JPS62196774A - Information deciding device - Google Patents

Abstract

PURPOSE:To make double check by comparing the code of identification number inputted by handwriting with existing code, and further, deciding whether handwriting of the code such as identification number, etc., inputted by handwriting is the person's own or not. CONSTITUTION:A user inputs the identification number by handwriting on a tablet 2 with a pen 1. A character recognizing section 3 recognizes inputted identification number by using a selected individual dictionary section 4 and sends the result of recognition and the degree of resemblance R to a CPU 5. Then, the identification number registered in a dictionary section 6 in a BUF and the result of recognition are compared, and when they coincide, compared with the value R for which the degree of pattern resemblance (r) is determined beforehand, and examine whether it is within allowable extent. In case where input is made by a person other than qualified persons, deviation from patterns registered in an individual dictionary section 4 is though to become large, and the value of (r) becomes larger. Accordingly, the person is decided to be a qualified person when r<=R.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to an information determination device for manually inputting information such as a password in an information system.

[Conventional technology] Conventionally, in various information systems, in order to maintain confidentiality, etc.
To determine whether a user is a qualified person, the user is usually asked to input a personal identification number and this is compared with a number previously registered in the central processing unit. There are various methods for entering a PIN number, but when manually writing it online, the entered character pattern is recognized against a character dictionary, and the result is compared with the registered number. and made a judgment.

[Problems to be Solved by the Invention] However, in this case, character patterns vary greatly from person to person, and on the other hand, character dictionaries are general and do not depend on individuals, so manually entered passwords may be misrecognized.
There is a drawback that there is a possibility that a person who is qualified may not be determined to be a qualified person, and it is difficult to prevent this from happening if the password is known by another person.

The present invention was developed in view of the above-mentioned shortcomings of the prior art.By performing character recognition using a dictionary for each individual, the risk of misrecognition is reduced, and the character pattern of the numbers input at the same time is registered in the dictionary. By taking into account the distance between the character patterns that are entered, the risk of being misidentified as a qualified person when input by someone other than a qualified person is reduced, and as a result, the confidentiality of the information system is improved. The purpose is to

[Means for Solving the Problem] As a means for solving this problem, in the embodiment of the present invention, there is a method in which the personal information of the user inputted by the user from the input section exists in the dictionary section. and a second determining means that determines whether the input personal information is the user's own information input from the input section.

[Operation] In the configuration of this embodiment, the handwritten code such as a PIN number is compared with the existing code, and it is further determined whether the handwriting of the handwritten code such as the PIN number is the person's own. A double check is performed by determining whether

[Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the system configuration of this embodiment. In the figure, 1 is a pen for inputting characters, 2 is a tablet for inputting characters, and 3 is a character recognition unit. 4 is a personal dictionary section for character recognition having character patterns or character characteristics for individual character recognition; 5 is a central processing unit C;
PU, 6 is a dictionary section in which personal data of qualified persons including personal codes, passwords, etc. of qualified persons are stored.

Next, the operation of this embodiment will be explained according to the flowchart shown in FIG. First, a personal code is manually entered from the tablet 2 into the area IN of the RAM 8 in step S1. This manual method is not special; for example, it may be performed manually using a keyboard or the like. In addition, the personal code in this case may be an employee number if it is within the company, or may be a number and name coded for the department.

Next, in step S2, the dictionary section 6 is searched for the manually entered personal code, and the password and personal dictionary selection information are stored in the RAM.
Put it in BUF of 8. The CPU 5 then operates the character recognition unit 3.
The user instructs the individual dictionary section 4 to be selected. Character recognition section 3
selects one of the individual dictionary units 4 according to selection information obtained via the CPU 5. Next, the process moves to step S3, and the user uses the pen 1 to input the password on the tablet 2 by hand. Then, in step S4, the character recognition section 3 uses the selected personal dictionary section 4 to recognize the manually entered password, and sends the recognition result and pattern similarity R to the CPU 5.

Here, especially in the case of character recognition, the distance between the human character pattern and the registered character pattern in the pattern space is determined regardless of the recognition method. Therefore, from the pattern pressure 11idi (i=1.2°3..., m) for each human character,
The value r obtained using a certain function f is defined as the pattern similarity, and is defined as in the following equation.

r=f (dl, d2, ..., d
m) r is a measure indicating how close the input character string as a whole is to the pattern registered in the individual dictionary section 4.

Next, in step S5, the recognition result is compared with the password registered in the dictionary section 6 included in the BUF, and it is checked whether or not they match. If they do not match, the process moves to step S8, and the recognition result is checked. Either the user is determined to be unqualified, or the user is requested to re-enter the information. If they match, the process moves to step S6, where the pattern similarity r is compared with a predetermined value R to see if it is within an allowable range.

If the input is made by someone other than a qualified person, it is considered that the deviation from the pattern registered in the personal dictionary section 4 will be large, and the value of r will be large. Therefore, in the case of r≦R, the process proceeds to step S7, and it is determined that the person is within the permissible range and is a qualified person. On the other hand, if r>R, the process moves to step S8 and it is determined that the person is not a qualified person.

As described above, by determining the handwriting of the handwritten password or its characteristics, it is possible to obtain higher confidentiality.

As mentioned above, Kimi-ryu uses a personal identification number (number) as a key to maintain confidentiality, but considering the flexibility of online handwritten input, it is possible to use letters, kana, and other characters instead of numbers. Of course, it is also possible to use kanji.

[Effects of the Invention] As described above, according to the present invention, it is possible to improve the confidentiality of information, and it is also possible to reduce the risk of misrecognition in character recognition.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the system configuration of an embodiment of the present invention, and FIG. 2 is a flowchart showing the flow of the operation according to FIG. In the figure, 1... Input pen, 2... Tablet, 3 Character recognition section, 4... Personal dictionary section, 5... CPU, 6...
. . . dictionary section, 7 . . . ROM, 8 . . RAM.

Claims (5)

[Claims] (1) A first determining means for determining whether or not personal information of an individual user inputted by the user from the input section exists in the dictionary section; An information determining device comprising: second determining means for determining whether or not the information belongs to the user. (2) The information determining device according to claim 1, wherein the personal information is input by hand. (3) The information determining device according to claim 1 or 2, wherein the second determining means determines from the handwriting of personal information input by the user into the input unit or its characteristics. . (4) The information determining device according to claim 3, further comprising a personal dictionary that stores handwriting or characteristics of personal information for each user. (5) The information determining device according to any one of claims 1 to 4, wherein the personal information is a password.