JP2014081877A - Information processing apparatus, signature information acquisition method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing apparatus, a signature information acquisition method, and a program capable of reducing incorrect recognition of a signature even if the signature is a simple handwritten signature such as an initial, so as to provide signature information capable of suppressing impersonating input by a stranger.SOLUTION: An information processing apparatus uses a touch sensor capable of multipoint simultaneous authentication to extract one stroke information as signature image information from information on input points that are simultaneously input by spread fingers at two or more points, acquires input point information from a relationship between the input points, and acquires the signature image information and the input point information as signature information. The information processing apparatus further performs authentication by collating and comparing the pieces of signature information.

Description

  The present invention relates to an information processing apparatus, a sign information acquisition method, and a program, and more particularly to an information processing apparatus, a sign information acquisition method, and a program that can acquire many features as data from a handwritten signature.

  Conventionally, many methods are used to authenticate an individual. For example, a method using a personal identification number or password is common, but safety is low in terms of leakage or theft. For this reason, development of personal authentication using static physical features such as fingerprints, irises, faces, or the shape of palms or finger blood vessels instead of passwords or passwords, or dynamic behavioral features Personal authentication using handwritten signatures (signatures) has been developed.

  And in authentication systems targeting a large number of unspecified individuals such as the reception authentication system and reservation authentication system of visitors, there are some people who feel psychologically strong in registering physical characteristics, so it is difficult to spread was there. However, authentication by handwritten signature has been performed for a long time, and has the advantage that it is relatively easy for users to accept.

  As an authentication system for such a handwritten signature, for example, in Patent Document 1, at the time of registration writing, a CCD camera and a pressure sensor are used to capture handwriting data of an image and writing pressure at each sampling time, and from these handwriting data, 5 authentication factors of moving vector, writing speed, pen holding tilt angle, black / white dot composition ratio and writing pressure are extracted and generated and registered as reference data for each person. A system for verifying signatures by comparing with reference data is disclosed.

  In Patent Document 2, the shape of a stroke of a dictionary registered character and the relative positional relationship between the apex angle of a triangle formed by three points selected by a predetermined method on the stroke and the length of two sides sandwiching the apex angle are disclosed. Expressed by a dictionary character feature matrix having one or both of the ratios as components, the input characters are also expressed by the input character feature matrix by the same processing, and the corresponding components of the input character feature matrix and the dictionary character feature matrix If the distance between the obtained matrices is smaller than a predetermined threshold distance, the input character input person is the same person as the corresponding dictionary registration character registrant. Is disclosed.

Japanese Patent Laid-Open No. 10-11574 Japanese Patent Laid-Open No. 10-63849

  Since the techniques described in Patent Documents 1 and 2 are intended for signature authentication by a single stroke such as pen input, the accuracy by authentication is improved by obtaining additional information other than stroke information of only handwriting. ing. However, the technique described in Patent Document 1 requires a pressure sensor for detecting writing pressure, and the technique described in Patent Document 2 prepares a dictionary character feature matrix for dictionary registered characters in advance. There is a problem that the apparatus becomes large and the necessary memory and processing load become large.

  On the other hand, in recent years, touch sensors having a multipoint simultaneous recognition function capable of simultaneously recognizing a plurality of input points have become common, and are also used for smartphones and large touch panel displays. However, when authenticating a handwritten signature using a large touch panel display, since it is easy for a third party to sniff the sign when inputting the sign, there is a high risk of impersonation input using the sign that the third party has seen. In addition, to prevent spoofing input by a third party, if the handwritten signature is complicated or the collation accuracy at the time of authentication is tightened, even if it is a signature of a legitimate sign input person, the signature recognition is good. There was a problem that we couldn't.

  The present invention has been made in view of these circumstances, and by using a touch sensor capable of simultaneous multi-point authentication, stroke information is obtained from information on input points that are simultaneously input at two or more points by extending a fingertip. In addition, by obtaining multiple pieces of information from the distance between each input point and the shape of the polygon connecting each input point, false recognition can be reduced from a simple handwritten signature of the initial level, and spoofed input by a third party An object of the present invention is to provide an information processing apparatus, a sign information acquisition method, and a program that can obtain suppressible sign information.

  In order to solve the above-mentioned problem, the first technical means of the present invention provides a touch sensor having a multi-point simultaneous recognition function for simultaneously recognizing a plurality of input points, and a sign by a plurality of fingers input to the touch sensor. An input coordinate detection unit that detects coordinate data of a plurality of corresponding input points every predetermined time, and one stroke information from the coordinate data of the plurality of input points every predetermined time detected by the input coordinate detection unit A signature image information extraction unit that identifies and extracts the one stroke information as signature image information, and is obtained at the same time among the coordinate data of the plurality of input points for each predetermined time detected by the input coordinate detection unit. An input point information extraction unit for extracting input point information obtained from the relationship of the coordinate data of the plurality of input points, and the sign image information and the input point information by the plurality of fingers. Is obtained is characterized by having a signature information determining unit for determining a sign of Inn.

  The second technical means further comprises a storage unit in the first technical means, and a collation unit that collates the confirmed sign information with the sign information stored in the storage unit in advance. It is a feature.

  According to a third technical means, in the first or second technical means, the input point information includes distance information between two input points of the plurality of input points.

  According to a fourth technical means, in any one of the first to third technical means, the input point information includes polygon area information having three or more input points of the plurality of input points as vertices. It is characterized by that.

  According to a fifth technical means, in any one of the first to fourth technical means, the input point information is information on an interior angle of a polygon having three or more input points of the plurality of input points as vertices. It is characterized by including.

  In a sixth technical means, in the technical means according to any one of the first to fifth aspects, the one stroke information is stroke information made up of coordinate data of midpoints of the coordinates of the plurality of input points every predetermined time. It is characterized by that.

  A seventh technical means is a sign information acquisition method for acquiring sign information from signs by a plurality of fingers input to a touch sensor having a multi-point simultaneous recognition function for simultaneously recognizing a plurality of input points. An input coordinate detecting step for detecting coordinate data of a plurality of contact points corresponding to the input signs by a plurality of fingers at a predetermined time, and a plurality of input points at the predetermined time detected at the input coordinate detecting step. A sign image extracting step of identifying one stroke information from the coordinate data and extracting the one stroke information as sign image information; and coordinates of the plurality of input points for each predetermined time detected by the input coordinate detection unit Among the data, an input point information extraction that extracts input point information obtained from the relationship of the coordinate data of the plurality of input points obtained at the same time. Step a is obtained by further comprising a signature information determination step of determining the input point information and the signature image data as signature information sign by the plurality of fingers.

  The eighth technical means is a program for causing a computer to execute the sign information acquisition method which is the seventh technical means.

  According to the present invention, the distance and angle between the fingertips does not spread beyond a certain distance and there are individual differences. For example, when inputting with two fingers widened or entering a hand From the information of the input point obtained based on the sign of, as well as obtaining one stroke information as sign information, and obtaining input point information such as the distance between each fingertip, the area of the polygon formed by the fingertip, and the angle, Sign information having many features can be obtained from a handwritten signature. For this reason, even if it is simple handwritten characters, such as an initial, misrecognition can be reduced and the sign information which can suppress the impersonation by the third party can be obtained. Further, by comparing the obtained signature information, it is possible to determine whether or not the person is an authorized input person.

It is a figure which shows the structure of the information processing apparatus which concerns on one Embodiment of this invention. It is a figure which shows an example of the signature input into the information processing apparatus of this invention. It is a figure for demonstrating the method to acquire signature information from the signature input into the information processing apparatus of this invention. It is a figure which shows the coordinate data of the input point for every predetermined time obtained from the signature shown in FIG. It is a figure which shows the other example of the signature input into the information processing apparatus of this invention. FIG. 6 is a diagram for explaining input point information obtained from a polygon formed by input points by each finger in the sign of FIG. 5. It is a flowchart which shows an example of a process in the case of performing signature registration by the information processing apparatus which concerns on one Embodiment of this invention. It is a flowchart which shows an example of a process in the case of performing signature authentication by the information processing apparatus which concerns on one Embodiment of this invention.

  Hereinafter, preferred embodiments of an information processing apparatus, a sign information acquisition method, and a program according to the present invention will be described with reference to the drawings. In the following description, the configurations denoted by the same reference numerals in different drawings are the same, and the description thereof may be omitted.

  FIG. 1 is a diagram showing a configuration of an information processing apparatus according to an embodiment of the present invention. The information processing apparatus 1 includes a touch panel 10, a signature information extraction unit 20, a storage unit 30, a signature information verification unit 40, and a communication unit 50. The touch panel 10 includes a multipoint authentication touch sensor 11 and a display 12.

  The multipoint authentication touch sensor 11 is of a projected capacitive type capable of simultaneously detecting multiple points designated by a plurality of fingertips, etc., and is a conventional resistive film type touch panel. Then, a detection signal indicating a plurality of input points (indicated points) is output simultaneously. For this reason, for example, even when two input points are simultaneously input by two fingertips, it is possible to identify the two input points based on the detection signal of the multipoint authentication touch sensor 11.

  The display 12 is configured by a display panel such as a liquid crystal display or an organic EL (electroluminescence) display for displaying an image. The multi-point authentication touch sensor 11 is provided on the display 12 and configured as a touch panel 10. In the present invention, the multipoint authentication touch sensor 11 may be a single touch sensor configured separately from the display 12, and the touch panel 10 combined with the display 12 as shown in the present embodiment. May be used. In the latter case, the touch panel 10 is a touch panel having a multipoint simultaneous authentication function.

  The signature information extraction unit 20 includes an input coordinate detection unit 21, a signature image information extraction unit 22, an input point information extraction unit 23, and a signature information determination unit 24. The input coordinate detection unit 21 has a function of receiving the detection signal from the multipoint authentication touch sensor 11 and detecting the coordinates of the input point every predetermined sampling time (predetermined time), and the detected coordinate information is not shown. It is recorded in the storage device or storage unit 30. The sign image information extraction unit 22 has a function of specifying one stroke information from the coordinate data of a plurality of input points for each sampling time detected by the input coordinate detection unit 21 and extracting the one stroke information as sign image information. Have

  The input point information extraction unit 23 is an input point obtained from the relationship between the coordinate data of a plurality of input points obtained at the same time among the coordinate data of the plurality of input points for each sampling time detected by the input coordinate detection unit 21. It has a function to extract information. Details of the input point information will be described later. The sign information confirmation unit 24 has a function of confirming the sign image information obtained by the sign image information extraction unit 22 and the input point information obtained by the input point information extraction unit 23 as sign information of a sign by a plurality of fingers Have

  The storage unit 30 is composed of an HDD (hard disk drive) or a semiconductor memory, and stores sign information 31 of a sign with a plurality of fingers obtained by the sign information extraction unit 20. The signature information matching unit 40 includes signature information 31 including signature image information and input point information stored in the storage unit 30 in advance, and signature image information obtained from signatures by a plurality of fingers input to the touch panel 10 and input. It has a function of collating and comparing each piece of sign information composed of point information and determining whether or not the information is within a predetermined coincidence range. Then, when the sign information is within a predetermined coincidence range, the operator is authenticated as an operator of a pre-registered sign.

  The information processing apparatus 1 includes a CPU (Central Processing Unit) or MPU (Micro Processing Unit) (not shown), a RAM (Random Access Memory) as a work area, and a ROM that stores various application programs and various setting contents. Each function of the signature information extraction unit 20 and the signature information collation unit 40 is realized by executing an application program stored in the ROM or the storage unit 30.

  The external interface 50 is an interface for connecting to an external information processing apparatus and a memory, and can be, for example, a network connection, a local connection, or an interface of a portable storage medium such as a memory stick. Then, the information processing apparatus 1 can send the signature information obtained by the signature information extraction unit 20 to another information processing apparatus via the external interface 50. Further, the information processing apparatus 1 can receive sign information obtained by another information processing apparatus via the external interface 50 and store the sign information in the storage unit 30 in advance as sign information.

  When the information processing apparatus 1 is used as a sign information registration apparatus, it is sufficient to have at least the touch panel 10 and the sign information extraction unit 20, and the obtained sign information is stored in the storage unit 30 or externally. You may make it send to another information processing apparatus via the interface 50. FIG. Further, when the information processing apparatus 1 is used as an authentication apparatus for sign information, in addition to the touch panel 10 and the sign information extraction unit 20, a storage unit 30 and a sign information collation unit 40 for storing the sign information 31 in advance are provided. Necessary.

Next, a method for obtaining signature information from a signature input on the touch panel 10 in the present invention will be described. FIG. 2 is a diagram illustrating an example of a signature input to the information processing apparatus of the present invention. 2, the stroke of the solid line L1, L2 denotes a two-stroke fingers of the index finger and the middle finger as a sign that is input to the touch panel 10, a solid line L1 from the input point a ₁ by the index finger to the input point a _m, the solid line L2 represents the stroke of the input point b ₁ to the input point b _m.

  When the user registers a signature using the information processing apparatus 1, the user can register the signature by signing on the touch panel 10 after selecting the icon 13 for sign registration. When authenticating a signature, the signature can be authenticated by selecting an icon 14 for signature authentication and then signing on a signature authentication screen displayed on the touch panel 10. The display of the icon 13 for sign registration and the icon 14 for sign authentication may be made selectable from a GUI screen (not shown) for entering the sign information input mode displayed on the touch panel 10.

  As shown in FIG. 2, for example, when the index finger and the middle finger are spread, the distance between the fingertips does not extend beyond a certain distance w, and there is an individual difference, but the distance becomes a certain distance w for each individual. . For this reason, if a sign is made by spreading the index finger and the middle finger, the locus of the sign becomes two parallel solid lines L1 and L2 separated by a distance w. In FIG. 2, the trajectory of the sign input by the index finger and the middle finger is displayed on the display 12, but it may be hidden because it is not seen by a third party.

FIG. 3 is a diagram for explaining a method for acquiring signature information from a signature input to the information processing apparatus of the present invention. Like the sign sign shown in FIG. 2 shown in FIG. 3, the stroke of which has been signed input with two fingers the touch panel 10, a solid line L1 from the input point a ₁ by the index finger to the input point a _m , the solid line L2 represents the stroke of the input point _{b 1} to the input point _{b m.} And the input coordinate detection part 21 acquires the coordinate data of the input point into which these signs are input for every predetermined sampling time.

FIG. 4 is a diagram showing the coordinate data of the input points every predetermined time obtained from the sign shown in FIG. If the signs shown in Figure 3 has been performed over the time _{t m} from the time _{t 1,} the input coordinate detection unit 21, coordinate data _Pa 1 input points _{a 1} input at a sampling time _{t 1 (xa 1,} ya ₁ ) and the input point _{b 1} coordinate data _{Pb 1 (xb 1, yb 1} ) is detected and further coordinate data _Pa 2 _(xa 2 input points _{a 2} at the next sampling time _{t 2,} ya ₂₎ and the input point b ₂ coordinate data Pb ₂ (xb ₂ , yb ₂ ) is detected. Then, the input coordinate detection unit 21, at time _{t m} the signature input ends, the coordinate data _{Pa m (xa m, ya m} ) of the input point _{a m} and coordinate data _Pb m _(xb m input point _{b m,} yb _The coordinate data is acquired every sampling time until _{m 2} ), and the coordinate data detected every sampling time shown in FIG. 4 is recorded in the storage unit 30 or a recording device (not shown).

  Next, the sign image information extraction unit 22 identifies one piece of stroke information obtained from the coordinate data detected by the input coordinate detection unit 21. Therefore, the sign image information extraction unit 22 determines which coordinate data at a certain sampling time is related to the coordinate data at the next sampling time based on the coordinate data detected by the input coordinate detection unit 21. Identify and group related coordinate data by stroke.

Specifically, first, the sign image information extraction unit 22 takes in the coordinate data Pa ₁ (xa ₁ , ya ₁ ) and the coordinate data Pb ₁ (xb ₁ , yb ₁ ) obtained at time t ₁ , and coordinates The data Pa ₁ (xa ₁ , ya ₁ ) is set as the start point of the stroke 1 corresponding to the solid line L1, and the coordinate data Pb ₁ (xb ₁ , yb ₁ ) is set as the start point of the stroke 2 corresponding to the solid line L2. Next, from the coordinate data Pa ₂ (xa ₂ , ya ₂ ) and the coordinate data Pb ₂ (xb ₂ , yb ₂ ) obtained at time t ₂ , the coordinate data Pa ₁ (xa ₁ , ya ₁ ) is the highest. The coordinate data close to the coordinate data Pb ₁ (xb ₁ , yb ₁ ) is captured as the end point coordinate data of the stroke 2 while the close coordinate data is captured as the end point coordinate data of the stroke ₁ . Normally, if the sampling time is sufficiently shorter than the time required for signing, the coordinate data Pa ₂ (xa ₂ , ya ₂ ) is used as the end point coordinate data of the stroke 1 and the coordinate data Pb ₂ (xb ₂ , yb ₂ ) is distributed to the strokes 1 and 2 as the end point coordinate data of the stroke 2.

Similarly, from the coordinate data Pa ₃ (xa ₃ , ya ₃ ) and the coordinate data Pb ₃ (xb ₃ , yb ₃ ) obtained at time t ₃ , to the end point coordinate data Pa ₂ (xa ₂ , ya ₂ ). The coordinate data closest to the coordinate data Pb ₂ (xb ₂ , yb ₂ ) is acquired as new end point coordinate data of the stroke 2 while the closest coordinate data is acquired as new end point coordinate data of the stroke 1. In a similar manner, repeated m times to capture the time _t coordinate data obtained by _{m Pa} m _(xa m, ya _m) and the coordinate data _Pb m _(xb m, yb _m) a.

In this way, the sign image information extraction unit 22 uses the obtained coordinate data of the input point as the stroke information of the stroke 1 corresponding to the solid line L1, Pa ₁ (xa ₁ , ya ₁ ), Pb ₂ (xb _{2 , yb 2) ··· Pa m (} xa m, the coordinate data group including the coordinate data of ya _m), as stroke information for the stroke 2 corresponding to the solid line _{L2, Pb 1 (xb 1,} yb 1), Pb 2 (Xb ₂ , yb ₂ )... Pb _m (xb _m , yb _m ) is assigned to a coordinate data group including coordinate data, and stroke information corresponding to the handwriting of each finger sign is obtained. Note that even when the sign is a point such as a period, it is handled as a stroke. In this case, the coordinate data of the stroke information obtained by the signature of each finger includes point coordinate data.

  Next, the sign image information extraction unit 22 selects one predetermined stroke information from the obtained plurality of stroke information, and specifies it as sign image information. As a selection method, of the stroke information obtained from the sign of each finger, for example, the stroke information of the sign whose start point is located on the leftmost side (in this case, the x coordinate value of the start point is the smallest) is the sign image. It can be specified as information, or the stroke information of a sign whose end point coordinate is located on the uppermost side (in this case, the value of the y coordinate of the end point is the largest) can be specified as sign information. Any method may be used for selecting the sign image data.

  Further, the stroke information specified as the sign image information may be stroke information including coordinate data of the midpoint of the coordinates of a plurality of input points. In this case, the average value of the x coordinate and the average value of the y coordinate of a plurality of coordinate data at a certain sampling time point detected by the input coordinate detection unit 21 are respectively expressed as the x coordinate value and the y coordinate at the midpoint at the sampling time point. By setting the value, it is possible to obtain stroke information including the coordinate data of the midpoint.

For example, in the sign shown in Figure 3, the coordinate midpoint c ₁ between the input point b ₁ input point a ₁ and the solid line L2 of the solid line L1 and a starting point, midpoint c of the input point b _m and the input point a _m Stroke information including a coordinate data group of strokes indicated by a broken line L3 having _m as an end point can be specified as sign image information. In this case, the coordinate data _{P c} of the midpoint _{c n} at any sampling time point _{t n,} calculated from _{((xa n + yb n)} / 2, (xa n + yb n) / 2).

  In this way, by using the stroke information composed of the coordinate data of the midpoint as the sign image data, the sign image information extraction unit 22 determines the correspondence between each coordinate data at a certain sampling time and each coordinate data at the next sampling time. There is no need to make a determination, and it is only necessary to calculate the coordinate data of the midpoint, so that even when the index increases, the processing load can be reduced. The stroke information in the present invention may not include time information. Therefore, the sign image information can be output as still image digital data such as JPEG or GIF.

Next, a method in which the input point information extraction unit 23 extracts input point information from the relationship between the coordinate data of a plurality of input points obtained at the same time (same sampling time) will be described. In the sign by two fingers shown in FIG. 3, among the coordinate data of a plurality of input points for each predetermined sampling time detected by the input coordinate detection unit 21, the distance w between two input points at an arbitrary sampling time tn. ^{_{(= √ ((xa n -xb}} n) 2 + (ya n -yb n) 2) are extracted as the input point information.

As described above, when the index finger and the middle finger are spread, the distance between the fingertips does not spread beyond a certain distance w, and the distance becomes a certain distance w. Therefore, the distance w may be obtained from the coordinate data at the sampling point t ₁ at the start point and the sampling point t _m at the end point. In addition, when signing, each individual can decide which finger to use for signing, and even with the same person, the thumb and index finger, thumb and middle finger, or thumb and little finger Since the distances are different, it is possible to reduce misrecognition by determining the finger to be used and performing authentication at the distance w even if the stroke has a similar sign.

  Then, the signature information confirmation unit confirms the information including the signature image information and the input point information extracted as described above as signature information of a signature by a plurality of fingers, and stores the signature in the storage unit 30. The information is stored or transmitted to an external information processing apparatus via the external interface 50. When signature verification is performed, the confirmed signature information is sent to the signature information verification unit 40.

  Next, a case where a sign is performed with three or more fingers will be described. FIG. 5 is a diagram showing another example of a signature input to the information processing apparatus according to the present invention, and shows a state where the signature is performed with five fingers in a state where the palm is spread. In FIG. 5, a solid line L11 indicates a stroke by a thumb, a solid line L12 indicates an index finger, a solid line L13 indicates a middle finger, a solid line L14 indicates a ring finger, and a solid line L15 indicates a stroke by a little finger. These solid lines L11, L12, L13, L14, and L15 are in a substantially parallel relationship. However, when a sign is displayed on the touch panel 10, the solid lines overlap each other, so that even if a third party steals, unlike a single sign, it is difficult to determine what kind of sign has been made. Has advantages. Therefore, spoofing by sign theft is difficult to be performed.

  When the sign is performed by three or more fingers, the number of coordinates of the input point increases and the stroke information increases. However, the input coordinate detection unit 21 detects the coordinates of the input point, and the sign image information extraction unit 22 detects the sign. The point of extracting the image information is basically the same as when the sign is made with two fingers. However, since the input point information obtained from the relationship between the coordinates of three or more input points has more types than the input point information obtained from the relationship between the coordinates of the two input points, the case of the two-point input point information More input point information can be extracted than.

FIG. 6 is a diagram for explaining the input point information obtained from the polygon formed by the input points by each finger in the sign of FIG. 6, the input point _d n at any sampling time point _{t n, e n, f n} , g n, shows a pentagon connecting _{h n,} as input point information, for example, between the two input points adjacent Information of the distances w1 to w5, the inner angles α1 to α5 of the polygon, or the area S of the polygon can be extracted. These pieces of input point information can be used as information for authenticating the individual because the length and angle of each finger when the palm is opened differ depending on the individual.

Incidentally, i _n a certain point in the pentagon shown in FIG. 5, 5 a rectangular midpoint, pentagonal x average value of each i _n the average value and y coordinate value of the x coordinate value of each vertex It has as coordinate value and y coordinate value. As the stroke information, the coordinate data of the middle point “in” is used as the stroke information. For example, even if the sign of five fingers is a discontinuous locus, the stroke information reflecting the shape of the stroke can be easily obtained. Can be obtained.

  Next, a signature registration process flow by the information processing apparatus of the present invention will be described. FIG. 7 is a flowchart showing an example of processing when signature registration is performed by the information processing apparatus according to an embodiment of the present invention. First, in step S1, it is determined whether or not the signature information input mode is set. The sign information input mode may be selected by operating the GUI screen displayed on the touch panel 10. If it is not the sign information input mode, it waits until the sign information input mode is selected.

  Next, the input coordinate detection unit 21 determines whether or not there is a touch input on the touch panel 10 (step S2). As described above, the sign input may be accepted after touching the sign registration icon 13 displayed on the touch panel before the user performs the touch input. If there is no touch input on the touch panel 10, the process waits until there is a touch input. If a touch input is detected (YES in step S <b> 2), the input coordinate detection unit 21 detects the input coordinates (step S <b> 3).

  Next, proceeding to step S4, the input coordinate detection unit 21 determines whether or not two or more input points are detected simultaneously. If two or more input points cannot be detected at the same time, the process moves to step S5, and a caution for prompting a signature by two or more fingers is displayed on the touch panel 10, and then the process returns to step S3. If two or more input points are detected at the same time in step S4, the process proceeds to step S6, where the sign image information extraction unit 22 extracts sign image information.

  After that, the sign information determination unit 24 extracts the input point information by the input point information extraction unit 23 (step S7), and the sign image information extracted by the sign image information extraction unit 22 and the input point information extraction unit 23 extract it. The entered point information is confirmed as sign information (step S8), and the confirmed sign information is registered in the storage unit 30, for example. (Step S9). Note that the extraction of the sign image information in step S6 and the extraction of the input point information in step S7 may be performed in order or in parallel. Here, the steps from Step S1 to Step S8 correspond to steps for obtaining signature information, and are indicated by Step S10 in FIG.

  Next, a signature registration process flow by the information processing apparatus of the present invention will be described. FIG. 7 is a flowchart showing an example of processing when signature registration is performed by the information processing apparatus according to an embodiment of the present invention. There are two cases of signature verification: a case where signature registration and signature verification are performed using the same information processing device, and a case where signature registration and signature verification are performed using different information processing devices. In the former case, it is assumed that the signature information acquired in advance by the own device is registered in the storage unit 30 of the own device. In the latter case, the sign information acquired by the other device is stored in the storage unit 30 of the own device. It is assumed that it is registered. Also in the following description, it is assumed that sign information to be collated is registered in the information processing apparatus in advance.

  First, in step S10, the signature information collation unit 40 acquires signature information. The acquisition of the sign information is the same as Step 10 for acquiring the sign information from Step S1 to Step S8 described with reference to FIG. However, the sign input may be accepted after the user touches the signature authentication icon 14 displayed on the touch panel before performing the touch input. Then, the sign information including the sign image information and the input point information obtained in step S10 is temporarily stored in the RAM or register of the information processing apparatus 1 (step S11).

  Next, the signature information collation unit 40 calls the signature information registered in the storage unit 30 (step S12), collates the obtained signature image information with the registered signature image information, and obtains the obtained signature image information. The input point information is compared with the registered input point information, and it is determined whether or not each is within a predetermined range of coincidence (step S14). Further, the result of collation is displayed on the display 12 or the like (step S14), and the process for sign collation is terminated.

  The information processing apparatus according to the present invention and the sign information acquisition method using the information processing apparatus according to the present invention have been described above, but the present invention is for causing a computer to execute the sign information acquisition method. The form as a program can also be taken. This program is stored in the storage unit 30 of the information processing apparatus having a computer as an application program, and can be read from the storage unit 30 and executed to implement the functions according to the present invention.

  As an embodiment of the present invention, a touch sensor having a multi-point simultaneous recognition function for simultaneously recognizing a plurality of input points, and coordinates of a plurality of input points corresponding to signs by a plurality of fingers inputted to the touch sensor An input coordinate detection unit that detects data every predetermined time, and one stroke information is identified from the coordinate data of the plurality of input points for each predetermined time detected by the input coordinate detection unit. Of the plurality of input points at a predetermined time among the coordinate data of the plurality of input points for each predetermined time detected by the input coordinate detection unit. An input point information extraction unit for extracting input point information obtained from the relationship; and the sign image information and the input point information as signature information of a sign by the plurality of fingers. It may be an information processing apparatus characterized by having a signature information determining unit for determining. Thereby, signature information having many features can be acquired from a handwritten signature. For this reason, even if it is simple handwritten characters, such as an initial, the misrecognition can be reduced and the sign information which can suppress the impersonation by the third party can be obtained.

  The information processing apparatus may further include a storage unit and a verification unit that performs verification of the determined signature information and the signature information stored in advance in the storage unit. Accordingly, it can be determined whether or not the user is a legitimate input person, and even a simple handwritten character such as an initial can reduce misrecognition and suppress spoofing by a third party.

  The input point information may include distance information between two input points among the plurality of input points. The input point information may include polygon area information having three or more input points as vertices among the plurality of input points. In addition, the input point information may include information on an inner angle of a polygon having three or more input points as vertices among the plurality of input points. Thus, characteristic data can be obtained from the handwritten signature.

  Further, the one stroke information may be stroke information including coordinate data of a midpoint of coordinates of a plurality of input points every predetermined time. Thereby, even if the number of fingers increases, stroke information can be obtained by a simple process.

  Further, the present invention is a sign information acquisition method for acquiring sign information from a plurality of finger signs input to a touch sensor having a multi-point simultaneous recognition function for simultaneously recognizing a plurality of input points. An input coordinate detecting step for detecting coordinate data of a plurality of contact points corresponding to the signatures made by the plurality of fingers every predetermined time, and coordinates of the plurality of input points detected at the predetermined time detected in the input coordinate detecting step A sign image extracting step of identifying one stroke information from the data and extracting the one stroke information as sign image information; and coordinate data of the plurality of input points for each predetermined time detected by the input coordinate detection unit Input point information extraction for extracting input point information obtained from the relationship between the coordinate data of the plurality of input points at a predetermined time Step a, it may be a sign information acquisition method characterized in that it comprises sign information determination step of determining the input point information and the signature image data as signature information sign by the plurality of fingers. Thereby, signature information having many features can be acquired from a handwritten signature. For this reason, even if it is simple handwritten characters, such as an initial, the misrecognition can be reduced and the sign information which can suppress the impersonation by the third party can be obtained. Furthermore, the present invention may be a program for causing a computer to execute the above sign information acquisition method.

DESCRIPTION OF SYMBOLS 1 ... Information processing apparatus, 10 ... Touch panel, 11 ... Multi-points authentication touch sensor, 12 ... Display, 13 ... Sign registration icon, 14 ... Sign authentication icon, 20 ... Sign information extraction part, 21 ... Input coordinate detection part, 22 ... Sign image information extraction unit, 23 ... input point information extraction unit, 24 ... sign information determination unit, 30 ... storage unit, 31 ... sign information, 40 ... sign information collation unit, 50 ... external interface.

Claims (8)

A touch sensor having a multi-point simultaneous recognition function for simultaneously recognizing a plurality of input points;
An input coordinate detection unit that detects coordinate data of a plurality of input points corresponding to signs by a plurality of fingers input to the touch sensor at predetermined time intervals;
A sign image information extracting unit for identifying one stroke information from the coordinate data of the plurality of input points for each predetermined time detected by the input coordinate detecting unit, and extracting the one stroke information as sign image information;
An input for extracting input point information obtained from the relationship of the coordinate data of the plurality of input points obtained at the same time among the coordinate data of the plurality of input points at the predetermined time detected by the input coordinate detection unit. A point information extraction unit;
An information processing apparatus, comprising: a sign information determination unit configured to determine the sign image information and the input point information as sign information of a sign by the plurality of fingers.   The information processing apparatus according to claim 1, further comprising: a storage unit; and a collation unit that collates the confirmed sign information with the sign information stored in advance in the storage unit.   The information processing apparatus according to claim 1, wherein the input point information includes distance information between two input points among the plurality of input points.   The information processing apparatus according to claim 1, wherein the input point information includes polygonal area information having three or more input points as vertices among the plurality of input points. .   5. The information processing according to claim 1, wherein the input point information includes information on an interior angle of a polygon having three or more input points of the plurality of input points as vertices. apparatus.   The information processing apparatus according to claim 1, wherein the one stroke information is stroke information including coordinate data of a middle point of a plurality of input points for each predetermined time. . A sign information acquisition method for acquiring sign information from a sign by a plurality of fingers input to a touch sensor having a multi-point simultaneous recognition function for simultaneously recognizing a plurality of input points,
An input coordinate detection step of detecting coordinate data of a plurality of contact points corresponding to signs by a plurality of fingers input to the touch panel every predetermined time;
A sign image extraction step of identifying one stroke information from the coordinate data of the plurality of input points for each predetermined time detected in the input coordinate detection step, and extracting the one stroke information as sign image information;
An input for extracting input point information obtained from the relationship of the coordinate data of the plurality of input points obtained at the same time among the coordinate data of the plurality of input points at the predetermined time detected by the input coordinate detection unit. A point information extraction step;
A sign information acquisition method comprising: a sign information confirmation step for confirming the sign image information and the input point information as sign information of the sign by the plurality of fingers.   The program for making a computer perform the signature information acquisition method of Claim 7.

Images