WO2017099555A1 - 시간분할 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 - Google Patents
시간분할 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 Download PDFInfo
- Publication number
- WO2017099555A1 WO2017099555A1 PCT/KR2016/014556 KR2016014556W WO2017099555A1 WO 2017099555 A1 WO2017099555 A1 WO 2017099555A1 KR 2016014556 W KR2016014556 W KR 2016014556W WO 2017099555 A1 WO2017099555 A1 WO 2017099555A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- segment
- block
- information
- handwritten signature
- area
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/333—Preprocessing; Feature extraction
- G06V30/347—Sampling; Contour coding; Stroke extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/45—Structures or tools for the administration of authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
- G06F3/041—Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0487—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
- G06F3/0488—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/30—Writer recognition; Reading and verifying signatures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/30—Writer recognition; Reading and verifying signatures
- G06V40/37—Writer recognition; Reading and verifying signatures based only on signature signals such as velocity or pressure, e.g. dynamic signature recognition
- G06V40/394—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/50—Maintenance of biometric data or enrolment thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0487—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
- G06F3/0488—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures
- G06F3/04883—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures for inputting data by handwriting, e.g. gesture or text
Definitions
- the present invention relates to a system and method for handwriting signature authentication, and more particularly, time to perform handwritten signature authentication by handwritten signature characteristic information based on a segment block including a segment divided into predetermined time units during handwriting signature.
- a split handwritten signature authentication system and method thereof are provided.
- the user In the off-line face-to-face, the user is authenticated using a resident registration card or a driver's license.
- a resident registration card or a driver's license In the off-line face-to-face, the user is authenticated using a resident registration card or a driver's license.
- a driver's license There is a demand for a method of checking whether the user using the service is the user in the online face-to-face.
- typical technologies include i-fin technology, SMS identity authentication technology, ARS identity authentication technology using ARS system, and identity authentication technology using digital signature (digital signature).
- SMS self-authentication technology sends an authentication number through a Short Message Service (SMS) message of a mobile communication terminal called a mobile phone, a mobile phone, a smartphone, and the like.
- SMS Short Message Service
- the above technologies have a risk that the mobile communication terminal may be lost or stolen by a third party when information is leaked.
- Handwriting signature authentication technology has an image comparison method for checking the conformity of the shape of the handwritten signature image, and the behavior characteristic comparison method that compares the handwriting signature characteristic data on the signature behavior characteristics of the signer at the time of signing.
- a handwritten signature authentication system to which an image comparison method is applied may similarly mimic a handwritten signature when only a third party imitates only the image shape of the handwritten signature, thereby making it easy to steal the third party.
- the behavior characteristic comparison method is applied to the handwritten signature authentication system rather than the image comparison method.
- Handwritten Signature Verification System with Behavior Characteristics Comparison Method When handwriting is signed, characteristics such as total pressure, speed, intersection point, and inflection point angle generated by signer's signature habit are extracted and stored. However, even when the behavior characteristic comparison method is applied, when a third party imitates the handwritten signature to correspond to the image of the handwritten signature, the behavior characteristic is frequently copied to some extent. Even in the conventional handwritten signature authentication system, even if the shape is completely different, if only the pattern of the overall behavioral characteristics is similar, there is a problem that it can be determined that it matches the registered signature signature behavioral characteristic information.
- the method of dividing the handwritten signature in the handwritten signature authentication system more accurately, and thus improving the security by increasing the authentication accuracy while having a higher recognition rate when handwritten signature authentication is required.
- an object of the present invention is to register a handwritten signature by storing handwritten signature characteristic information based on a time division segment block including segments divided by a predetermined time unit at the time of handwriting signature, and user when requesting handwritten signature authentication.
- the present invention provides a time division segment block-based handwritten signature authentication system and method for performing signature authentication.
- a time division segment block-based handwritten signature authentication system comprises: handwriting touch data including position data and pressure data on touched positions corresponding to a handwritten signature signed by a signer A handwritten signature input unit including a touch input unit for outputting signature input data; A register for registering handwritten signature characteristic information for each signer; And counting and storing the time when the handwritten signature input data including the touch data and the handwritten signature input data, which are input from the handwritten signature input unit, are input, and the segment reference time (t 1) by the handwritten signature input data and the counted time.
- Recognize the segment in units generate the handwritten signature image and segment image by recognizing the handwritten signature, generate a handwritten signature block including the handwritten signature and a segment block including the segment image, the segment block and handwritten Collecting signature characteristic information ( ⁇ ) including information of each signature block and correlation information between the segment blocks, and correlation information between the segment block and the handwritten signature block, and mapping the signature signature information to the signer identification information of the signer; After registering in the register, request for handwriting signature verification
- the signature information of the signer who collects handwritten signature characteristic information ( ⁇ ) including correlation information between the handwritten signature block and the segment blocks from touch data input through the touch input unit of the handwritten signature input unit and generates a handwritten signature authentication.
- a handwritten signature authentication unit for performing handwritten signature authentication based on a handwritten signature segment according to the degree of conformity compared with handwritten signature characteristic information ( ⁇ ) collected by loading registered handwritten signature characteristic information ( ⁇ ') corresponding to. It is done.
- the handwritten signature authenticator counts and stores the time when handwritten signature input data and handwritten signature input data inputted from the handwritten signature input data inputted from the touch input unit of the handwritten signature input unit, and stores the time. Recognizing the segment in units of the segment reference time (t 1 ) according to the time to be generated, recognizes the handwritten signature to generate a handwritten signature image and segment image, and includes a handwritten signature block including the handwritten signature image and the segment image After generating the segment block, the handwritten signature full block feature information (Q) which is the feature information of the handwritten signature block including the handwritten signature image for the handwritten signature, and the whole segment block which is the characteristic information of the handwritten signature segments constituting the handwritten signature Segment block feature information (V) of the number, the number A handwritten signature feature extraction unit for extracting handwritten signature property information ( ⁇ ) including block correlation feature information (C) by correlation between the original signature block and the segment blocks; Handwritten signature segment block authentication that compares handwritten signature characteristic information ( ⁇ ) extracted through the handwritten signature characteristic extraction unit
- the handwritten signature characteristic extraction unit may include a handwritten signature start detection unit for detecting the start of a handwritten signature from the touch data; A handwritten signature end detection unit for detecting the end of the handwritten signature as the handwritten signature end point when the touch data is not input for a predetermined time while the touch data is input; Counting and storing the touch data input from the touch data and the input time of the touch data, and recognizing the segment in units of the segment reference time (t 1 ) by the touch data and the counted time, and recognizing the handwritten signature A time division segment detection unit generating and outputting a handwritten signature image and a segment image; A segment counting unit for counting the number of segments detected by the time division segment detecting unit; The segment receives the image segment blocks containing the segment image (s i) to generate, and the generated segment blocks (s i) the characteristic information of the segment block characteristic information (v i) a segmented block generation and to generate for the characterized information (v i) segment blocks characterized in detecting the full segment block
- the handwritten signature full block feature detection unit calculates the area of the handwritten signature block S to further generate and output the handwritten signature full block area information space S , and the segment block feature detection unit receives the segment image.
- a segment block generator configured to generate and output the segment block s i including the segment image; It said segment blocks (s i) for receiving input, the location of each corner of the segment block segment block location information (p i) detected by the output segment block edge position detecting unit for a; Said segment blocks (s i) and the segment block location information (p i) said segment blocks for receiving at least one of (i s) segment block area information by calculating the area of the ( A segment block area characteristic detector for generating and outputting a); The handwritten signature full block area information (space S ) and the segment block area information (from the handwritten signature full block feature detection unit) An area ratio characteristic detection unit configured to generate a segment block area ratio information ⁇ i by calculating a segment block area ratio to the entire block area of the handwritten signature; And the segment
- the block is a rectangular block, wherein the segment block generating unit generates a rectangular segment block including the segment by passing through the highest point, the lowest point, the leftmost point, and the rightmost point of the corresponding segment. .
- the handwritten signature full block feature detection unit calculates the area of the handwritten signature block S , further generates and outputs handwritten signature full block area information space S
- the segment block correlation detection unit comprises a segment block (s). i) and positioned adjacent to the adjacent segment block (s j), the segment block among the (s i) and includes and cross (overlap) adjacent segment blocks to form a region (s j) comprising if determining that there is, and, and An intersection area detector for calculating an area of the intersection area and outputting intersection area information ⁇ ij ;
- the handwritten signature total block area information (space S ), segment block area information ( ) And the intersection area information ( ⁇ ij ) are input, the ratio of the intersection area ( ⁇ ij ) to the entire block area of the signature is calculated, and the signature block intersection area ratio information (r ij ) is generated.
- a cross-area ratio information cross-area ratio detector for generating a cross-section ratio information; Include block segments (s i) of the adjacent block segments (s j), and whether cross that information whether the segment blocks containing related information (O ij) for generating and outputting segments of the block are included in the reference relationship detection; Block segments (s i) the segment position relationship detection unit generates the location information (POS ij) the information block segments the position of the adjacent block segment (s j) on the basis of the output; Segment blocks (s i) of the adjacent segment block basis (s j) the segment blocks (s i) which side edges and cross whether representing position information of the segment block edge position relation information corner generating and outputting a (EDGE ij) of A positional relationship detector; And the intersection area information (
- the time division segment detector in response to the input of the touch data input in real time, characterized in that the segments are sequentially recognized by the segment reference time (t 1 ) by the input touch data.
- the time division segment detection unit counts and stores the input touch data and the input time of the touch data, and when a handwritten signature image is detected by the touch data, the time division segment detection unit is based on the handwritten signature recording time required to record the handwritten signature image.
- the segment is divided into units of the segment reference time t 1 to recognize the segment.
- the segment block generation unit may include a sub segment block for a sub segment generated by a segment divided in the generated segment block ( ) And bacony area blocks ( ) Is further detected, and the segment block edge position detector is further configured to provide the sub-segment block ( Subsegment block location information for ) And the bacony region block ( Area region location information () ) Is further detected and outputted, and the segment block area characteristic detection unit comprises: a sub-segment block ( ) And bacony area blocks ( Subarea block area information ( ) And vacancy area block area information ( ) Is generated and output, and the area ratio characteristic detection unit includes a plurality of sub-segment blocks (s i ) in one segment block s i .
- the segment block area ( Subsegment block area ) The area of the vacancy area block relative to the ratio and the area of the segment block ( ), Calculate the ratio of subsegment block area ratio information ( ) And bacony area block area ratio information ( ), And the segment block feature information generation unit includes a sub segment block ( Subsegment block location information for ), Bacony area block ( Area region location information () ), Subsegment block area information ( ), Baconsea area block area information ( ), Subsegment block area ratio information ( ) And bacony area block area ratio information ( ) To generate and output all segment block feature information (V) further reflected.
- a time division segment block-based handwritten signature authentication method includes: counting a time when handwritten signature input data and handwritten signature input data inputted from a handwritten signature input unit and including touch data are counted; Store together, and recognize the segment by the segment reference time (t 1 ) unit by the handwritten signature input data and the counted time, the handwritten signature is generated to generate a handwritten signature image and segment image, the handwriting including the handwritten signature A handwriting block generating a signature block and a segment block including the segment, the handwriting including information of each of the segment block and the handwritten signature block, correlation information between the segment blocks, and correlation information between the segment block and the handwritten signature block; Collect signature attribute information ( ⁇ ) Registration process by mapping the signer identification information registered in the registration section of the nominee; And handwritten signature characteristic information ( ⁇ ) including correlation information between the handwritten signature block and the segment blocks from touch data input through the touch input unit of the handwritten signature input unit when a handwritten signature authentication request is generated, and
- Handwritten signature authentication that performs handwritten signature authentication based on the handwritten signature segment according to the matching degree by comparing the registered handwritten signature property information ( ⁇ ') corresponding to the requested signer's identification information with the handwritten signature property information ( ⁇ ) collected. It characterized in that it comprises a process.
- the handwritten signature registration process may include: a registration request monitoring step of monitoring whether a handwritten signature registration request is generated; Signer identification information obtaining step of obtaining signer identification information of a signer to be registered when a handwritten signature registration request is generated; A handwritten signature characteristic data acquiring step of acquiring the handwritten signature characteristic information ( ⁇ ) from touch data input from a touch input unit for a handwritten signature by a signer; And a handwritten signature registration step of mapping the handwritten signature characteristic information to the signer identification information and storing the registered signature property information in a register.
- the handwritten signature authentication process includes: a handwritten signature authentication request monitoring step of monitoring whether a handwritten signature authentication request is generated; A signer identification information acquiring step of acquiring signer identification information of the signer when a handwritten signature authentication request is generated; Obtaining handwritten signature characteristic information for obtaining the handwritten signature characteristic information ( ⁇ ) from touch data input from a touch input unit with respect to the handwritten signature by the signer; A registration handwriting signature property information loading step of loading registration handwriting signature property information ( ⁇ ') registered in advance corresponding to the obtained signer identification information; And a handwritten signature authentication step of comparing the obtained handwritten signature characteristic information ( ⁇ ) and the loaded registered handwritten signature characteristic information ( ⁇ ') to perform handwritten signature authentication and output the authentication result. .
- the handwriting signature characteristic information ( ⁇ ) acquiring step may include: a handwriting signature tracking step of initiating tracking of the handwritten signature from touch data of the handwritten signature input data inputted from the handwritten signature input unit; When the handwriting signature tracking is started, the time when the handwritten signature input data and the handwritten signature input data input through the time division segment detection unit are counted and stored together, and the segment reference time ( a segment detection step of recognizing and outputting segments in units of t 1 ); A segment count step of counting the number of segments detected by the time division segment detector; A segment block receiving the segment image, generating a segment block s i including the segment image, and generating and outputting segment block feature information v i , which is feature information about the generated segment block s i .
- Handwritten signature block detection step of generating a handwritten signature block S including the obtained handwritten signature image, and generating and outputting handwritten signature full block feature information Q which is characteristic information of the handwritten signature block S ;
- a segment block correlation detection step of generating and outputting block correlation feature information (C) which is information according to a correlation between the entire handwritten signature block and the segment blocks; And generate full segment block feature information (V) including segment block feature information (v i ) for all segments, wherein the handwritten signature full block feature information (Q), the full segment block feature information (V), and And handwriting signature characteristic information generation step of generating and outputting handwritten signature characteristic information ( ⁇ ) including the block correlation characteristic information (C).
- the handwritten signature full block feature detection step may further include generating the handwritten signature full block area information space S by calculating the area of the handwritten signature block S , and outputting the segmented block feature.
- a segment block generation step of receiving an input and generating and outputting the segment block s i including the segment image;
- a segment block edge position detecting step of receiving the segment block s i and detecting and outputting segment block position information pi which is position information of each corner of the segment block; Said segment blocks (s i) and the segment block location information (p i) said segment blocks for receiving at least one of (i s) segment block area information by calculating the area of the ( Generating and outputting a segment block area characteristic;
- the handwritten signature total block area information (space S ) and the segment block area information detected in the handwritten signature full block feature detection step ( ) Segment block area ( S ) compared to the total block area ( S ) of the handwritten signature Calculating a ratio to generate and output segment block area ratio information ⁇
- the block is a rectangular block, and in the segment block generation step, the segment block generator generates a rectangular segment block including the segment by passing through the highest point, the lowest point, the leftmost point, and the rightmost point of the segment. Characterized in that.
- the handwriting signature full block feature detection step may further include generating and outputting handwritten signature total block area information space S by calculating the area of the handwriting signature block S
- the segment block correlation detection step may include: segment block If judges adjacent that segment block (s j) to form a (s i) and positioned adjacent to the adjacent segment block (s j) s of the segment blocks including (s i), and and the intersection (overlap) area, and An intersection area detection step of calculating the area of the inclusion and intersection areas and outputting intersection area information ⁇ ij ;
- the handwritten signature total block area information (space S ), segment block area information ( ) And the intersection area information ( ⁇ ij ) are input, the ratio of the intersection area ( ⁇ ij ) to the entire block area of the signature is calculated, and the signature block intersection area ratio information (r ij ) is generated.
- Detection include block segments (s i) of the adjacent block segments (s j), and whether cross that information whether the segment blocks containing related information (O ij) for generating and outputting segments of the block are included in the reference phase relation; Block segments (s i) of the adjacent segment blocks based on position information of the segment block position information generating and outputting the segment (POS ij) of (s j) the positional relationship detecting step; Segment blocks (s i) of the adjacent segment block basis (s j) the segment blocks (s i) which side edges and cross whether representing position information of the segment block edge position relation information corner generating and outputting a (EDGE ij) of Positional relationship detection step; And the intersection area (s i) of the adjacent block segments (s j), and whether cross that information whether the segment blocks containing related information (O ij) for generating and outputting segments of the block are included in the reference phase relation; Block segments (s i) of the adjacent segment blocks based on position information of the segment block position information
- the segment block generating step may include: a segment block detecting step of receiving the segment image and generating and outputting the segment block s i including the segment image; A sub-segment block for the sub-segment generated by the segment divided within the generated segment block ( Detecting a sub segment block; And a vacancy region block, which is an empty space region generated by the sub segment block, in the generated segment block.
- segment blocks (s i) for receiving input, the location of each corner of the segment block segment block location information (p i Generating segment block edge position information for detecting and outputting;
- the sub-segment block Subsegment block location information for Generating sub-segment block edge position information for detecting and outputting;
- the bacony region block Area region location information () (B) generating and outputting the basic region block edge position information, wherein the segment block area characteristic detecting step includes at least one of the segment block s i and the segment block position information p i .
- Segment block area information is calculated by calculating an area of the segment block s i . Generating segment block area characteristic information for outputting the generated segment data;
- the sub-segment block Subarea block area information ( Generating sub-segment block area specifying information for generating and outputting the information;
- the bacony region block Calculate the area of the And generating vacancy area block area characteristic information, wherein the area ratio characteristic detecting step comprises: handwritten signature total block area information space S detected in the handwritten signature full block feature detection step and the segment;
- the segment block area Vacancy area block area compared to Calculate the ratio of (B) generating and outputting the basic region block area ratio feature information
- the generating of the segment block feature information includes: ) And bacony area blocks ( Subsegment block location information for ), Bacony area block ( Area region location information () ), Subsegment block area information ( ), Baconsea area block area information ( ), Subsegment block area ratio information ( ) And bacony area block area ratio information ( ) To generate and output all segment block feature information (V) further reflected.
- the present invention divides segments by dividing them into predetermined time units according to the handwritten signature time of the handwritten signature recorded by the signer at the time of the handwritten signature, and each of the features and blocks of the segment block including the divided time division segments and the entire handwritten signature block.
- the handwritten signature authentication element can be specialized to perform the authentication, thereby increasing the recognition rate of the handwritten signature.
- FIG. 1 is a diagram illustrating a configuration of a time division segment block based handwritten signature authentication system according to the present invention.
- FIG. 2 is a diagram illustrating a configuration of a handwritten signature characteristic acquisition unit of a time division segment block based handwritten signature authentication system according to the present invention.
- FIG. 3 is a diagram for describing a time division segment block formation method and time division segment block feature information elements of a handwritten signature according to an embodiment of the present invention.
- FIG. 4 is a diagram for describing a method of forming a time division segment block and a time division segment block feature information element of FIG. 3 in more detail.
- FIG. 5 is a diagram illustrating a configuration of a time division segment block feature detection unit of a handwritten signature characteristic acquisition unit according to the present invention.
- FIG. 6 is a diagram illustrating a detailed configuration of a segment block correlation detection unit of a handwritten signature characteristic acquisition unit according to the present invention.
- FIG. 7 is a diagram illustrating a method of generating segment block inclusion relationship information, which is one of correlation information between segment blocks, according to an exemplary embodiment.
- FIG. 8 is a diagram for describing a method of generating segment block position relation information, which is one of correlation information among handwritten signature segment blocks, according to an exemplary embodiment.
- FIG. 9 is a diagram for describing a method of generating segment block edge positional relationship information, which is one of correlation information among handwritten signature segment blocks, according to an exemplary embodiment.
- FIG. 10 is a flowchart illustrating a time division segment block based handwritten signature authentication method according to the present invention.
- FIG. 11 is a flowchart illustrating a method of collecting handwritten signature data of a time division segment block based handwritten signature authentication method according to the present invention.
- FIG. 12 is a flowchart illustrating a method of generating a time division segment block according to the first embodiment of the present invention.
- FIG. 13 is a flowchart illustrating a method of generating a time division segment block according to a second embodiment of the present invention.
- the handwritten signature time division segment corresponds to the handwritten signature time at which the signer signs the handwriting signature at the time of handwriting signature.
- the segment block in the present invention means a block of the polygon including the segment
- FIG. 1 is a diagram illustrating a configuration of a time division segment block based handwritten signature authentication system according to the present invention.
- a time division segment block-based handwritten signature authentication system includes a register 100, a handwritten signature inputter 400, and a handwritten signature authenticator 500, and according to an embodiment, the inputter 200. ), And further includes an output unit (300).
- the registration unit 100 may be a mobile communication terminal called a personal computer (PC), a hard disk such as a notebook, a portable memory such as a universal serial bus (USB), a security token, a mobile phone, a mobile phone, a smartphone, or the like. It can be configured in a storage medium such as a SIM card, a micro SD card of a mobile communication terminal, a trust zone of a mobile communication terminal, a hard disk of an online server, etc., and stores handwritten signature characteristic information ( ⁇ ) for a handwritten signature.
- PC personal computer
- a hard disk such as a notebook
- USB universal serial bus
- security token a mobile phone
- a mobile phone a mobile phone, a smartphone, or the like.
- It can be configured in a storage medium such as a SIM card, a micro SD card of a mobile communication terminal, a trust zone of a mobile communication terminal, a hard disk of an online server, etc., and stores handwritten signature characteristic information ( ⁇ ) for a handwritten signature.
- the handwritten signature characteristic information ( ⁇ ) includes handwritten signature full block feature information (Q), full segment block feature information (V), and block correlation feature information (C). Detailed information included in the above information will be described in detail with reference to FIGS. 2 to 5 to be described later.
- the input unit 200 includes a key input device having a plurality of keys capable of generating a plurality of commands and outputting key data (key signals) for a pressed key, and is integrally formed with a screen to provide position data of a touched position. It may be a touch pad for outputting, a receiving device for receiving data through wired or wireless communication with an external device, and the like.
- a handwritten signature authenticator 500 may issue a command such as a handwritten signature registration command or a handwritten signature authentication command by a user's request. Will output
- the input unit 200 may be a force terminal, a payment terminal, a mobile communication terminal, or the like, when the handwritten signature authenticator 500 is configured in the form of a server.
- the output unit 300 is configured to output the handwritten signature image, the handwritten signature characteristic information, and the handwritten signature authentication result obtained when the handwritten signature authenticator 500 is configured in the mobile communication terminal. It may be a display device such as a liquid crystal display (LCD), and when the handwritten signature authenticator 500 is configured in a server, a short message service (SMS) message indicating a handwritten signature authentication result.
- SMS short message service
- Message transmission server for transmitting mobile communication messages such as, long message service (LMS) message, multimedia message service (MMS) message, application server for transmitting push message, email server, It may be a mobile communication terminal for receiving and displaying the authentication result.
- the handwriting signature input unit 400 is composed of a personal terminal, a personal computer, a mobile communication terminal, a POS, a payment terminal, etc. of a user terminal and a member receiving the handwriting signature, or connected to a separate device for a handwriting signature provided by a user.
- Outputting handwritten signature input data for acquiring a handwritten signature image may include at least one of the scan unit 410 and the touch input unit 420.
- the touch input unit 420 may be necessarily included.
- the touch input unit 420 may track a handwritten signature, and may be a touch pad, a touch screen, a smart pen, and the like that can collect image characteristic and behavior characteristic information about the handwritten signature and the handwritten signature segment.
- the scan unit 410 scans the paper on which the handwritten signature is written and outputs the scan data to the handwritten signature authenticator 500.
- the touch input unit 420 may be a touch pad, a touch screen, or the like.
- the handwriting signature authentication unit may include touch data including continuous position data and pressure data for a handwritten signature input by a user by hand. Output as 500.
- the handwritten signature authenticator 500 includes a controller 510, a handwritten signature characteristic extractor 520, and a handwritten signature segment block authenticator 560.
- the handwritten signature authenticator 500 may be configured based on an application in a mobile communication terminal and a computer, or may be configured in one or more of an application and a web server based on a server side, and may be configured in a POS and a payment terminal. It may be configured in a form. According to the configuration of the present invention, the configuration of the application server, the web server, and the firmware device based on the application, the firmware, or the web server itself is obvious to those skilled in the art, and thus the detailed description thereof will be omitted.
- the controller 510 controls the overall operation of the handwritten signature authenticator 500.
- the controller 510 determines whether the command received from the input unit 200 is handwritten signature registration or handwritten signature authentication, and controls segment-based handwritten signature registration and handwritten signature authentication operations according to the present invention according to the determined command.
- the control result is output to the output unit 300.
- the handwritten signature characteristic extractor 520 extracts and outputs the time signature segment block-based handwritten signature characteristic information ( ⁇ ) from the handwritten signature input data input from the touch input unit 420 of the handwritten signature input unit 400.
- the handwritten signature characteristic extractor 520 includes a handwritten signature tracker 530, a handwritten signature image acquirer 540, and a handwritten signature characteristic acquirer 550.
- the handwritten signature tracking unit 530 detects continuous position data from the touch data output from the touch input unit 420 of the handwritten signature input unit 400 and provides it to the handwritten signature image acquisition unit 540.
- the handwritten signature image acquisition unit 540 receives the position data from the handwritten signature input unit 400 or the scan data input from the handwritten signature tracker 530, and obtains and outputs a handwritten signature image from the scan data and the position data.
- the handwritten signature image acquisition unit 540 may acquire the traced handwritten signature image through the scan unit 410, and generate and track location data input in real time through the touch input unit 420 and the handwritten signature tracker 530. You could do it.
- the handwritten signature characteristic acquisition unit 550 receives touch data continuously output from the touch input unit 420 of the handwritten signature input unit 400, counts the time of the input touch data, and schedules the input touch data. Split by time unit to identify the handwritten segment, count the number n of the handwritten segment identified, and generate a handwritten segment image for the handwritten segment identified.
- the handwriting signature property acquisition unit 550 may determine a block (hereinafter, referred to as a “segment block”) of a polygon (hereinafter, assuming a “square”) including the corresponding segment image for each of the handwritten signature segment images generated.
- Segment block a block of a polygon (hereinafter, assuming a “square”) including the corresponding segment image for each of the handwritten signature segment images generated.
- V segment block feature information
- Q Generate handwritten signature full block
- a detailed configuration of the handwritten signature characteristic acquisition unit 550 will be described in detail with reference to FIGS. 2, 3, and 5 to be described later.
- the handwritten signature segment block authentication unit 560 receives the signer identification information from the handwritten signature input unit 400 when a handwritten signature authentication command is generated from the control unit 510, and handwritten signature characteristic information ( ⁇ ) from the handwritten signature characteristic extraction unit 520. ), The registered handwritten signature characteristic information ⁇ 'corresponding to the signer identification information is loaded from the registration unit 100, and the registered handwritten signature characteristic information ⁇ ' and the handwritten signature characteristic extractor 520 are loaded. Handwritten signature authentication is performed by comparing inputted handwritten signature characteristic information ( ⁇ ) to compare whether or not a predetermined criterion is matched.
- the handwritten signature segment block authentication unit 560 extracts the handwritten signature property information Q 'of the handwritten signature property information ⁇ ' and the handwritten signature property information extracted through the handwritten signature property extractor 520. Compares the handwritten signature full block feature information Q of ( ⁇ ), compares the loaded full segment block feature information V 'with the extracted full segment block feature information V, and loads the loaded block correlation feature information (C ') and the extracted block correlation feature information (C) are compared to determine whether or not a predetermined criterion is equal to or greater than to perform handwritten signature authentication.
- FIG. 2 is a diagram illustrating a structure of a handwritten signature characteristic acquisition unit of a time division segment block-based handwritten signature authentication system according to the present invention
- FIG. 3 is a method of forming a segment block and a segment block feature according to an embodiment of the present invention.
- FIG. 4 is a diagram for describing information elements
- FIG. 4 is a diagram for describing the method of forming a time division segment block and the time division segment block feature information element of FIG. 3 in more detail.
- the handwritten signature characteristic acquisition unit 550 will be described with reference to FIGS. 2 to 4.
- the handwriting signature characteristic acquisition unit 550 may include a handwriting signature start detection unit 610, a handwriting signature end detection unit 620, a time division segment detection unit 630, a segment count unit 640, a segment block feature detection unit 650, and a handwriting signature.
- a full block feature detector 660, a segment block correlation detector 670, and a handwritten signature block characteristic information generator 680 are included.
- the handwriting signature start detection unit 610 receives continuous handwritten signature input data according to the handwriting signature from the touch input unit 400. Receive input.
- the handwriting signature start detection unit 610 detects the handwriting signature start point 3 as shown in FIG. 3.
- the handwritten signature start point 3 is the start point of the first handwritten signature segment.
- the handwritten signature start detector 610 outputs handwritten signature start point information, which is also first handwritten signature segment start point information, and outputs a handwritten signature start point detection signal to the time division segment detector 630.
- the handwritten signature end detection unit 620 determines that the handwritten signature is terminated when the touch data, which is the handwritten signature input data, is not input from the touch input unit 420 for a predetermined time or longer and corresponds to the position information of the last input touch data. That is, the point of reference numeral 5 of FIG. 3 is determined as the end point, and the determined handwritten signature end point information is output.
- the time division segment detector 630 drives a timer (not shown), and counts a time.
- the start point and the end point of the handwritten signature segment are detected in units of units, and a handwritten signature detection signal is output to the segment count unit 640 every time a handwritten signature segment is detected.
- the last handwritten signature segment may be shorter than the split time.
- the time division segment detection unit 630 counts the time when the signer starts handwritten signature and checks whether the count time t c is the division time t 1 , thereby dividing the count time.
- the handwriting signature segment block 2-1 is generated, the count time is initialized again, and the count time is again counted so that the count time t c is the split time t 1 .
- the handwriting signature segment block 2-2 is generated after detecting the second handwriting signature segment by checking the recognition.
- the time division segment detection unit 630 detects the handwritten signature segment block 2-3 and the handwritten segment segment 2-4 in the same manner.
- the last handwritten signature segment block 2-4 will be generated by the handwritten signature end point information input from the handwritten signature end detector 520.
- the time division segment detector 630 outputs start point information of the start point 3 of the detected handwritten signature segment and end point information of the end point 4.
- the time division segment detection unit 630 detects four handwritten signature segments and outputs a detection signal for each detection.
- time division segment detector 630 generates a segment image and outputs the segment image to the segment block feature detector 650 while tracking the handwritten signature segment.
- the segment counting unit 640 counts the number of segments n each time the segment detection signal is input from the time division segment detection unit 630, and counts the number of times when the handwriting signature end notification signal is input from the handwriting signature end detection unit 620. n) Output the information. In the case of FIG. 3, the segment count unit 640 outputs 4 as the number n information.
- the segment block feature detector 650 When the segment image is input from the time division segment detector 630, the segment block feature detector 650 generates a segment block s 0 (2-1) including the input segment image, and generates a segment for the generated segment block.
- the block feature information v 0 is generated, and when the segment block feature information v i is obtained for each of the entire handwritten signatures, the entire segment block feature information V is generated and output.
- the detailed configuration and operation of the segment block feature detector 650 will be described later with reference to FIG. 4.
- the segment block may be composed of polygons of various shapes, such as a rectangle and a pentagon, but it may be preferable that the segment block is formed in a quadrangle form as shown in FIGS. 3 and 4 so as to easily apply the same rules to all segments constituting the handwritten signature. .
- the segment block (2) of the rectangular form according to the present invention is divided into a written signature by a certain time unit, including all the segmented image segment and the area is minimized, the top point of the segment image (top i ), It may be a block formed to pass through the lowest point (bottom i ), the leftmost point (left i ) and the rightmost point (right i ).
- the handwritten signature full block feature detector 660 synthesizes segment images input from the handwritten signature image acquirer 540 or the time division segment detector 630 to generate a handwritten signature image, and includes the entire handwritten signature image.
- the handwritten signature full block S is generated, and the generated handwritten signature full block feature information Q is generated and output.
- the handwritten signature full block feature information (Q) is position information [ ⁇ (X 1 , Y 1 ), (X 2 , Y 2 ), (X) of each corner of the handwritten signature full block S as shown in Equation 2 below. 3 , Y 3 ), (X 4 , Y 4 ) ⁇ ] and the area information space S of the entire handwritten signature S.
- the segment block correlation detector 670 receives the segment block feature information v i from the segment block feature detector 650, and the handwritten signature block area of the handwritten signature block 1 from the handwritten signature full block feature detector 660.
- information (space s) of the input received segment blocks (s i) different at least one segment blocks adjacent to each (s j) with correlation information and the segment block and contains the correlation information, which is formed between the entire handwritten signature segment blocks formed between Block correlation feature information (C) is generated and output.
- the handwritten signature block characteristic information generator 680 receives the full segment block feature information V from the segment block feature detector 650 and the handwritten signature full block feature information Q from the handwritten signature full block feature detector 660. ), And receives the block correlation feature information (C) from the segment block correlation detector 670, the entire segment block feature information (V), the handwritten signature whole block feature information (Q), and the block correlation feature information. Handwritten signature characteristic information ( ⁇ ) including (C) is generated and output.
- FIG. 5 is a diagram illustrating a configuration of a segment block feature detection unit of a handwritten signature characteristic acquisition unit according to the present invention.
- the segment block feature detector 650 includes a segment block generator 651, a segment block edge position detector 652, a segment block area feature detector 653, an area ratio feature detector 654, and a segment block feature information generator 655. ).
- the segment block generator 651 When the segment image is input from the time division segment detector 630, the segment block generator 651 generates a segment block s i including the segment image to generate a segment block edge position detector 652 and a segment block area characteristic detector ( 653) and the segment block feature information generation unit 655.
- Segment block generating unit 651 as shown in FIG. 4 (B) If the segment image of the segment blocks (s i) is separated (removed write) (hereinafter referred to a discrete segment image "sub-segments" hereinafter), a sub-segment Further generates a sub-segment block 2-2-1 and a sub-segment block 2-2-2, wherein the sub-segment block 2-2-1 and the sub-segment block 2-2-2 It may further generate a vacancy area (va) block (7-1, 7-2, 7-3) for the empty space generated by.
- a vacancy area (va) block (7-1, 7-2, 7-3) for the empty space generated by.
- Segment block edge position detection unit 652 outputs (p i), the segment block position information of each corner of the segment blocks (s i) when the segment blocks (s i) received from the segment block generator 651.
- the segment block position information pi may be expressed as Equation 3 below when the block is a rectangle.
- the segment block edge position detection unit 652 is a sub-segment block including a sub-segment when the segments are separated as shown in FIG. (i is the index of the segment block, x is the index of the sub-segment, which represents the x-th subsegment of the i-th segment), and the sub-segment block ( Subsegment block location information for ) May be further outputted, and the bacony region block (which is an empty space formed by the sub-segment block) Vacancy for (7-1, 7-2, 7-3) (i is the index of the segment block, y is the index of the bacony area block, representing the y-th bacony area block of the i-th segment) Area block location information ( You might print more).
- sub segment block position information ( ) May be represented by Equation 4 below when the block is a rectangle.
- total subsegment block position information for all segment blocks ( ) May be represented by Equation 5 below when the block is a rectangle.
- the bacony area block position information (Where y is an index corresponding to the number of bacony regions) can be represented by Equation 6 below when the block is a rectangle.
- Equation 7 the entire vacancy region block position information ( ⁇ P i ) for all segment blocks may be represented by Equation 7 below when the block is a rectangle.
- the entire handwritten signature that is, the entire segment block position information P for all the segments may be represented by Equation 8 below.
- Segment block area characteristic detecting unit 653 is a block segments (s i) to calculate the area of the segment block area information received from the segment block generation section 651 ( )
- the segment block area characteristic detector 653 includes a sub segment block ( (x is an index corresponding to the number of sub-segment blocks) and the vacancy area block ( (y is an index corresponding to the number of bacony regions) to calculate subsegment block area information ( ) And vacancy area block area information ( ) To generate the segment block area information ( ) May be included in the output, or may be output as separate information.
- the area ratio characteristic detection unit 654 receives segment block area information (from the segment block area characteristic detection unit 653). ) And handwritten signature block area information (space S ) from the handwritten signature full block feature detection unit 660 to calculate the segment block area ratio to the area of the handwritten signature block to obtain segment block area ratio information ⁇ i .
- the segment block feature information generation unit 655 receives the segment block position information pi from the segment block edge position detection unit 652 and receives the segment block area information (from the segment block area characteristic detection unit 653). ) To receive input, from the area ratio characteristic detecting unit 654 receives the segment block area ratio information ( ⁇ i), segment block characteristic information (v i) to generate, and the segment block characteristic information for an entire segment block (v i ) Is generated and outputs all segment block feature information (V).
- the segment block feature information v i may be represented by Equation 9 below.
- the segment block feature information generation unit 655 may include a plurality of sub segment blocks (s) in one segment block s 1 (2-2) as shown in FIG. ), The sub-segment block area ratio information (in the total segment block feature information V). ) And bacony area block area ratio information ( ) May be included.
- each sub segment block feature information ( ) And their set ( ) May be represented as in Equation 10 below.
- ) And their set ( ) May be expressed as in Equation 11 below.
- the entire segment block feature information (V) may be expressed as in Equation 12 below.
- FIG. 6 is a diagram illustrating a detailed configuration of a segment block correlation detector of a handwritten signature characteristic acquisition unit according to the present invention
- FIG. 7 is a diagram illustrating segment block inclusion relationship information which is one of correlation information between segment blocks according to an embodiment of the present invention.
- FIG. 8 is a view for explaining a generation method
- FIG. 8 is a view for explaining a method for generating segment block position relation information, which is one of correlation information between handwritten signature segment blocks, according to an embodiment of the present invention.
- FIG. 10 is a diagram illustrating a method of generating segment block edge position relation information, which is one of correlation information between handwritten signature segment blocks
- FIG. 10 is a diagram illustrating an example of handwritten signature according to an embodiment of the present invention. .
- FIGS. 6 to 10 will be given with reference to FIGS. 6 to 10.
- the segment block correlation detector 670 includes an intersection area detector 671, an intersection area detector 672, a segment block inclusion relationship detector 673, a segment position relationship detector 674, an edge position relationship detector 675, and a correlation.
- the relationship feature information generator 677 is included.
- the cross-area detection section 671 receives the segment blocks (s i) from the segment block feature detection unit 650, the analysis across the input segment blocks (s i) and the adjacent at least one adjacent segment block (s j) It checks whether there is an area, and if there is an intersection area, calculates the intersection area of the corresponding segment block s j and the adjacent segment block s j , and generates and outputs the calculated intersection area information ⁇ ij .
- the intersection area detecting unit 671 examines the adjacent segment blocks s 1 and 2-2 adjacent to the segment blocks s 0 and 2-1 of FIG. Check for existence. Since the segment block s 0 (2-1) and the adjacent segment block (s 1 ) (2-2) have intersection areas, the intersection area detection unit 671 makes the segment block (s 0 ) 2-1 and the segment The intersection area of the blocks s 1 and 2-2 is calculated, and the intersection area information ⁇ 01 is output.
- intersection area detector 671 calculates the intersection area and outputs the intersection area information ⁇ 12 .
- intersection area detector 671 may output the intersection area information ⁇ 23 and 6 for the segment blocks s 2 and 2-3 and the segment blocks s 3 and 2-4.
- intersection area detecting unit 671 may include a sub-segment block (B) of FIG. 4. ), If present, subsegment block ( ) And subsegment blocks ( ), If there is an intersection area between ) And subsegment blocks ( )on Sub intersection area information for ) May be configured to output
- the intersection area ratio detector 672 receives segment block area information (from the area segment block area characteristic detector 653 of the segment block feature detector 650). ), The handwritten signature full block feature detection unit 660 receives the handwritten signature full block area information (space S ), and the cross-sectional area detection unit 671 receives the cross-sectional area information ( ⁇ ij ). Generate and output information.
- the intersection area ratio information includes a segment block area (s i ) of a segment block s i .
- the crossing area ratio detection unit 672 is a sub-segment block (from the area segment block area characteristic detection unit 653 of the segment block feature detection unit 650). And Subsegment block area information for And ) Is inputted to the sub-segment block from the intersection area detector 671. ) And subsegment blocks ( Sub-intersection information for ) And the sub segment block ( Subsegment block area of Sub-intersection information for Subsegment block intersection area ratio information ), Adjacent subsegment block intersection area ratio information ( ), Sub-segment block total cross area ratio information ( ) May be further calculated and output.
- Containing segment block relationship detection unit 673 outputs the segment block (s i) adjacent block segments (s j) the inclusion relationship, including segment information block relationship according to the determination result (O ij) it is determined for the.
- the segment block inclusion relationship information is stored as state information such as three pieces, namely, including (IN), without (external) (EX), and intersection (INTER).
- the segment block inclusion relationship detecting unit 673 when the adjacent segment block s j is included in the segment block s i , the segment block inclusion relationship detecting unit 673 generates the inclusion IN as segment block inclusion relationship information. If the adjacent segment block s j exists outside the segment block s i as shown in (b), it generates the inclusion (EX) as the segment block inclusion relationship information, and the adjacent segment block s j as shown in ( c). ) And the segment block s j overlap to generate an intersection as the segment block inclusion relationship information.
- the segment block inclusion relationship detection unit 673 includes a sub-segment block ( i ) in the segment block si. Adjacent subsegment blocks in Information about the sub-segment block inclusion information according to the determination result. ) May be configured to output Like the segment block inclusion relationship information O ij , the sub-segment block inclusion relationship information is stored as state information such as three types, namely, inclusion (IN), inclusion (external presence) EX, and intersection (INTER).
- state information such as three types, namely, inclusion (IN), inclusion (external presence) EX, and intersection (INTER).
- Segment position relationship detection unit 674 generates and outputs the segment block (s i), the segment location relation information (POS ij) location information of a block adjacent segments (s j), based on the.
- a segment location relation detector 674 when present in the right segment blocks (s i) to an adjacent segment block (s j), based on the as shown in (a) of FIG segment location relation information Create and output information 'R' meaning right as (POS ij ) . If adjacent segment block s j is on the left side of segment block s i as shown in ( b ), 'L' means left. It is output as segment position information (POS ij ), and as shown in ( c ), 'U' if adjacent segment (s j ) exists at the top and 'D' if it exists at the bottom as segment position relation information (POS ij ). Output
- the segment position relationship detecting unit 674 is a sub-segment block (s i ) in the segment block si.
- Relative subsegment blocks () ) On the right, sub-segment block positional information ( Information R 'representing the right side is generated and output, and the adjacent subsegment block ( ) This subsegment block ( ), The left side of the sub-segment block positional information ( ), And adjacent subsegment blocks ( ) Is at the top and 'U' is at the bottom, and 'D' is at the bottom. May be configured to output
- Edge position relationship detection unit 675 is a segment blocks (s i) and the adjacent segment block (s j) receiving input from the cross-area ratio detecting section 672 whether or not there is cross, the segment block (s i) and the adjacent segment blocks (s j) in this case, there is cross-segment blocks (s i) of the adjacent segment block basis (s j) which edges (eDGE) and the edge position relationship information indicating the intersection of the segment blocks (s i) (eDGE ij Create and print
- a corner position relationship detection unit 675 if existing under the left relative to the adjacent segment block (s j), the segment block (s i) as shown in (a) of Figure 9, the edge position ⁇ L, D ⁇ is generated and output as the relationship information EDGE ij .
- And edge position relationship detection unit 675 is a, the corner position relation information (EDGE ij) if it exists on the right side relative to the adjacent segment block (s j), the segment block (s i) as shown in (b) of Fig. 9 ⁇ Generate and print R, U ⁇ .
- the edge position relationship detection unit 675 may include a sub-segment block ( i ) in the lower center of the adjacent segment block s j based on the segment block s i as shown in ( c ) of FIG. 9. ), ⁇ L, R, D ⁇ is generated and output as edge positional relationship information EDGE ij .
- the edge position relationship detection unit 675 may determine the edge position relationship information EDGE ij when the adjacent segment block s j exists in the upper center with respect to the segment block s i as shown in FIG. 9 (d). ⁇ L, R, U ⁇ is generated and output.
- the edge position detection unit 675 is a sub-segment block ( ) And adjacent subsegment blocks ( ) Is input from the crossing area ratio detection unit 672, and whether or not the intersections are crossed. ) And adjacent subsegment blocks ( ) Intersect the subsegment block ( Relative subsegment blocks () ) This subsegment block ( Subsegment block edge positional information (which indicates which edge (EDGE) it intersects) May be configured to generate and output
- the edge position relationship detecting unit 675 includes adjacent subsegment blocks (A) of FIG. 9. ) This subsegment block ( Sub-segment block edge position information ( ) And generate ⁇ L, D ⁇ as output.
- the edge position relationship detection unit 675 may include adjacent sub-segment blocks (B) of FIG. 9. ) This subsegment block ( ), When present in the upper right corner with reference to the subsegment block edge positional information ( Generate ⁇ R, U ⁇ as
- the edge position relationship detection unit 675 may include adjacent sub-segment blocks (C) of FIG. ) This subsegment block (Sub-segment block edge position information ( And generate ⁇ L, R, D ⁇ as
- the corner position relationship detector 675 includes an adjacent sub segment block (D) as shown in FIG. ) This subsegment block (Sub-segment block edge positional information ( And generate ⁇ L, R, U ⁇ as shown in FIG. )
- the correlation feature information generation unit 677 is the intersection area ratio information input from the intersection area ratio detector 672, the segment block inclusion relationship detector 673, the segment position relationship detector 674, and the edge position relationship detector 675. Receives the segment block containment relation information (O ij ), the segment block position relation information (POS ij ), and the segment block edge position relation information (EDGE ij ), and generates correlation feature information (c ij ) including the information. When block correlation feature information (c ij ) is generated for all segment blocks, all block correlation feature information (C) is generated and output.
- the block correlation feature information c ij and the entire block correlation feature information C may be expressed as in Equations 13 and 14 below.
- the correlation feature information generation unit 677 may include the intersection area ratio detection unit 672, the segment block inclusion relationship detection unit 673, the segment position relationship detection unit 674, and the edge position relationship detection unit 675 of FIG. 4.
- the sub segment block exists as shown in b
- the sub segment block area ratio information ( ), Subsegment block intersection area ratio information ( ), Adjacent subsegment block intersection area ratio information ( ), Sub-segment block total cross area ratio information ( ), Subsegment block containment information ( ), Subsegment block positional information ( ) And subsegment block edge positional information ( ) Receives sub-segment block correlation feature information including the information ) And subsegment block correlation feature information (for all subsegment blocks). ) May be included in the overall block correlation characteristic information (C).
- the sub-segment block correlation characteristic information ( ) May be expressed as Equation 15 below, and the sub-segment block correlation feature information ( The entire block correlation characteristic information (C) including) may be expressed as in Equation 16 below.
- FIG. 10 is a diagram illustrating a time division segment block based handwritten signature authentication method according to the present invention.
- the controller 510 determines whether a handwritten signature registration is requested by the handwritten signature image registration command from the input unit 200 (S111), or whether a handwritten signature authentication request is generated by the handwritten signature authentication command. Inspect (S113).
- the controller 510 requests to input the signer identification information of the registrant, that is, the signer (S115), and checks whether the signer identification information is input (S117).
- the signer requests to sign by hand (S118).
- the controller 510 After requesting the handwriting signature, the controller 510 performs time division segment block-based hand signature signature information collection routine to collect time division segment block-based hand signature signature information ( ⁇ ) (S119), and collects the time division segment block.
- Base handwritten signature characteristic information ( ⁇ ) is mapped to the signer identification information and stored in the registration unit 100 (S121).
- the controller 510 requests to input the signer identification information (S123), and then checks whether the signer identification information is input (S125).
- the controller 510 requests to input a handwritten signature through the output unit 300 (S126).
- the controller 510 After the handwritten signature request, the controller 510 performs a time division segment block based hand signature characteristic information collection routine through the handwritten signature characteristic extractor 520 to collect time division segment block based hand signature characteristic information ( ⁇ ). After the step S127, the registered signature property information ⁇ 'registered in correspondence with the input signer identification information is loaded from the registration unit 100 through the handwritten signature segment block authentication unit 560 (S129).
- the controller 510 compares the registered handwritten signature characteristic information ⁇ ' and the handwritten signature characteristic information ⁇ through the handwritten signature segment block authentication unit 560 (S131). ). The controller 510 may further compare the conventional registered handwritten signature behavior feature information with the collected handwritten signature behavior feature information.
- the control unit 510 determines whether or not a predetermined reference match degree is preset for each item through the handwritten signature segment block authentication unit 560 (S133), and performs an authentication failure process when the preset reference match degree is lower (S137), and in advance. If the set criterion is equal to or greater than the authentication success process (S135).
- FIG. 11 is a diagram illustrating a method of collecting handwritten signature data of a time division segment block based handwritten signature authentication method according to the present invention.
- a method of collecting handwritten signature information of a time division segment block-based handwritten signature authentication method will be described.
- the controller 510 may include touch signature, which is handwritten signature input data, from the touch input unit 420 through at least one of the handwritten signature tracker 530, the handwritten signature image acquirer 540, and the handwritten signature characteristic acquirer 550. Monitor whether it starts to be input (S211).
- the controller 510 When the touch data input for the handwritten signature starts to be monitored, the controller 510 performs a segment and a handwritten signature detection routine to detect a segment, a segment block, a sub segment, a sub segment block, a bacsi region block, and the like.
- Information including a segment image, a sub-segment image, a handwritten signature image, a segment block, a sub-segment block, and a total handwritten signature block S is detected and stored (S215).
- the segment and handwritten signature detection routine will be described in detail with reference to FIGS. 12 and 13 below.
- the controller 510 detects and stores the segment block number information n (S217).
- the controller 510 calculates the area of all the handwritten signature blocks S and generates the handwritten signature full block area information space S (S219).
- Handwritten signature when the entire block area information (space S) is created the control unit 510 is handwritten signature entire block area (space S) than segment block area ( ), The ratio of segment block area ratio information ( ⁇ i ), sub-segment block area ratio information ( ) And bacony area block area ratio information ( ) Is generated (S221).
- the controller 510 initializes variables (i, j, m, etc.) (S225).
- the controller 510 checks whether there is an adjacent segment block s j having an overlapping region for each of the handwritten signature segment blocks s i (S227). .
- the controller 510 counts the number m of adjacent segment blocks (S229).
- the area of the inclusion and overlap regions formed by the calculation is calculated to generate cross area information ⁇ ij (S231).
- the controller 510 When the cross- sectional area information ⁇ ij is generated, the controller 510 generates segment block cross- sectional area ratio information, which is ratio information of the cross- sectional area ⁇ ij that is an overlapping area with respect to the area of the segment block s i . ) Is generated (S233).
- control unit 510 is adjacent segment block intersection area ratio information which is ratio information of the area intersecting the area ⁇ ij with respect to the area of the segment block s i and the overlapping area, that is, the adjacent segment block s j having the intersection area. ( ) Is generated (S235).
- the control unit 510 includes segment block inclusion relationship information (O ij ), which is information on whether the adjacent handwritten signature segment block s j is included or not, based on the segment block s i for each segment block s i . s i ) Segment block positional relationship information POS ij , which is segment block relative position information indicating a position of a reference adjacent segment block, and an adjacent segment block s j are connected to either edge EDGE of the reference segment block s i . Segment block edge positional relationship information EDGE ij indicating the positional information of the edge indicating whether or not to intersect is generated (S237, S239, S241).
- Control unit 510 S225 to S251 through the processes all segment blocks (s i: i ⁇ n), each segment blocks for each (s i: i ⁇ n), one adjacent any adjacent segment blocks based on the (s j: Generate block correlation feature information (c ij ) with j ⁇ n).
- the controller 510 When the information is generated, the controller 510 generates block correlation feature information C (S253), and finally generates handwritten signature characteristic information ( ⁇ ) including all of the information and stores it in the register 100. (S255).
- a real-time segment detection method for detecting a segment by a predetermined time unit based on input touch data and a segment segmentation detection method for detecting a segment by a predetermined time unit after acquiring the entire handwritten signature Etc.
- FIG. 12 is a flowchart illustrating a method of generating a time division segment block according to a first embodiment of the present invention, and illustrates a method of generating a time division segment block using a real time segment detection method.
- the controller 510 counts the segment identification time t c through the time division segment detector 630 (S311).
- the controller 510 starts tracking the handwritten signature by the input touch data (S313).
- the controller 510 determines whether the segment identification time t c , which is the count time, is the segment reference time t 1 (S315).
- the controller 510 counts the number of segments n (S317).
- the controller 510 After counting the number of segments, the controller 510 generates and stores a handwritten signature segment image (S319). As shown in FIG. 4, the segment image is generated by touch data inputted up to a segment reference time t 1 by tracking a written signature recorded from a segment start point.
- the generation control section 510 After the image segment, the generation control section 510 generates a block segment (s i) (S321). Said segment blocks (s i) is the highest point (top), the lowest point (bottom), comprising the left-most point (left), the right-most point (right), and the minimum area as shown in (a) of Fig. 4 It may be formed to have.
- the controller 510 detects segment block position information pi (S323).
- the controller 510 determines the segment block area information (pi). ), That is, calculated (S325).
- the controller 510 determines whether the handwritten signature is terminated through the handwritten signature end detection unit 620 (S327).
- the controller 510 increases the value of the segment index variable i by 1 (S329), initializes the segment identification time t c (S331), and repeats the process after S311. .
- segment detection is performed when the touch data is continuously input, that is, when the handwritten signature does not end, the segment reference time t 1 . Thereafter it will be configured to count the segment identification time t c immediately after initializing the value of the segment index variable i immediately.
- the controller 510 stores the counted number of segments n (S333), generates a handwritten signature image, and stores the generated handwritten signature image (S335).
- FIG. 13 is a flowchart illustrating a method of generating a time division segment block according to a second embodiment of the present invention, and illustrates a method of generating a segment block using a segment division detection method.
- the controller 510 counts the handwritten signature recording time t r (S411).
- the controller 510 starts tracking the handwritten signature (S413).
- the controller 510 checks whether the handwritten signature is terminated (S415), and when the handwritten signature is terminated, generates a handwritten signature image and stores the handwritten signature recording time t r (S417). .
- the controller 510 divides the handwritten signature image stored with the time information into predetermined time units to generate and store the segmented image (S419).
- the controller 510 When the segment image is generated, the controller 510 generates and stores a segment block s i for each segmented segment image (S421).
- the control unit 510 is stored and then counting the number of segment blocks (S425).
- the controller 510 detects and stores segment block position information and segment block area information of each segment block (S427, S429).
- the present invention is not limited to the above-described typical preferred embodiment, but can be carried out in various ways without departing from the gist of the present invention, various modifications, alterations, substitutions or additions in the art réelle who has this can easily understand it. If the implementation by such improvement, change, replacement or addition falls within the scope of the appended claims, the technical idea should also be regarded as belonging to the present invention.
Abstract
Description
Claims (17)
- 서명자가 서명하는 수기서명에 대응하여 터치되는 위치들에 대한 위치데이터 및 압력데이터를 포함하는 터치데이터를 수기서명 입력 데이터로서 출력하는 터치입력부를 포함하는 수기서명 입력부;서명자별 수기서명 특성 정보를 등록하는 등록부; 및상기 수기서명 입력부로부터 입력되고 터치데이터를 포함하는 수기서명 입력 데이터 및 수기서명 입력 데이터가 입력되는 시간을 카운트하여 함께 저장하고, 상기 수기서명 입력 데이터 및 카운트되는 시간에 의해 세그먼트 기준 시간(t1)단위로 세그먼트를 인식하고, 수기서명을 인식하여 수기서명 이미지 및 세그먼트 이미지를 생성하고, 상기 수기서명을 포함하는 수기서명 블록 및 상기 세그먼트 이미지를 포함하는 세그먼트 블록을 생성하며, 상기 세그먼트 블록 및 수기서명 블록 각각의 정보 및 상기 세그먼트 블록들 상호간 상관관계 정보와, 상기 세그먼트 블록 및 수기서명 블록 상호간 상관관계 정보를 포함하는 수기서명 특성 정보(Σ)를 수집하여 상기 서명자의 서명자 식별정보에 맵핑하여 상기 등록부에 등록한 후, 수기서명 인증 요청 발생 시 상기 수기서명 입력부의 터치입력부를 통해 입력되는 터치데이터로부터 수기서명 블록 및 세그먼트 블록들 상호간의 상관관계 정보를 포함하는 수기서명 특성 정보(Σ)를 수집하고 수기서명 인증을 요청한 서명자의 식별정보에 대응하는 등록 수기서명 특성 정보(Σ')를 로드하여 수집된 수기서명 특성 정보(Σ)와 비교하여 일치도에 따른 수기서명 세그먼트에 기반하여 수기서명 인증을 수행하는 수기서명 인증부를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제1항에 있어서,상기 수기서명 인증부는,상기 수기서명 입력부의 터치 입력부로부터 입력되는 수기서명 입력데이터로부터 입력되는 수기서명 입력 데이터 및 수기서명 입력 데이터가 입력되는 시간을 카운트하여 함께 저장하고, 상기 수기서명 입력 데이터 및 카운트되는 시간에 의해 세그먼트 기준 시간(t1)단위로 세그먼트를 인식하고, 수기서명을 인식하여 수기서명 이미지 및 세그먼트 이미지를 생성하고, 상기 수기서명 이미지를 포함하는 수기서명 블록 및 상기 세그먼트 이미지를 포함하는 세그먼트 블록을 생성한 후, 수기서명에 대한 수기서명 이미지를 포함하는 수기서명 블록의 특징 정보인 수기서명 전체 블록 특징 정보(Q), 수기서명을 구성하는 상기 수기서명 세그먼트들의 특징 정보인 전체 세그먼트 블록들의 전체 세그먼트 블록 특징 정보(V), 상기 수기서명 블록 및 세그먼트 블록들 간의 상관관계에 의한 블록 상관관계 특징 정보(C)를 포함하는 수기서명 특성 정보(Σ)를 추출하는 수기서명 특성 추출부;상기 수기서명 특성 추출부를 통해 추출된 수기서명 특성 정보(Σ)와 미리 등록되어 있는 등록 수기서명 특성 정보(Σ')를 비교하여 각각 미리 설정된 일치도에 따른 수기서명 인증을 수행하는 수기서명 세그먼트 블록 인증부; 및등록 요청 발생 시 상기 수기서명 특성 추출부를 통해 추출되는 수기서명 특성 정보를 상기 등록부에 저장하여 등록시키고, 수기서명 인증 요청 발생 시 상기 수기서명 세그먼트 블록 인증부를 제어하여 수기서명 인증을 수행시키는 제어부를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제1항에 있어서,상기 수기서명 특성 추출부는,상기 터치데이터로부터 수기서명의 시작을 검출하는 수기서명 시작 검출부;상기 터치데이터의 입력되던 중 일정 시간 터치데이터가 입력되지 않으면 최종 터치데이터 입력 위치를 수기서명 종료 지점으로 수기서명이 종료를 검출하는 수기서명 종료 검출부;상기 터치데이터로부터 입력되는 터치데이터 및 터치데이터가 입력되는 시간을 카운트하여 함께 저장하고, 상기 터치데이터 및 카운트되는 시간에 의해 세그먼트 기준 시간(t1)단위로 세그먼트를 인식하고, 수기서명을 인식하여 수기서명 이미지 및 세그먼트 이미지를 생성하여 출력하는 시간분할 세그먼트 검출부;상기 시간분할 세그먼트 검출부에서 검출된 세그먼트 개수를 카운트하는 세그먼트 카운트부;상기 세그먼트 이미지를 입력받아 상기 세그먼트 이미지를 포함하는 세그먼트 블록(si)을 생성하고, 생성된 세그먼트 블록(si)에 대한 특징 정보인 세그먼트 블록 특징 정보(vi)를 생성하고 생성된 세그먼트 블록 특징 정보(vi)들을 포함하는 전체 세그먼트 블록 특징 정보(V)를 생성하여 출력하는 세그먼트 블록 특징 검출부;획득된 수기서명 이미지를 포함하는 수기서명 블록(S)을 생성하고, 상기 수기서명 블록(S)의 특징 정보인 수기서명 전체 블록 특징 정보(Q)를 생성하여 출력하는 수기서명 전체 블록 특징 검출부;상기 수기서명 전체 블록과 상기 세그먼트 블록들 상호간의 상관관계에 따른 정보인 블록 상관관계 특징 정보(C)를 생성하여 출력하는 세그먼트 블록 상관관계 검출부; 및상기 수기서명 전체 블록 특징 정보(Q), 상기 전체 세그먼트 블록 특징 정보(V) 및 상기 블록 상관관계 특징 정보(C)를 포함하는 수기서명 특성 정보(Σ)를 생성하여 출력하는 수기서명 블록 특성 정보 생성부를 포함하는 수기서명 특성 획득부를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제3항에 있어서,상기 수기서명 전체 블록 특징 검출부는,상기 수기서명 블록(S)의 면적을 계산하여 수기서명 전체 블록 면적 정보(spaceS)를 더 생성하여 출력하고,세그먼트 블록 특징 검출부는,상기 세그먼트 이미지를 입력받고, 상기 세그먼트 이미지를 포함하는 상기 세그먼트 블록(si)을 생성하여 출력하는 세그먼트 블록 생성부;상기 세그먼트 블록(si)을 입력받고, 세그먼트 블록의 각 모서리의 위치정보인 세그먼트 블록 위치 정보(pi)를 검출하여 출력하는 세그먼트 블록 모서리 위치 검출부;상기 세그먼트 블록(si) 및 상기 세그먼트 블록 위치 정보(pi) 중 적어도 하나 이상을 입력받아 상기 세그먼트 블록(si)의 면적을 계산하여 세그먼트 블록 면적 정보()를 생성하여 출력하는 세그먼트 블록 면적 특성 검출부;상기 수기서명 전체 블록 특징 검출부로부터 수기서명 전체 블록 면적 정보(spaceS) 및 상기 세그먼트 블록 면적 정보()를 입력받아 상기 수기서명 전체 블록 면적 대비 세그먼트 블록 면적 비율을 계산하여 세그먼트 블록 면적 비율 정보(Δi)를 생성하여 출력하는 면적 비율 특성 검출부; 및수기서명의 각 세그먼트별 상기 세그먼트 블록 모서리 위치 정보(pi), 상기 세그먼트 블록 면적 정보() 및 세그먼트 블록 면적 비율 정보(Δi)를 포함하는 세그먼트 블록 특징 정보(vi)를 생성하고, 전체 수기서명의 세그먼트들에 대한 전체 세그먼트 블록 특징 정보(V)를 생성하여 출력하는 세그먼트 블록 특징 정보 생성부를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제4항에 있어서,상기 블록은 사각형의 블록이되,상기 세그먼트 블록 생성부는,해당 세그먼트의 최상위지점, 최하위지점, 최좌측지점 및 최우측지점을 통과하여 상기 세그먼트를 포함하는 사각형의 세그먼트 블록을 생성하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제3항에 있어서,상기 수기서명 전체 블록 특징 검출부는,상기 수기서명 블록(S)의 면적을 계산하여 수기서명 전체 블록 면적 정보(spaceS)를 더 생성하여 출력하고,상기 세그먼트 블록 상관관계 검출부는,세그먼트 블록(si)과 인접하게 위치하는 인접 세그먼트 블록(sj)들 중 상기 세그먼트 블록(si)과 포함 및 교차(중첩)영역을 형성하는 인접 세그먼트 블록(sj)이 있는지를 판단하고, 있으면 포함 및 교차영역의 면적을 계산하여 교차 면적 정보(δij)를 출력하는 교차 면적 검출부;상기 수기서명 전체 블록 면적 정보(spaceS), 세그먼트 블록 면적 정보() 및 상기 교차 면적 정보(δij)를 입력받고, 상기 수기서명 전체 블록 면적에 대한 교차 면적(δij) 비율을 계산하여 수기서명 블록 교차 면적 비율 정보(rij)를 생성하고, 상기 세그먼트 블록 면적() 대비 상기 교차 면적(δij) 비율을 계산하여 세그먼트 블록 교차 면적 비율 정보()를 생성하며, 상기 인접 세그먼트 블록 면적(spacesj) 대비 상기 교차 면적(δij) 비율을 계산하여 인접 세그먼트 블록 교차 면적 비율 정보()를 생성하는 교차 면적 비율 정보 교차 면적 비율 검출부;세그먼트 블록(si)을 기준으로 인접 세그먼트 블록(sj)이 포함되는지의 여부 및 교차하는지의 여부 정보인 세그먼트 블록 포함관계 정보(Oij)를 생성하여 출력하는 세그먼트 블록 포함관계 검출부;세그먼트 블록(si)을 기준으로 인접 세그먼트 블록(sj)의 위치를 정보인 세그먼트 블록 위치 정보(POSij)를 생성하여 출력하는 세그먼트 위치관계 검출부;세그먼트 블록(si)을 기준으로 인접 세그먼트 블록(sj)이 세그먼트 블록(si)의 어느 쪽 모서리와 교차하는지를 나타내는 위치정보인 세그먼트 블록 모서리 위치관계 정보(EDGEij)를 생성하여 출력하는 모서리 위치관계 검출부; 및상기 교차 면적 정보(δij), 수기서명 블록 교차 면적 비율 정보(rij), 세그먼트 블록 교차 면적 비율 정보(), 인접 세그먼트 블록 교차 면적 비율 정보(), 세그먼트 블록 포함관계 정보(Oij), 세그먼트 블록 위치 정보(POSij), 세그먼트 블록 모서리 위치관계 정보(EDGEij)를 포함하는 블록 상관관계 특징 정보(C)를 생성하여 출력하는 상관관계 특징 정보 생성부를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제3항에 있어서,상기 시간분할 세그먼트 검출부는,실시간으로 입력되는 터치데이터의 입력 시 상기 입력되는 터치데이터에 의해 상기 세그먼트 기준 시간(t1) 단위로 순차적으로 세그먼트를 인식하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제3항에 있어서,상기 시간분할 세그먼트 검출부는,입력되는 터치데이터 및 터치데이터의 입력 시간을 카운트하여 저장하고, 상기 터치데이터에 의해 수기서명 이미지가 검출되면 상기 수기서명 이미지를 기록하는 데 소요된 수기서명 기록 시간에 기준하여 상기 세그먼트 기준 시간(t1) 단위로 분할하여 세그먼트를 인식하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 제4항에 있어서,상기 세그먼트 블록 생성부는,상기 세그먼트 블록 모서리 위치 검출부는,상기 세그먼트 블록 면적 특성 검출부는,면적 비율 특성 검출부는,하나의 세그먼트 블록(si)에 복수개의 서브 세그먼트 블록()이 존재하는 경우, 세그먼트 블록 면적() 대비 서브 세그먼트 블록 면적() 비율 및 세그먼트 블록 면적 대비 베이컨시 영역 블록 면적()의 비율을 계산하여 서브 세그먼트 블록 면적 비율 정보() 및 베이컨시 영역 블록 면적 비율 정보()를 출력하며,상기 세그먼트 블록 특징 정보 생성부는,서브 세그먼트 블록()에 대한 서브 세그먼트 블록 위치 정보(), 베이컨시 영역 블록()의 베이컨시 영역 블록 위치 정보(), 서브 세그먼트 블록 면적 정보(), 베이컨시 영역 블록 면적 정보(), 서브 세그먼트 블록 면적 비율 정보() 및 베이컨시 영역 블록 면적 비율 정보()가 더 반영된 전체 세그먼트 블록 특징 정보(V)를 생성하여 출력하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 시스템.
- 수기서명 입력부로부터 입력되고 터치데이터를 포함하는 수기서명 입력 데이터 및 수기서명 입력 데이터가 입력되는 시간을 카운트하여 함께 저장하고, 상기 수기서명 입력 데이터 및 카운트되는 시간에 의해 세그먼트 기준 시간(t1)단위로 세그먼트를 인식하고 수기서명을 인식하여 수기서명 이미지 및 세그먼트 이미지를 생성하고, 상기 수기서명을 포함하는 수기서명 블록 및 상기 세그먼트를 포함하는 세그먼트 블록을 생성하며, 상기 세그먼트 블록 및 수기서명 블록 각각의 정보 및 상기 세그먼트 블록들 상호간 상관관계 정보와, 상기 세그먼트 블록 및 수기서명 블록 상호간 상관관계 정보를 포함하는 수기서명 특성 정보(Σ)를 수집하여 서명자의 서명자 식별정보에 맵핑하여 등록부에 등록하는 등록 과정; 및수기서명 인증 요청 발생 시 상기 수기서명 입력부의 터치입력부를 통해 입력되는 터치데이터로부터 수기서명 블록 및 세그먼트 블록들 상호간의 상관관계 정보를 포함하는 수기서명 특성 정보(Σ)를 수집하고 수기서명 인증을 요청한 서명자의 식별정보에 대응하는 등록 수기서명 특성 정보(Σ')를 로드하여 수집된 수기서명 특성 정보(Σ)와 비교하여 일치도에 따른 수기서명 세그먼트에 기반하여 수기서명 인증을 수행하는 수기서명 인증 과정을 포함하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제10항에 있어서,상기 수기서명 등록 과정은,수기서명 등록 요청이 발생되는지를 모니터링하는 등록 요청 모니터링 단계;수기서명 등록 요청이 발생되면 등록할 서명자의 서명자 식별정보를 획득하는 서명자 식별정보 획득단계;서명자에 의한 수기서명에 대해 터치입력부로부터 입력되는 터치데이터로부터 상기 수기서명 특성 정보(∑)를 획득하는 수기서명 특성 데이터 획득 단계; 및상기 서명자 식별정보에 상기 수기서명 특성 정보를 맵핑하여 등록부에 저장하여 등록하는 수기서명 등록 단계를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제10항에 있어서,상기 수기서명 인증 과정은,수기서명 인증 요청의 발생여부를 모니터링 하는 수기서명 인증 요청 모니터링 단계;수기서명 인증 요청이 발생되면 서명자의 서명자 식별정보를 획득하는 서명자 식별정보 획득단계;상기 서명자에 의한 수기서명에 대해 터치입력부로부터 입력되는 터치데이터로부터 상기 수기서명 특성 정보(∑)를 획득하는 수기서명 특성 정보 획득 단계;상기 획득된 서명자 식별정보에 대응하여 미리 등록되어 있는 등록 수기서명 특성 정보(∑')를 로드하는 등록 수기서명 특성 정보 로드 단계; 및상기 획득된 수기서명 특성 정보(∑)와 상기 로드된 등록 수기서명 특성 정보(∑')를 비교하여 수기서명 인증을 수행하고 그 인증 결과를 출력하는 수기서명 인증 단계를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제10항 또는 제12항에 있어서,상기 수기서명 특성 정보(∑) 획득 단계는,수기서명 입력부로부터 입력되는 수기서명 입력데이터의 터치데이터로부터 수기서명의 추적을 개시하는 수기서명 추적단계;상기 수기서명 추적이 개시되면 시간분할 세그먼트 검출부를 통해 입력되는 수기서명 입력 데이터 및 수기서명 입력 데이터가 입력되는 시간을 카운트하여 함께 저장하고, 상기 수기서명 입력 데이터 및 카운트되는 시간에 의해 세그먼트 기준 시간(t1)단위로 세그먼트를 인식하여 출력하는 세그먼트 검출단계;상기 시간분할 세그먼트 검출부에서 검출된 세그먼트 개수를 카운트하는 세그먼트 카운트 단계;상기 세그먼트 이미지를 입력받아 상기 세그먼트 이미지를 포함하는 세그먼트 블록(si)을 생성하고, 생성된 세그먼트 블록(si)에 대한 특징 정보인 세그먼트 블록 특징 정보(vi)를 생성하여 출력하는 세그먼트 블록 특징 검출 단계;획득된 수기서명 이미지를 포함하는 수기서명 블록(S)을 생성하고, 상기 수기서명 블록(S)의 특징 정보인 수기서명 전체 블록 특징 정보(Q)를 생성하여 출력하는 수기서명 전체 블록 특징 검출 단계;상기 수기서명 전체 블록과 상기 세그먼트 블록들 상호간의 상관관계에 따른 정보인 블록 상관관계 특징 정보(C)를 생성하여 출력하는 세그먼트 블록 상관관계 검출단계; 및전체 세그먼트들에 대한 세그먼트 블록 특징 정보(vi)들을 포함하는 전체 세그먼트 블록 특징 정보(V)를 생성하고, 상기 수기서명 전체 블록 특징 정보(Q), 상기 전체 세그먼트 블록 특징 정보(V) 및 상기 블록 상관관계 특징 정보(C)를 포함하는 수기서명 특성 정보(Σ)를 생성하여 출력하는 수기서명 블록 특성 정보 생성단계를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제13항에 있어서,상기 수기서명 전체 블록 특징 검출 단계는,상기 수기서명 블록(S)의 면적을 계산하여 수기서명 전체 블록 면적 정보(spaceS)를 더 생성하여 출력하고,세그먼트 블록 특징 검출 단계는,상기 세그먼트 이미지를 입력받고, 상기 세그먼트 이미지를 포함하는 상기 세그먼트 블록(si)을 생성하여 출력하는 세그먼트 블록 생성 단계;상기 세그먼트 블록(si)을 입력받고, 세그먼트 블록의 각 모서리의 위치정보인 세그먼트 블록 위치 정보(pi)를 검출하여 출력하는 세그먼트 블록 모서리 위치 검출 단계;상기 세그먼트 블록(si) 및 상기 세그먼트 블록 위치 정보(pi) 중 적어도 하나 이상을 입력받아 상기 세그먼트 블록(si)의 면적을 계산하여 세그먼트 블록 면적 정보()를 생성하여 출력하는 세그먼트 블록 면적 특성 검출 단계;상기 수기서명 전체 블록 특징 검출 단계에서 검출된 수기서명 전체 블록 면적 정보(spaceS) 및 상기 세그먼트 블록 면적 정보()를 입력받아 상기 수기서명 전체 블록 면적(spaceS) 대비 세그먼트 블록 면적() 비율을 계산하여 세그먼트 블록 면적 비율 정보(Δi)를 생성하여 출력하는 면적 비율 특성 검출 단계; 및수기서명의 각 세그먼트별 상기 세그먼트 블록 모서리 위치 정보(pi), 상기 세그먼트 블록 면적 정보() 및 세그먼트 블록 면적 비율 정보(Δi)를 포함하는 세그먼트 블록 특징 정보(vi)를 생성하고, 전체 수기서명의 세그먼트들에 대한 전체 세그먼트 블록 특징 정보를 생성하여 출력하는 세그먼트 블록 특징 정보 생성 단계를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제14항에 있어서,상기 블록은 사각형의 블록이되,상기 세그먼트 블록 생성 단계에서 세그먼트 블록 생성부는,해당 세그먼트의 최상위지점, 최하위지점, 최좌측지점 및 최우측지점을 통과하여 상기 세그먼트를 포함하는 사각형의 세그먼트 블록을 생성하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제13항에 있어서,상기 수기서명 전체 블록 특징 검출 단계는,상기 수기서명 블록(S)의 면적을 계산하여 수기서명 전체 블록 면적 정보(spaceS)를 더 생성하여 출력하고,상기 세그먼트 블록 상관관계 검출 단계는,세그먼트 블록(si)과 인접하게 위치하는 인접 세그먼트 블록(sj)들 중 상기 세그먼트 블록(si)과 포함 및 교차(중첩)영역을 형성하는 인접 세그먼트 블록(sj)이 있는지를 판단하고, 있으면 포함 및 교차영역의 면적을 계산하여 교차 면적 정보(δij)를 출력하는 교차 면적 검출 단계;상기 수기서명 전체 블록 면적 정보(spaceS), 세그먼트 블록 면적 정보() 및 상기 교차 면적 정보(δij)를 입력받고, 상기 수기서명 전체 블록 면적에 대한 교차 면적(δij) 비율을 계산하여 수기서명 블록 교차 면적 비율 정보(rij)를 생성하고, 상기 세그먼트 블록 면적() 대비 상기 교차 면적(δij) 비율을 계산하여 세그먼트 블록 교차 면적 비율 정보()를 생성하며, 상기 인접 세그먼트 블록 면적() 대비 상기 교차 면적(δij) 비율을 계산하여 인접 세그먼트 블록 교차 면적 비율 정보()를 생성하는 교차 면적 비율 검출 단계;세그먼트 블록(si)을 기준으로 인접 세그먼트 블록(sj)이 포함되는지의 여부 및 교차하는지의 여부 정보인 세그먼트 블록 포함관계 정보(Oij)를 생성하여 출력하는 세그먼트 블록 포함관계 검출 단계;세그먼트 블록(si)을 기준으로 인접 세그먼트 블록(sj)의 위치 정보인 세그먼트 블록 위치 정보(POSij)를 생성하여 출력하는 세그먼트 위치관계 검출 단계;세그먼트 블록(si)을 기준으로 인접 세그먼트 블록(sj)이 세그먼트 블록(si)의 어느 쪽 모서리와 교차하는지를 나타내는 위치정보인 세그먼트 블록 모서리 위치관계 정보(EDGEij)를 생성하여 출력하는 모서리 위치관계 검출 단계; 및상기 교차 면적 정보(δij), 수기서명 블록 교차 면적 비율 정보(rij), 세그먼트 블록 교차 면적 비율 정보(), 인접 세그먼트 블록 교차 면적 비율 정보(), 세그먼트 블록 포함관계 정보(Oij), 세그먼트 블록 위치 정보(POSij), 세그먼트 블록 모서리 위치관계 정보(EDGEij)를 포함하는 블록 상관관계 특징 정보(C)를 생성하여 출력하는 상관관계 특징 정보 생성 단계를 포함하는 것을 특징으로 하는 시간분할 세그먼트 블록 기반 수기서명 인증 방법.
- 제14항에 있어서,상기 세그먼트 블록 생성 단계는,상기 세그먼트 이미지를 입력받고, 상기 세그먼트 이미지를 포함하는 상기 세그먼트 블록(si)을 생성하여 출력하는 세그먼트 블록 검출 단계;상기 세그먼트 블록 모서리 위치 검출 단계는,상기 세그먼트 블록(si)을 입력받고, 세그먼트 블록의 각 모서리의 위치정보인 세그먼트 블록 위치 정보(pi)를 검출하여 출력하는 세그먼트 블록 모서리 위치 정보 생성 단계;상기 세그먼트 블록 면적 특성 검출 단계는,상기 세그먼트 블록(si) 및 상기 세그먼트 블록 위치 정보(pi) 중 적어도 하나 이상을 입력받아 상기 세그먼트 블록(si)의 면적을 계산하여 세그먼트 블록 면적 정보()를 생성하여 출력하는 세그먼트 블록 면적 특성 정보 생성 단계;면적 비율 특성 검출 단계는,상기 수기서명 전체 블록 특징 검출 단계에서 검출된 수기서명 전체 블록 면적 정보(spaceS) 및 상기 세그먼트 블록 면적 정보()를 입력받아 상기 수기서명 전체 블록 면적(spaceS) 대비 세그먼트 블록 면적() 비율을 계산하여 세그먼트 블록 면적 비율 정보(Δi)를 생성하여 출력하는 면적 비율 특성 정보 생성 단계;하나의 세그먼트 블록(si)에 복수개의 서브 세그먼트 블록()이 존재하는 경우, 세그먼트 블록 면적() 대비 서브 세그먼트 블록 면적()의 비율을 계산하여 서브 세그먼트 블록 면적 비율 정보()생성하여 출력하는 서브 세그먼트 면적 비율 특성 정보 생성 단계; 및하나의 세그먼트 블록(si)에 복수개의 서브 세그먼트 블록()이 존재하는 경우, 세그먼트 블록 면적() 대비 베이컨시 영역 블록 면적()의 비율을 계산하여 베이컨시 영역 블록 면적 비율 정보()를 생성하여 출력하는 베이컨시 영역 블록 면적 비율 특성 정보 생성 단계를 포함하되,상기 세그먼트 블록 특징 정보 생성 단계는
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2018529598A JP6695578B2 (ja) | 2015-12-11 | 2016-12-12 | 時分割セグメントブロック基盤手記署名認証システムおよび方法 |
US16/060,825 US10515257B2 (en) | 2015-12-11 | 2016-12-12 | Handwritten signature authentication system and method based on time-division segment block |
EP16873420.0A EP3388963B1 (en) | 2015-12-11 | 2016-12-12 | Segment-based handwritten signature authentication system and method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150177503A KR101598331B1 (ko) | 2015-12-11 | 2015-12-11 | 시간분할 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 |
KR10-2015-0177503 | 2015-12-11 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017099555A1 true WO2017099555A1 (ko) | 2017-06-15 |
Family
ID=55541715
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2016/014556 WO2017099555A1 (ko) | 2015-12-11 | 2016-12-12 | 시간분할 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 |
Country Status (5)
Country | Link |
---|---|
US (1) | US10515257B2 (ko) |
EP (1) | EP3388963B1 (ko) |
JP (1) | JP6695578B2 (ko) |
KR (1) | KR101598331B1 (ko) |
WO (1) | WO2017099555A1 (ko) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017222125A1 (ko) * | 2016-06-20 | 2017-12-28 | 주식회사 피노텍 | 식별코드를 이용한 자필서명 검증 시스템 및 방법 |
KR101990072B1 (ko) * | 2016-08-09 | 2019-06-18 | 주식회사 피노텍 | 식별코드를 이용한 자필서명 검증 시스템 및 방법 |
KR101789298B1 (ko) * | 2016-06-20 | 2017-11-20 | 주식회사 피노텍 | 식별코드를 이용한 자필서명 검증 시스템 및 방법 |
CN106127162B (zh) * | 2016-06-28 | 2022-06-24 | 联想(北京)有限公司 | 一种签名识别方法和电子设备 |
CN107784219B (zh) * | 2016-08-25 | 2019-04-30 | 腾讯科技(深圳)有限公司 | 一种字符验证码实现方法、装置及终端 |
KR101911679B1 (ko) * | 2017-02-10 | 2018-10-25 | 주식회사 하이딥 | 터치 압력 센서를 이용한 사용자 인증 방법 및 장치 |
KR102456020B1 (ko) | 2017-08-17 | 2022-10-19 | 삼성전자주식회사 | 전자 문서에 서명을 포함시키기 위한 전자 장치 및 그 전자 장치의 제어 방법 |
KR102111858B1 (ko) * | 2018-08-16 | 2020-05-15 | 한국과학기술원 | 기계학습을 이용한 획 기반 수기 서명 인증 방법 및 시스템 |
CN112686140A (zh) * | 2020-12-29 | 2021-04-20 | 平安普惠企业管理有限公司 | 电子签名的识别方法、系统、设备及存储介质 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000353243A (ja) * | 1999-06-11 | 2000-12-19 | Toshiba Corp | 署名照合装置とその方法、ならびにプログラム記憶媒体 |
JP2007110180A (ja) * | 2005-10-11 | 2007-04-26 | Hitachi Ltd | 署名履歴保管装置、署名履歴保管方法およびそのプログラム |
JP2007299226A (ja) * | 2006-04-28 | 2007-11-15 | Canon Inc | 画像処理装置、画像処理方法、サイン登録プログラム及び記憶媒体 |
WO2015030500A1 (ko) * | 2013-08-30 | 2015-03-05 | 삼성전자 주식회사 | 전자 장치 및 전자 장치의 서명 입력 처리 방법 |
KR20150041569A (ko) * | 2013-10-08 | 2015-04-16 | 삼성전자주식회사 | 서명 등록 방법, 서명 인증 방법 및 그 장치 |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5091975A (en) * | 1990-01-04 | 1992-02-25 | Teknekron Communications Systems, Inc. | Method and an apparatus for electronically compressing a transaction with a human signature |
DE69329380T2 (de) * | 1993-06-30 | 2001-03-01 | Ibm | Verfahren zum Segmentieren von Bildern und Klassifizieren von Bildelementen zur Dokumentverarbeitung |
EP0737342A1 (en) * | 1993-12-17 | 1996-10-16 | Quintet, Incorporated | Method of automated signature verification |
US5745598A (en) * | 1994-03-11 | 1998-04-28 | Shaw; Venson Ming Heng | Statistics based segmentation and parameterization method for dynamic processing, identification, and verification of binary contour image |
US6236740B1 (en) * | 1995-04-07 | 2001-05-22 | Michael E. Lee | Signature verification apparatus and method utilizing relative angle measurements |
US5745592A (en) * | 1995-07-27 | 1998-04-28 | Lucent Technologies Inc. | Method for detecting forgery in a traced signature by measuring an amount of jitter |
US5909500A (en) * | 1996-01-02 | 1999-06-01 | Moore; Steven Jerome | Method and apparatus for detecting forged signatures |
US5930380A (en) * | 1997-02-11 | 1999-07-27 | Lucent Technologies, Inc. | Method and apparatus for verifying static signatures using dynamic information |
GB0201232D0 (en) * | 2002-01-19 | 2002-03-06 | Queen Mary & Westfield College | Authentication systems |
KR100608775B1 (ko) | 2004-11-23 | 2006-08-02 | 삼성전자주식회사 | 필체를 이용한 사용자 인증 시스템 및 방법 |
RO121497B1 (ro) | 2005-02-09 | 2007-06-29 | Softwin S.R.L. | Sistem informatic şi metodă pentru achiziţia, analiza şi autentificarea semnăturii olografe |
KR100957059B1 (ko) | 2007-12-04 | 2010-05-13 | 한미아이티 주식회사 | 터치 스크린을 통해 입력된 문자 데이터를 처리하는 시스템및 방법 |
KR101062011B1 (ko) | 2009-11-16 | 2011-09-05 | 한국조폐공사 | 서명정보 입력부를 구비한 스마트 카드 및 사용자 인증방법 |
US9589120B2 (en) * | 2013-04-05 | 2017-03-07 | Microsoft Technology Licensing, Llc | Behavior based authentication for touch screen devices |
US20140331313A1 (en) * | 2013-05-03 | 2014-11-06 | Vialab, Inc. | Authentication of signature using acoustic wave analysis |
WO2015162607A1 (en) * | 2014-04-23 | 2015-10-29 | Signpass Ltd. | Methods and systems for signature analysis and authentication |
KR101585842B1 (ko) | 2015-10-05 | 2016-01-15 | 주식회사 시큐브 | 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 |
-
2015
- 2015-12-11 KR KR1020150177503A patent/KR101598331B1/ko active IP Right Grant
-
2016
- 2016-12-12 JP JP2018529598A patent/JP6695578B2/ja active Active
- 2016-12-12 WO PCT/KR2016/014556 patent/WO2017099555A1/ko active Application Filing
- 2016-12-12 US US16/060,825 patent/US10515257B2/en active Active
- 2016-12-12 EP EP16873420.0A patent/EP3388963B1/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000353243A (ja) * | 1999-06-11 | 2000-12-19 | Toshiba Corp | 署名照合装置とその方法、ならびにプログラム記憶媒体 |
JP2007110180A (ja) * | 2005-10-11 | 2007-04-26 | Hitachi Ltd | 署名履歴保管装置、署名履歴保管方法およびそのプログラム |
JP2007299226A (ja) * | 2006-04-28 | 2007-11-15 | Canon Inc | 画像処理装置、画像処理方法、サイン登録プログラム及び記憶媒体 |
WO2015030500A1 (ko) * | 2013-08-30 | 2015-03-05 | 삼성전자 주식회사 | 전자 장치 및 전자 장치의 서명 입력 처리 방법 |
KR20150041569A (ko) * | 2013-10-08 | 2015-04-16 | 삼성전자주식회사 | 서명 등록 방법, 서명 인증 방법 및 그 장치 |
Also Published As
Publication number | Publication date |
---|---|
EP3388963A4 (en) | 2019-08-28 |
EP3388963A1 (en) | 2018-10-17 |
US20180357469A1 (en) | 2018-12-13 |
US10515257B2 (en) | 2019-12-24 |
KR101598331B1 (ko) | 2016-03-14 |
JP2018538623A (ja) | 2018-12-27 |
JP6695578B2 (ja) | 2020-05-20 |
EP3388963B1 (en) | 2023-08-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017099555A1 (ko) | 시간분할 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 | |
WO2017061758A1 (ko) | 세그먼트 블록 기반 수기서명 인증 시스템 및 방법 | |
WO2018230875A1 (ko) | 단말 및 그 제어 방법 | |
WO2017039287A1 (ko) | 세그먼트 기반 수기서명 인증 시스템 및 방법 | |
WO2019216499A1 (ko) | 전자 장치 및 그 제어 방법 | |
WO2020017706A1 (en) | Electronic device and method for controlling the same | |
WO2018088806A1 (ko) | 영상처리장치 및 영상처리방법 | |
WO2020153552A1 (ko) | 기록된 거래의 수정이 가능한 블록체인을 위한 방법 및 장치 | |
WO2021049791A1 (ko) | 증강현실과 가상현실을 이용한 문서의 처리 시스템 및 그 방법 | |
WO2018088809A1 (en) | Method of displaying user interface related to user authentication and electronic device for implementing same | |
WO2017099314A1 (ko) | 사용자 정보를 제공하는 전자 장치 및 방법 | |
WO2020013545A1 (en) | Apparatus and method for authenticating object in electronic device | |
WO2020017902A1 (en) | Method and apparatus for performing user authentication | |
WO2018151356A1 (ko) | 다중 스케일 곡률 기반 가시적 벡터 모델 해싱 방법 | |
WO2017142299A1 (ko) | 공간분할 세그먼트에 대한 동적이동 추적 기반 수기서명 인증 시스템 및 방법 | |
WO2017039103A1 (ko) | 이동 단말기 및 그의 제어 방법 | |
WO2020168606A1 (zh) | 广告视频优化方法、装置、设备及计算机可读存储介质 | |
WO2020004686A1 (ko) | 블록체인 dPoC 합의알고리즘 기반 합의 인증 수행 방법 | |
WO2022268023A1 (zh) | 指纹识别方法、装置、电子设备和可读存储介质 | |
WO2020189827A1 (ko) | 전자 장치 및 그 제어 방법 | |
WO2018199576A1 (en) | Method and apparatus for performing authentication based on biometric information | |
WO2020071693A1 (ko) | 전자 장치, 서버 및 이를 이용한 서명 인증 방법 | |
WO2019074185A1 (en) | ELECTRONIC APPARATUS AND CONTROL METHOD THEREOF | |
EP3659073A1 (en) | Electronic apparatus and control method thereof | |
WO2020171549A1 (en) | Apparatus for searching for content using image and method of controlling same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16873420 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2018529598 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2016873420 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2016873420 Country of ref document: EP Effective date: 20180711 |