# US 8,787,671 B2 · Character recognition preprocessing method and apparatus > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,787,671 B2 - Document ID: US8787671B2 - Status: Granted - Kind: B2 - Application number: 13/052,851 - Filed: 2011-03-21 - Published: 2011-09-22 - Granted: 2014-07-22 - Earliest claimed: 2010-03-19 - Assignees: Samsung Electronics Co., Ltd.; Industry Foundation of Chonnam National University - Applicants: Hee-Bum Ahn; Jong-Hyun Park; Soo-Hyung Kim; Hyung-Jung Yang; Guee-Sang Lee - Inventors: Hee-Bum Ahn; Jong-Hyun Park; Soo-Hyung Kim; Hyung-Jung Yang; Guee-Sang Lee - Web page: https://patentgrep.com/patent/US8787671B2 ## CPC - G06V 30/1478: {of characters or characters lines} - G06V 30/166: Normalisation of pattern dimensions - G06V 30/10: Character recognition - G06V 30/1607: {Correcting image deformation, e.g. trapezoidal deformation caused by perspective} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Disclosed is a character recognition preprocessing method and apparatus for correcting a nonlinear character string into a linear character string. A binarized character string region is divided into character regions on a character-by-character basis. Upper and lower feature points of each character region are derived, and an upper boundary line, which is a curve connecting the upper feature points of the character regions, and a lower boundary line, which is a curve connecting the lower feature points of the character regions, are generated by applying cubic spline interpolation. Nonlinearity is corrected through adaptive region enlargement by using the maximum horizontal length and the maximum height of the divided character regions. ## Claims ### Claim 1 A character recognition preprocessing method in a character recognition preprocessing apparatus, the method comprising the steps of: detecting a character string region including a character string from an image, and dividing the character string region into character regions corresponding to respective characters constituting the character string in the character string region; extracting upper feature points and lower feature points for each character region; generating an upper boundary line connecting the upper feature points and a lower boundary line connecting the lower feature points; determining a layout of a reference region; dividing a real character string region, which is formed by the upper and lower boundary lines and includes the character string, into correction character regions corresponding to the respective characters, based on the upper and lower feature points; mapping an upper boundary line of each correction character region to an upper reference straight line of the reference region by interpolation, and mapping a lower boundary line of each correction character region to a lower reference straight line of the reference region by interpolation; and applying an interpolation value used in the mapping to respective pixels included in each correction character region, thereby mapping each character region to the reference region. ### Claim 2 (depends on 1) The method as claimed in claim 1, wherein the upper feature points include both end points of an upper boundary of the corresponding character region, and the lower feature points include both end points of a lower boundary of the corresponding character region. ### Claim 3 (depends on 1) The method as claimed in claim 1, wherein Red, Green, Blue (RGB) values of the pixels included in each correction character region are identical to RGB values of corresponding pixels of the reference region. ### Claim 4 (depends on 1) The method as claimed in claim 1, wherein the reference region comprises a rectangular region, and horizontal and vertical lengths of the reference region are determined using a horizontal length and a height of each character region. ### Claim 5 (depends on 1) The method as claimed in claim 1, wherein the upper and lower boundary lines are generated by cubic spline interpolation. ### Claim 6 A character recognition preprocessing apparatus comprising: a camera for generating an image; and a character recognition preprocessing unit for detecting a character string region including a character string from an image, dividing the character string region into character regions corresponding to respective characters constituting the character string in the character string region, extracting upper feature points and lower feature points for each character region, generating an upper boundary line connecting the upper feature points and a lower boundary line connecting the lower feature points, determining a layout of a reference region, dividing a real character string region, which is formed by the upper and lower boundary lines and includes the character string, into correction character regions corresponding to the respective characters, based on the upper and lower feature points, mapping an upper boundary line of each correction character region to an upper reference straight line of the reference region by interpolation, mapping a lower boundary line of each correction character region to a lower reference straight line of the reference region by interpolation, and applying an interpolation value used in the mapping to respective pixels included in each correction character region, thereby mapping each character region to the reference region. ### Claim 7 (depends on 6) The apparatus as claimed in claim 6, wherein the upper feature points include both end points of an upper boundary of the corresponding character region, and the lower feature points include both end points of a lower boundary of the corresponding character region. ### Claim 8 (depends on 6) The apparatus as claimed in claim 6, wherein Red, Green, Blue (RGB) values of the pixels included in each correction character region are identical to RGB values of corresponding pixels of the reference region. ### Claim 9 (depends on 6) The apparatus as claimed in claim 6, wherein the reference region comprises a rectangular region, and horizontal and vertical lengths of the reference region are determined using a horizontal length and a height of each character region. ### Claim 10 (depends on 6) The apparatus as claimed in claim 6, wherein the upper and lower boundary lines are generated by cubic spline interpolation.