DETAILED ACTION
Contents
Notice of Pre-AIA or AIA Status 2
Election/Restrictions 2
Claim Rejections - 35 USC § 101 2
Claim Rejections - 35 USC § 102 3
Claim Rejections - 35 USC § 103 9
Conclusion 17
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Election/Restrictions
Applicant’s election without traverse of group I, claims 1-11, 19-20 in the reply filed on 10/1/25 is acknowledged.
Claims 1-20 are currently pending. Claims 12-18 are withdrawn from consideration.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 19 and all corresponding dependent claims are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter as follows. Claims 1 and 19 are directed to an abstract idea by recognizing text by analyzing symbol representations when character codes are missing or invalid. The steps constitute a mental process and data analysis which are forms of abstract ideas. Claims 1 and 19 do not integrate the abstract idea into a practical application. The claims have functional steps without reciting any specific technological improvement to computer functionality. The claims have generic components that perform the abstract steps. Claims 1 and 19 do not recite additional elements that amount to significantly more than the abstract idea. The additional elements are well-understood, routine and conventional computer operations. Implementing the abstract idea on a generic computer system does not provide an inventive concept. Thus, claims 1 and 19 and all corresponding dependent claims lack the inventive concept to transform the abstract idea into patent eligible subject matter.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless - (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention.(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-2, 19 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Pouyadou et al (US 2017/0161595 A1).
Regarding claim 1, Pouyadou discloses a method to perform text recognition, the method comprising: accessing a description of a symbol in a page description file for a document (see 0018, abstract; A method for extracting text from a page description language (“PDL”) document includes capturing a PDL file generated by a print driver for a printed document. The PDL file contains a set of drawing instructions for drawing a run of text formed from a series of glyphs. The PDL file is parsed to extract the drawing instructions of each glyph. The glyph drawing instructions are compared with a database of glyph characteristics. When a match is found between the drawing instructions and the database of glyph characteristics a corresponding text character associated with that set of characteristics is extracted. Where no match is found, the drawing instructions may be rendered as a bitmap and features extracted from the bitmap are compared with stored features to identify a glyph with similar features, or if there is none, an error may be recorded. A text summary may be generated….. The exemplary method provides for direct character recognition from page description language (PDL) documents. The method may include parsing a PDL file which is created when a user initiates a print job, for example by clicking “print” from an application. The PDL document to be parsed may be composed of drawing instructions. For each visual representation of a symbol (glyph) that is processed, the character it represents is found by matching the glyph characteristics with a reference database. The reference database can be based on glyph lookup in a database of fonts. The font can use the TrueType/OpenType technology in which case the recognition is made directly by the quadratic splines control points of the glyph. In other cases, recognition is made by a semi-exact visual similarity.);
identifying, responsive to a character code failure, the symbol using a vectorized drawing instruction (VDI) for the symbol (see 0018; The exemplary method provides for direct character recognition from page description language (PDL) documents. The method may include parsing a PDL file which is created when a user initiates a print job, for example by clicking “print” from an application. The PDL document to be parsed may be composed of drawing instructions. For each visual representation of a symbol (glyph) that is processed, the character it represents is found by matching the glyph characteristics with a reference database. The reference database can be based on glyph lookup in a database of fonts. The font can use the TrueType/OpenType technology in which case the recognition is made directly by the quadratic splines control points of the glyph. In other cases, recognition is made by a semi-exact visual similarity.), wherein the character code failure comprises one of: an absence of a character code in the description of the symbol, or a bad character code in the symbol description of the symbol (see 0017, 0028; Aspects of the exemplary embodiment relate to a method, apparatus and computer-readable medium for applying local character recognition to typical office documents at or about the time they are printed. The method is general enough to be usable in other contexts where similar font technologies are used and when glyph-to-character information is lost……. A glyph index 30, which stores the indices and instructions for rendering the corresponding characters, may be sent to the printer, e.g., in the PDL file 26. This index is often not specific to the font; therefore it is not possible to use a simple index-to-glyph mapping table. Most print drivers construct the glyph index 30 incrementally as they produce the PDL document 20. The indices are typically allocated whenever a character has to be drawn for the first time for a given font during the job. For example, in the above example, ‘e’ happens to be the first character to be displayed for the selected font during the job, therefore acquires the index 1, h is the 12th and acquires the index 12, and so on. From this, it should be clear that even if the hooking of text operators is possible, there is no practical way to go back to the character to be drawn from the glyph indices only.); and identifying a text of the document using the identified symbol (see 0009; In accordance with another aspect of the exemplary embodiment, a system for extracting text from a page description language (PDL) of a document includes a capture component that captures a PDL file generated by a print driver for a printed document. The PDL file contains a set of drawing instructions for drawing a run of text formed from a series of glyphs in the printed document. A parser parses the PDL file to intercept the drawing instructions of each glyph. A comparison component compares the drawing instructions of each glyph with a database of glyph characteristics. An extraction component extracts a text character associated with each glyph when a match is found between the drawing instructions and the database of glyph characteristics based on the comparison of the drawing instructions with the database of glyph characteristics. A summary component generates a text summary of the extracted text characters associated with each glyph. A processor implements the capture component, parser, comparison component, extraction component, and summary component).
Regarding claim 2, Pouyadou discloses matching the VDI for the symbol to a representation of a target VDI stored in a database (see 0018, 0038, 0029); and identifying the symbol based on the target VDI (see 0009, 0039).
Regarding claim 19, Pouyadou discloses a system comprising: a memory (see 0030; memory); and a processing device communicatively coupled to the memory (see 0030; processor), the processing device to: access a description of a symbol in a page description file for a document (see 0009; In accordance with another aspect of the exemplary embodiment, a system for extracting text from a page description language (PDL) of a document includes a capture component that captures a PDL file generated by a print driver for a printed document. The PDL file contains a set of drawing instructions for drawing a run of text formed from a series of glyphs in the printed document. A parser parses the PDL file to intercept the drawing instructions of each glyph. A comparison component compares the drawing instructions of each glyph with a database of glyph characteristics. An extraction component extracts a text character associated with each glyph when a match is found between the drawing instructions and the database of glyph characteristics based on the comparison of the drawing instructions with the database of glyph characteristics. A summary component generates a text summary of the extracted text characters associated with each glyph. A processor implements the capture component, parser, comparison component, extraction component, and summary component); identify, responsive to a character code failure, the symbol using a vectorized drawing instruction (VDI) for the symbol, wherein the character code failure comprises one of: an absence of a character code in the description of the symbol, or a bad character code in the symbol description of the symbol (see 0017, 0029, 0018; Aspects of the exemplary embodiment relate to a method, apparatus and computer-readable medium for applying local character recognition to typical office documents at or about the time they are printed. The method is general enough to be usable in other contexts where similar font technologies are used and when glyph-to-character information is lost…… It should be noted that, even if processed by single text drawing operators, glyphs can be stored in many formats. Typical formats include pure bitmaps, TrueType contours (which are basically a set of Bézier points used to draw the glyph curves), and/or PostScript instructions used in the PostScript PDL and its variations, such as PDF. In this case, the glyph curves are drawn using a subset of PostScript instructions embedded in the glyph description. Other variations include, for example, Adobe Type 2, CFF, and Chameleon fonts, whose formats may or may not be proprietary…. The exemplary method provides for direct character recognition from page description language (PDL) documents. The method may include parsing a PDL file which is created when a user initiates a print job, for example by clicking “print” from an application. The PDL document to be parsed may be composed of drawing instructions. For each visual representation of a symbol (glyph) that is processed, the character it represents is found by matching the glyph characteristics with a reference database. The reference database can be based on glyph lookup in a database of fonts. The font can use the TrueType/OpenType technology in which case the recognition is made directly by the quadratic splines control points of the glyph. In other cases, recognition is made by a semi-exact visual similarity.); and identify a text of the document using the identified symbol (see 0009, 0041; In accordance with another aspect of the exemplary embodiment, a system for extracting text from a page description language (PDL) of a document includes a capture component that captures a PDL file generated by a print driver for a printed document. The PDL file contains a set of drawing instructions for drawing a run of text formed from a series of glyphs in the printed document. A parser parses the PDL file to intercept the drawing instructions of each glyph. A comparison component compares the drawing instructions of each glyph with a database of glyph characteristics. An extraction component extracts a text character associated with each glyph when a match is found between the drawing instructions and the database of glyph characteristics based on the comparison of the drawing instructions with the database of glyph characteristics. A summary component generates a text summary of the extracted text characters associated with each glyph. A processor implements the capture component, parser, comparison component, extraction component, and summary component….. For each identified character, as the character is identified or once the parser system 44 finishes parsing PDL document 20 and all text characters have been extracted, the summary component 46 generates a text summary 92 of the extracted text characters associated with each glyph. For each match identified by the parser for a given text-related operator, the text summary component 46 adds the extracted character to the text summary and its location in the document page. The text summary 92 can be in a format which permits it to be read by a computing device, such as computing device 12, and/or a user operating the device 12. The text summary 92 may be stored as an Extensible Markup Language (“XML”) file.).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimedinvention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) in view of Chellapilla (IEEE: “Fast optical character recognition through glyph hashing for document conversion”).
Regarding claim 3, Pouyadou teaches all elements as mentioned above in claim 2. Pouyadou does not teach expressly computing a first hash value for the VDI for the symbol; and matching the first hash value with a second hash value for the target VDI stored in the database.
Chellapilla, in the same field of endeavor, teaches computing a first hash value for the VDI for the symbol (see section 3, 3.1, 3.1.1); and matching the first hash value with a second hash value for the target VDI stored in the database (see section 3, 3.1, 3.2.2., 3.3).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou to utilize the cited limitations as suggested by Chellapilla. The suggestion/motivation for doing so would have been to enhance the speed of the recognition (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou, while the teaching of Chellapilla continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claims 4-6 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) in view of Carbune et al (CL: “Fast Multi-language LSTM-based Online Handwriting Recognition”).
Regarding claims 4-6, Pouyadou teaches all elements as mentioned above in claim 1. Pouyadou does not teach expressly processing the VDI for the symbol using a neural network model to generate probabilities that the symbol corresponds to one or more candidate symbols; and using the generated probabilities to identify the symbol;
a first subnetwork processing the VDI for the symbol in a first direction, a second subnetwork processing the VDI for the symbol in a second direction, and a third subnetwork processing combined outputs of the first subnetwork and the second subnetwork;
a recurrent network, a long short-term memory network, a network with self-attention, or a transformer network.
Carbune, in the same field of endeavor, teaches processing the VDI for the symbol using a neural network model to generate probabilities that the symbol corresponds to one or more candidate symbols; and using the generated probabilities to identify the symbol (see section 2, 3);
a first subnetwork processing the VDI for the symbol in a first direction, a second subnetwork processing the VDI for the symbol in a second direction, and a third subnetwork processing combined outputs of the first subnetwork and the second subnetwork (see section 3);
a recurrent network, a long short-term memory network, a network with self-attention, or a transformer network (see section 3-4).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou to utilize the cited limitations as suggested by Carbune. The suggestion/motivation for doing so would have been to enhance the recognition times (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou, while the teaching of Carbune continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claims 7-8 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) with Carbune et al (CL: “Fast Multi-language LSTM-based Online Handwriting Recognition”) and further in view of Smith (IEEE: “An Overview of the Tesseract OCR Engine”).
Regarding claims 7-8, Pouyadou with Carbune teaches all elements as mentioned above in claim 4. Pouyadou with Carbune does not teach expressly selecting, based on the generated probabilities, a plurality of the candidate symbols; and selecting the symbol from the plurality of the candidate symbols, using at least one of: a degree of font similarity of the plurality of the candidate symbols and one or more reference symbols of the document, a degree of language similarity of the plurality of the candidate symbols and the one or more reference symbols of the document, or a degree of semantic similarity of the plurality of the candidate symbols and the one or more reference symbols of the document;
identifying the one or more reference symbols using one or more VDIs stored in a database; or identifying, with at least a threshold confidence, the one or more reference symbols using the neural network model.
Smith, in the same field of endeavor, teaches selecting, based on the generated probabilities, a plurality of the candidate symbols; and selecting the symbol from the plurality of the candidate symbols, using at least one of: a degree of font similarity of the plurality of the candidate symbols and one or more reference symbols of the document, a degree of language similarity of the plurality of the candidate symbols and the one or more reference symbols of the document, or a degree of semantic similarity of the plurality of the candidate symbols and the one or more reference symbols of the document (see section 5.2, 6, 7);
identifying the one or more reference symbols using one or more VDIs stored in a database; or identifying, with at least a threshold confidence, the one or more reference symbols using the neural network model (see section 5.2, 6, 7).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou with Carbune to utilize the cited limitations as suggested by Smith. The suggestion/motivation for doing so would have been to enhance the accuracy which is significantly improved (see Section 9). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou with Carbune, while the teaching of Smith continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) with Carbune et al (CL: “Fast Multi-language LSTM-based Online Handwriting Recognition”) and further in view of Ha (NE: “A Neural Representation of Sketch Drawings”).
Regarding claim 9, Pouyadou with Carbune teaches all elements as mentioned above in claim 4. Pouyadou with Carbune does not teach expressly teach (i) a training input comprising a VDI for a training symbol, and (i) a target output comprising identity of the training symbol.
Ha, in the same field of endeavor, teaches (i) a training input comprising a VDI for a training symbol (see section 1, intro, 3.2), and (i) a target output comprising identity of the training symbol (see section 3.1, abstract, 1, 4).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou with Carbune to utilize the cited limitations as suggested by Ha. The suggestion/motivation for doing so would have been to easily enable the model to learn, thus be more robust to recognize the target (see section 3.4). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou with Carbune, while the teaching of Ha continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) with Carbune et al (CL: “Fast Multi-language LSTM-based Online Handwriting Recognition”) and further in view of Nagy (IJDAR: “Document Analysis Systems that Improve with Use”).
Regarding claim 10, Pouyadou with Carbune teaches all elements as mentioned above in claim 4. Pouyadou with Carbune does not teach expressly teach determining that the symbol has been misidentified; obtaining a ground truth identity for the symbol; and re-training the neural network model using the ground truth identity for the symbol.
Nagy, in the same field of endeavor, teaches determining that the symbol has been misidentified (sees pg. 1); obtaining a ground truth identity for the symbol (see pg. 1-2); and re-training the neural network model using the ground truth identity for the symbol (see pg. 3, section 2.4).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou with Carbune to utilize the cited limitations as suggested by Nagy. The suggestion/motivation for doing so would have been to reduce the likelihood that the system will misidentify again and again (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou with Carbune, while the teaching of Nagy continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) in view of Chao et al (US 2003/0068099 A1).
Regarding claim 11, Pouyadou teaches all elements as mentioned above in claim 1. Pouyadou does not teach expressly teach at least one of: copying, using the identified text of the document, a first portion of the document to a new location within the document or to a new document; storing, using the identified text of the document, a second portion of the document; or printing, using the identified text of the document, a third portion of the document.
Chao, in the same field of endeavor, teaches at least one of: copying, using the identified text of the document, a first portion of the document to a new location within the document or to a new document (see abstract, 0031, 0022, 0028); storing, using the identified text of the document, a second portion of the document (see 0018, 0006); or printing, using the identified text of the document, a third portion of the document (see 0002).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou to utilize the cited limitations as suggested by Chao. The suggestion/motivation for doing so would have been to verify the accuracy of the extraction (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou, while the teaching of Chao continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Pouyadou et al (US 2017/0161595 A1) in view of Nagy (IJDAR: “Document Analysis Systems that Improve with Use”).
Regarding claim 20, Pouyadou teaches all elements as mentioned above in claim 19. Pouyadou does not teach expressly teach match the VDI for the symbol to a representation of a target VDI stored in a database; or process the VDI for the symbol using a neural network model to generate probabilities that the symbol corresponds to one or more candidate symbols.
Nagy, in the same field of endeavor, teaches match the VDI for the symbol to a representation of a target VDI stored in a database (see section 1, 2, 2.3, 2.4); or process the VDI for the symbol using a neural network model to generate probabilities that the symbol corresponds to one or more candidate symbols (see section 1, 2, 2.3, 2.4).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Pouyadou to utilize the cited limitations as suggested by Nagy. The suggestion/motivation for doing so would have been to reduce the likelihood that the system will misidentify again and again (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Pouyadou, while the teaching of Nagy continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Conclusion
Claims 1-11, 19-20 are rejected.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to EDWARD PARK. The examiner’s contact information is as follows:
Telephone: (571)270-1576 | Fax: 571.270.2576 | Edward.Park@uspto.gov
For email communications, please notate MPEP 502.03, which outlines procedures pertaining to communications via the internet and authorization. A sample authorization form is cited within MPEP 502.03, section II.
The examiner can normally be reached on M-F 9-6 CST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrew Moyer, can be reached on (571) 272-9523. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/EDWARD PARK/
Primary Examiner, Art Unit 2666