the art in
`
`technical areas pertaining to the’914 Patent at the relevant time frame of the
`
`alleged invention would have had a combination of experience and education in
`
`computer science. This would typically consist of at least a bachelor’s degree in
`
`computer science or a related engineering field and 2-3 years of work and/or
`
`research experience in the field of computer science, image processing or a related
`
`field.
`
`V. Claim Construction
`
`I have been informed that a claim subject to inter partes review receives the
`
`broadest reasonable construction in light of the specification of the patent in which
`
`it appears. I have also been informed that other terms in the claims which have not
`
`been construed should be given its plain and ordinary meaning under the broadest
`
`reasonable construction. I abided by these principles while providing my analysis
`
`and opinions. I have been informed that it would be useful to provide some
`
`guidance in this proceeding with respect to the term “code line” and asked me to
`
`consider the corresponding construction. As part of that, I have considered the
`
`term’s (“code line”) context within the claim, use within the specification, and my
`
`10
`
`Page 10 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`understanding of how one of ordinary skill in the art would understand the term
`
`around the time of the purported invention.
`
`I have considered whether a broadest reasonable interpretation of “code line”
`
`would be broad enough to cover “encoded information that can be detected and
`
`decoded by a computer system to extract the information encoded therein.” I
`
`believe that it would, since such an interpretation is consistent with the
`
`specification of the ’914 Patent (See Id., col. 23, ll. 1-8), and the understanding one
`
`of ordinary skill in the art when applying the broadest reasonable construction.
`
`VI. Claimed Subject Matter of the ’914 Patent Disclosed by the Prior art
`
`A.
`
`Brief Summary of Prior art:
`
`I understand that the following references are prior art to the ’914 Patent. In
`
`my opinion these references either alone or in combination disclose all features of
`
`the claims 1-10 of the’914 Patent.
`
`1. Hoyos (Ex. 1001):
`
`Hoyos disclosure relates to the same field as the ’914 Patent, which is
`
`automatically recognizing a bill or a coupon used in the sale or purchase of goods
`
`and services. Ex. 1001 at ¶ [0001].
`
`Hoyos disclose describes “a scanner configured to receive a bill or coupon.
`
`The coupon is processed by … [applying] connected component analysis,
`
`segmentation, coupon matching, and data extraction to determine an associated
`11
`
`Page 11 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`vendor and customer account information. This information is used to complete a
`
`payment transaction.” See id., Abstract.
`
`Fig. 1:
`
`In the above process, the image scanned by the scanner undergoes
`
`preprocessing in the preprocessor, where the scanned coupon is corrected. See id.,
`
`¶ [0050]. Further, connected component analysis is performed to extract data from
`
`the image, which underwent correction. See id., ¶¶ [0033], [0043]-[0046] and
`
`[0050]. The useful regions in the coupon image are identified, and the data
`
`corresponding to the regions is extracted (using OCR) and compared against the
`
`data in the database. Id., ¶¶ [0007], [0144] and [0145].
`
`Fig. 3.
`
`12
`
`Page 12 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Further, Hoyos discloses a database of vendor data which is used to detect a
`
`match that determines a vendor on the coupon image. See id., ¶ [0007]. Further, the
`
`results of the analysis, i.e. vendor’s ID, an account number, and name and address
`
`information are provided to complete a payment transaction. See id., ¶ [0035],
`
`Abstract.
`
`2.
`
`Pandian (Ex. 1003):
`
`Pandian disclosure relates to the field of document recognition based
`
`document management system, where the documents are images captured using
`
`electronic document sources such as facsimile images, scanned images. Ex. 1003
`
`at ¶ [0002] and Abstract.
`
`Pandian disclosure describes “modules for image capture, image
`
`enhancement, image identification, optical character recognition, data extraction
`
`and quality assurance. The system captures data from electronic documents as
`
`diverse as facsimile images, scanned images and images from document
`
`management systems. It processes these images and presents the data in, for
`
`example, a standard XML format ... [t]he system can extract images directly from a
`
`facsimile machine, a scanner or a document management system for processing.”
`
`Id., Abstract.
`
`In the above process, Pandian describes that the captured images are corrected
`
`for skewed errors, water marks, holes punched … etc. by the image enhancement
`
`13
`
`Page 13 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`module 32 to make the optical character recognition more accurate. See Id., ¶
`
`[0079]
`
`Figure 2B.
`
`Further, corrected images are processed by OCR module 37, where OCR is
`
`performed and a feedback relating to the quality of the OCR is generated. If the
`
`quality of the OCR is lesser than the desired degree of quality, then the feedback
`
`loop module 39 sends the results to the enhancement module to further enhance the
`
`image and perform OCR again. This process is repeated until the OCR output of
`
`desired quality is obtained. Further, Pandian also describes that the image
`
`identification module can distinguish between different types of document images
`
`such as Bank of America statement, Citibank statement and Utility bill. See id., ¶¶
`
`[0138]-[0141].
`
`14
`
`Page 14 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`3. Baker (Ex. 1004):
`
`Baker discloses “a system for evaluating barcoded mail which includes an
`
`imaging device to provide an image signal corresponding to an image of a barcode
`
`or an address of a mail piece.” Ex. 1004 at Abstract.
`
`Baker further describes that the barcoded mail are evaluated using OCR.
`
`Specifically, the processor 50 in Baker determines the destination address on the
`
`image of the mail. Further, the address information is decoded from the barcode on
`
`the image. Finally the address is evaluated by comparing the destination address
`
`determined using OCR with the address information decoded from the barcode.
`
`See Id., Col. 6, ll.65 - Col. 7, ll. 4. Baker further describes that these addresses can
`
`include ZIP code or “ZIP+4” for detailed destination address. See id., col. 1, ll. 16-
`
`22.
`
`4.
`
`Sipe (Ex. 1005):
`
`Sipe discloses a process of creating a address database by taking daily address
`
`information and updating the United States Postal Service (USPS) NCOA
`
`database. Sipe also teaches that this process will enable the collection of address
`
`change information at near real time, validate the change information, and
`
`distribute this new address database to licensed users on a daily or more frequent
`
`basis. Ex. 1005 at Abstract.
`
`15
`
`Page 15 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`5.
`
`Schwalb (Ex. 1006):
`
`Schwalb discloses a method of searching a database using fuzzy rules. The
`
`method accepts a word or word phrase such as a person’s name or address and
`
`returns fuzzy rules for searching the database. Further Schwalb discloses selecting
`
`and modifying the fuzzy rules to improve the accuracy of search results from the
`
`database. Schwalb also identifies that there are problems associated with searching
`
`databases for addresses. Ex. 1006 at Abstract and col. 1, ll. 38-40.
`
`6. Du (Ex. 1007):
`
`Du discloses a method for displaying, on the portable imaging device,
`
`information associated with an item selected by the user from a remote location. In
`
`the above method Du describes that “a user at the location of the first entity
`
`operates a portable imaging device to capture an image of identifying data, such as
`
`a barcode, that identifies a selected item. The captured image is then
`
`communicated to a server operated by a second entity that is different than the first
`
`entity to obtain item information (e.g., price, availability, etc.) associated with the
`
`selected item. The item information is communicated back to the portable imaging
`
`device for display to the user while the user remains at the location of the first
`
`entity”. Ex. 1007 at Abstract.
`
`7. Hung (Ex. 1008):
`
`16
`
`Page 16 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Hung discloses a scanner that can read two-dimensional barcodes from
`
`reflective or emissive electronic displays. Ex. 1008 at Abstract. The scanner can
`
`also adapt to read barcodes from displays with non-square pixels. More
`
`specifically, the scanner captures an image of a barcode displayed on an electronic
`
`screen. The image is provided to the digital processor, where the known barcode
`
`elements are measured and the image data is scaled to produce barcode elements
`
`within the aspect ratio tolerance. Id., col. 13, ll. 1-3. The processor determines if
`
`the two-dimensional barcode elements are square or not by comparing the lengths
`
`of adjoining sides of the regions. Further, in case of not a square the processor
`
`scales the image either by lengthening one axis or/and shortening the other. See id.,
`
`col. 11, ll. 61 to col. 12, ll. 2.
`
`8. Bressan (Ex. 1009):
`
`Bressan discloses a system for updating a contacts database. Bressan’s
`
`system uses a portable imager to acquire a digital image of a business card. Ex.
`
`1009 at Abstract. Thereafter, the image pre-processor performs pre-processing on
`
`the digital image, where the pre-processing includes shadow correction, re-sizing
`
`the image or other corrections. Id., col. 4, ll. 26-31.
`
`Fig.1:
`
`17
`
`Page 17 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Further, an image segmenter extracts text image segments from the digital
`
`image, and an optical character recognizer (OCR) generates one or more textual
`
`content candidates for each text image segment. Thereafter, a content selector
`
`selects a textual content candidate for each text image segment based on the scores
`
`assigned by a scoring processor. Finally, an interface is configured to update the
`
`contacts list based on the selected textual content candidates. Id., Abstract.
`
`B.
`
`Hoyos discloses Claims 1 and 10
`
`Claim 1
`
`Claim 1(a):
`
`18
`
`Page 18 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`“A method of processing a remittance coupon captured by a mobile device”
`Hoyos discloses a method of recognizing a coupon by generating an
`
`electronic representation of the coupon (See e.g., Ex. 1001, ¶¶ [0001] and [0007]),
`
`and using the recognized information in the coupon to complete a payment
`
`transaction (See id., Abstract). This, in my opinion discloses processing a
`
`remittance coupon.
`
`Id., Fig. 1:
`
`Hoyos further discloses a scanner that scans the image and generates an
`
`electronic representation of the coupon. Id., ¶ [0032]. Further, it is also stated in the
`
`Specification of the ’914 Patent that the mobile device can be a scanner. See Ex.
`
`1002, col. 7, ll. 56-59. Thus, in my opinion the scanner is the mobile device that
`
`captures an image of the coupon and provides the image data to an associated
`
`computer for further processing. See id., ¶ [0004].
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration I am of the opinion that Hoyos discloses a
`
`method of processing a remittance coupon captured by a mobile device.
`19
`
`Page 19 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Claim 1(b):
`
`“receiving an image of a remittance coupon captured by a mobile device”
`
`In furtherance to my analysis provided above with respect to Claim 1(a),
`
`Hoyos discloses that the resulting image data from the scanner is provided to an
`
`associated computer. See id., ¶¶ [0004] and [0016].
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration I am of the opinion that Hoyos discloses the
`
`resulting image data, which is scanned by the scanner, of the remittance coupon is
`
`received by the associated computer. Therefore, Claim 1(b) is disclosed in Hoyos.
`
`Claim 1(c):
`
`“correcting at least one aspect of the image to produce a corrected image”
`
`Hoyos discloses that the image received from the scanner undergoes
`
`preprocessing (in the preprocessor 310 shown in Fig.3 of Hoyos) during which the
`
`coupon image undergoes skew correction (in the skew correction 410 as show in
`
`Fig.4 of Hoyos). Further, Hoyos clearly discloses that the coupon image is skew
`
`corrected in the preprocessor 310. See id., ¶ [0050].
`
`Figs. 3 and 4 (annotation added):
`
`20
`
`Page 20 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration I am of the opinion that Hoyos discloses
`
`correcting at least one aspect of the image to produce a corrected image. Therefore,
`
`Claim 1(c) is disclosed in Hoyos.
`
`Claim 1(d):
`
`“performing a first content recognition pass on the corrected image to extract
`content from the remittance coupon”
`
`Hoyos discloses that the data extracted during the process executed in the
`
`connected components block 416 is passed to a segmentator. See id., ¶ [0045].
`
`According to at least the FIGs. 3and 4 of Hoyos, the data extraction in the
`
`connected components block 416 is performed on the coupon image which is
`
`21
`
`Page 21 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`corrected at skew correction block 410. Thus, in my opinion data is extracted from
`
`the corrected image by performing a first content recognition pass which is
`
`performed by the connected components block 416 of Hoyos.
`
`My opinion is further reinforced by the disclose of Hoyos at ¶¶ [0045] and
`
`[0046], which in summary discloses that the extracted data is used to determine the
`
`type of content extracted for e.g. barcode, table, text area … etc, which in turn is
`
`used to determine the coupon type.
`
`FIGs. 3and 4 (annotation added).
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration I am of the opinion that Hoyos discloses
`
`Claim 1(d).
`
`Claim 1(e):
`
`22
`
`Page 22 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`“identifying an address of a biller on the remittance coupon by
`comparing address content in the extracted content with an address
`database”
`Hoyos discloses that “optical character recognition engine is applied to the
`
`relevant portion of the coupon image” to extract “[f]or example, the company
`
`name, its zip code, and its address”. See id., ¶ [0144]. Therefore, according to
`
`Hoyos the address content is part of the content extracted from the coupon image.
`
`Further, “[t]he resulting character string [of the optical character recognition
`
`engine] is compared to database entry.” Id. at ¶ [0144]. Thus, in my opinion the
`
`address content which is part of the extracted content is also compared to the
`
`address entries in the database. My analysis and opinion is further reinforced by
`
`the fact that Hoyos includes a database of vendor data. See id., ¶ [0007].
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration, I am of the opinion that Claim 1(e) is disclosed
`
`in Hoyos.
`
`Claim 1(f):
`
`“determining biller profile information of the biller, including an identity
`of the biller on the remittance coupon, by comparing the identified
`address of the biller with a database of biller profile information”
`
`23
`
`Page 23 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`In furtherance to my analysis provided above with respect to Claim 1(e),
`
`Hoyos discloses that “coupon is scanned and compared to a database of coupon
`
`data. The comparison is used to determine the coupon type and associated
`
`vendor”. See id., ¶ [0030]. Clearly, the associated vendor’s information is
`
`determined by comparing the coupon image with a database. This opinion is also
`
`based on Hoyos stating that “[t]he computer extracts prominent features from the
`
`image in order to determine (1) the company that issued the bill.” See id., ¶ [0004]
`
`(emphasis added). Thus, in my opinion vendor’s information which is the biller
`
`profile information of the biller is determined by comparing the coupon image to a
`
`database of vendor data. See id., ¶ [0007].
`
`Further, comparing the coupon image to a database of vendor data includes
`
`“[using] portions of text in the coupon image that is useful in the identification …
`
`For example … zip code, and its address are typical of useful regions in the
`
`identification process.” Id., ¶ [0144] (emphasis added). OCR is performed on the
`
`address field and “[t]he resulting character string is compared to database entry”
`
`(Id., ¶ [0144]). Hence, I am of the opinion that the resulting character string of the
`
`address is used with a database of vendor data to detect a match that determines a
`
`vendor. See id., ¶ [0007].
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration, I am of the opinion that Claim 1(f) is disclosed
`
`in Hoyos.
`
`24
`
`Page 24 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Claim 1(g):
`
`“producing a set of billing information, including the extracted content and
`the identity of the biller, for processing a payment of the bill”
`
`See Hoyos at Fig. 1 (annotation added):
`
`Hoyos discloses that “at block 118, the results of the coupon analysis are
`
`provided ... these will include an OCR line that includes the vendor's ID, an
`
`account number, an amount due, and name and address information” Id., ¶ [0035]
`
`(emphasis added). Thus, in my opinion, as the result of the coupon analysis Hoyos
`
`system produces identity of the biller (vendor’s ID) and the extracted content such
`
`as the name and address information, an amount due … etc. Further, Hoyos
`
`discloses that this information is used to complete a payment transaction. Id.,
`
`Abstract and Fig. 1. Therefore, in my opinion producing a set of billing
`
`information for processing a payment of the bill is disclosed in Hoyos.
`25
`
`Page 25 of 71
`
`ROTHSCHILD EXHIBIT 1016
`
`
`
`Declaration of Dr. Fang Qiu in support of Petition for Inter Partes Review of
`U.S. Patent No. 8,379,914
`
`Therefore, In view of the above analysis and my discussion of Hoyos in
`
`section VI.A(1) of this Declaration, I am of the opinion that Claim 1(g) is disclosed
`
`in Hoyos.
`
`Claim 1(h):
`
`“using the biller profile information of the biller to perform a second
`content recognition pass on the corrected image to extract content from
`the remittance coupon, wherein the biller profile information includes at
`least one of a remittance coupon format, a remittance coupon mask, a
`location of at least one field on the remittance coupon and a format of at
`least one field”
`
`Hoyos discloses that the fields which are of interest in the coupon image are
`
`identified based on the coordin