# US 8,786,607 B2 · Apparatus and method of generating personal fonts > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,786,607 B2 - Document ID: US8786607B2 - Status: Granted - Kind: B2 - Application number: 12/320,360 - Filed: 2009-01-23 - Published: 2009-07-30 - Granted: 2014-07-22 - Earliest claimed: 2008-01-28 - Assignees: Samsung Electronics Co., Ltd.; Korea Advanced Institute of Science - Applicants: Sung-Jung Cho; Jin-Hyung Kim; Dae-Hwan Kim - Inventors: Sung-Jung Cho; Jin-Hyung Kim; Dae-Hwan Kim - Web page: https://patentgrep.com/patent/US8786607B2 ## CPC - G06F 40/109: Font handling; Temporal or kinetic typography - G06F 18/28: Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries - G06T 11/23: using straight lines or curves - G06V 30/1914: {Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries, e.g. user dictionaries} - G06V 30/347: {Sampling; Contour coding; Stroke extraction} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract An apparatus and method of generating personal fonts is provided, which can modify a trace of handwriting inputted by a user into an elegant one and generate diverse personal fonts in accordance with user's style of handwriting. The apparatus for generating personal fonts includes a character input unit to receive an input of a character trace from a user, a character recognition unit to recognize a representative character corresponding to an input character from the input character trace, a representative trace generation unit to generate a representative trace expressing a trace of the representative character, and a trace modification unit to modify the trace of the input character by combining a weight value of the generated representative trace with the trace of the input character. ## Claims ### Claim 1 An apparatus for generating personal fonts, comprising: a character input unit to receive an input of a character trace, which has at least one feature point, from a user; a character recognition unit to recognize a representative character corresponding to an input character from the input character trace; a representative trace generation unit to generate a representative trace expressing a trace of the representative character; a trace modification unit to modify a position of the at least one feature point on the trace of the input character by combining a weight value of the generated representative trace with the trace of the input character using at least one processor to generate the personal fonts; and a correspondence analysis unit to analyze correspondence between constituent elements of the input character trace and the generated representative trace if the correspondence between the constituent elements is not provided by comparing the trace of the input character and the generated representative trace. ### Claim 2 (depends on 1) The apparatus of claim 1, wherein the character recognition unit selects and recognizes a character model having the minimum distance by comparing distances between the traces of the input characters and traces of the pre-stored character models. ### Claim 3 (depends on 1) The apparatus of claim 1, wherein the character recognition unit recognizes the representative character by performing pattern recognition with respect to the trace of the input character by using a Bayesian network. ### Claim 4 (depends on 1) The apparatus of claim 1, wherein the representative trace generation unit generates the representative trace by statistically modeling the constituent elements of the representative character and position relations among the constituent elements of the representative character by using a Bayesian network. ### Claim 5 (depends on 1) The apparatus of claim 1, wherein the correspondence analysis unit uses a dynamic programming matching method. ### Claim 6 (depends on 1) The apparatus of claim 1, wherein the trace modification unit linearly or non-linearly combines and generates corresponding constituent elements of a modified input character in accordance with correspondence between the trace of the input character and the trace of the representative character. ### Claim 7 (depends on 1) The apparatus of claim 1, further comprising a personal font storage unit to store a modified trace of the input character and the representative character. ### Claim 8 (depends on 1) The apparatus of claim 1, further comprising a personal font display unit to display a modified trace of the input character on a display screen. ### Claim 9 A method of generating personal fonts, comprising: receiving an input of a character trace, which has at least one feature point, from a user; recognizing a representative character corresponding to an input character from the input character trace; generating a representative trace expressing a trace of the representative character; modifying a position of the at least one feature point on the trace of the input character by combining a weight value of the generated representative trace with the trace of the input character to generate the personal fonts; and analyzing the correspondence if the correspondence between constituent elements of the input character trace and the representative character is not provided by comparing the trace of the input character and the generated representative trace, wherein the method is performed using at least one processor. ### Claim 10 (depends on 9) The method of claim 9, wherein the recognizing comprises selecting and recognizing a character model having the minimum distance by comparing distances between the traces of the input characters and traces of the pre-stored character models. ### Claim 11 (depends on 9) The method of claim 9, wherein the recognizing comprises recognizing the representative character by performing pattern recognition with respect to the trace of the input character by using a Bayesian network. ### Claim 12 (depends on 9) The method of claim 9, wherein the generating comprises generating the representative trace by statistically modeling constituent elements of the representative character and position relations among the constituent elements of the representative character by using a Bayesian network. ### Claim 13 (depends on 9) The method of claim 9, wherein the analyzing comprises using a dynamic programming matching method. ### Claim 14 (depends on 9) The method of claim 9, wherein the modifying comprises linearly or non-linearly combining and generating corresponding constituent elements of a modified input character trace in accordance with correspondence between the trace of the input character and the trace of the representative character. ### Claim 15 (depends on 9) The method of claim 9, further comprising storing a modified trace of the input character and the representative character. ### Claim 16 (depends on 9) The method of claim 9, further comprising displaying a modified trace of the input character on a display screen.