# US 2010/0169077 A1 · Method, system and computer readable recording medium for correcting ocr result > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2010/0169077 A1 - Document ID: US20100169077A1 - Status: Published application - Kind: A1 - Application number: 12/650,062 - Filed: 2009-12-30 - Published: 2010-07-01 - Earliest claimed: 2008-12-30 - Assignees: Nhn Corporation - Inventors: Byoung Seok YANG; Hee Cheol Seo; Do Gil Lee; Ki Joon Sung - Web page: https://patentgrep.com/patent/US20100169077A1 ## CPC - G06V 30/153: {using recognition of characters or words} - G06V 30/158: {using character size, text spacings or pitch estimation} - G06V 30/10: Character recognition - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Disclosed is a method, system and computer readable recording medium for correcting an OCR result. According to an exemplary embodiment of the present invention, there is provided a method for correcting an OCR result, the method including performing character recognition on content including character information using an OCR technique, removing extra carriage return information from the content, outputting the character recognition result, and correcting word spacing on the outputted result. ## Claims ### Claim 1 A method for correcting an optical character recognition (OCR) result, the method comprising: performing character recognition on a content using an OCR technique, the content comprising character information; removing extra carriage return information from the result of performing character recognition, wherein the extra carriage return information comprises carriage return information other than carriage return information indicating a start of a paragraph or an end of a paragraph; outputting a character recognition result after removing the extra carriage return information; and correcting word spacing on the outputted character recognition result. ### Claim 2 (depends on 1) The method of claim 1, wherein the carriage return information comprises an “Enter” character. ### Claim 3 (depends on 1) The method of claim 1, wherein performing character recognition comprises: analyzing the character information; and extracting at least one character corresponding to the character information. ### Claim 4 (depends on 1) The method of claim 1, further comprising: recognizing an indentation in the result of character recognition, and a carriage return preceding the indentation, wherein the carriage return information indicating the start of the paragraph or the end of the paragraph comprises information corresponding to the carriage return preceding the indentation. ### Claim 5 (depends on 4) The method of claim 4, wherein recognizing the indentation comprises recognizing a space 0.5 times larger than a height of a previous line at a starting point of a line. ### Claim 6 (depends on 1) The method of claim 1, further comprising: recognizing, in a character recognition result, a carriage return that follows a first line, wherein the first line ends without reaching a right side of the content, wherein the carriage return information indicating the start or the end of the paragraph comprises information corresponding to the carriage return that follows the first line. ### Claim 7 (depends on 6) The method of claim 6, wherein the right side of the content comprises a left side of a right margin disposed in the content. ### Claim 8 (depends on 1) The method of claim 1, wherein correcting word spacing comprises: associating a flag for determining word spacing with each character comprising the outputted character recognition result; and correcting the word spacing by determining whether to perform the word spacing for each character based on the flag. ### Claim 9 (depends on 1) The method of claim 1, wherein correcting word spacing is performed using a Hidden Markov Model (HMM). ### Claim 10 (depends on 1) The method of claim 1, further comprising: outputting a corrected OCR result based on the corrected word spacing and outputted character recognition result; and using the corrected OCR result for Internet search services. ### Claim 11 A system for correcting an optical character recognition (OCR) result, the system comprising: a character recognition unit to perform character recognition on a content using an OCR technique, the content comprising character information, to remove extra carriage return information from the result of performing character recognition, the extra carriage return information comprising carriage return information other than carriage return information indicating a start of a paragraph or an end of a paragraph, and to output a character recognition result after removal of the extra carriage return information; and a language model unit to receive the outputted character recognition result and to correct word spacing on the outputted character recognition result. ### Claim 12 (depends on 11) The system of claim 11, wherein the carriage return information comprises an “Enter” character. ### Claim 13 (depends on 11) The system of claim 11, wherein the character recognition unit is configured to analyze the character information and to extract at least one character corresponding to the character information. ### Claim 14 (depends on 11) The system of claim 11, wherein the character recognition unit is configured to recognize an indentation in the result of character recognition, and to recognize a carriage return preceding the indentation, and wherein the carriage return information indicating the start of the paragraph or the end of the paragraph comprises information corresponding to the carriage return preceding the indentation. ### Claim 15 (depends on 14) The system of claim 14, wherein recognizing the indentation comprises recognizing a space 0.5 times larger than a height of a previous line at a starting point of a line. ### Claim 16 (depends on 11) The system of claim 11, wherein the character recognition unit is configured to recognize, in a character recognition result, a carriage return that follows a first line, wherein the first line ends without reaching a right side of the content, and wherein the carriage return information indicating the start or the end of the paragraph comprises information corresponding to the carriage return that follows the first line. ### Claim 17 (depends on 16) The system of claim 16, wherein the right side of the content comprises a left side of a right margin disposed in the content. ### Claim 18 (depends on 11) The system of claim 11, wherein the language model unit is configured to associate a flag for determining word spacing with each character comprising in the outputted character recognition result, and to correct the word spacing by determining whether to perform the word spacing for each character based on the flag. ### Claim 19 (depends on 11) The system of claim 11, wherein the language model unit is configured to correct word spacing using a Hidden Markov Model (HMM). ### Claim 20 (depends on 11) The system of claim 11, further comprising a search service providing unit, wherein the language model unit is configured to output a corrected OCR result based on the corrected word spacing and outputted character recognition result, and the search service providing unit is configured to use the corrected OCR result for Internet search services. ### Claim 21 A non-transitory computer-readable recording medium comprising an executable program which, when executed, performs the steps of: performing character recognition on a content using an OCR technique, the content comprising character information; removing extra carriage return information from the result of character recognition, wherein the extra carriage return information comprises carriage return information other than carriage return information indicating a start of a paragraph or an end of a paragraph; outputting a character recognition result after removing the extra carriage return information; and correcting word spacing on the outputted character recognition result.