JPH07141506A - Information processing equipment - Google Patents

Abstract

(57) [Summary] [Purpose] Just by inputting a face image at the position of the system to be used, the user can verify the person himself without having to remember his / her own personal identification code (character password). It is possible to prevent being unprotected. [Structure] The face image captured by the still video camera 8 is stored in the A / D converter 2 in the image processing board 7.
1 is converted into digital information and sent to the feature extraction circuit 22. Here, the digitally converted face image information is
It is converted into parameters that best represent the characteristics of the individual. Next, in the matching circuit 23, the parameters created by the characteristic extraction circuit 22 and the previously registered characteristic parameters of the specific user are matched (pattern matching). If they match, the matching is successful.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing apparatus having a protection function at the time of opening a file or logging in, and more particularly to an information processing apparatus characterized by its user verification input method and processing. .

[0002]

2. Description of the Related Art Conventionally, an information processing device such as a computer or a document processing device such as a Desk Top Publishing is often used as a shared machine for some users. In order to make the file created by the system confidential, use a character input device such as a keyboard to enter the password code specified by the system user as a password, and check this against the code registered in advance in the device. By doing so, the user was recognized.

[0003]

However, the password processing by character input as in the above-mentioned conventional example is
If the system user forgets the personal identification code that he / she has decided, he / she will be unable to open his / her file or log in to the system, and will have to waste time for decrypting the personal identification code. Had to be forced. Also, anyone other than the person can open the file or log in to the system if he / she knows the password of the person, so it cannot be said that the file is truly protected.

An object of the present invention is to provide an information processing apparatus which solves the above problems.

[0005]

In order to achieve the above object, the present invention is an information processing apparatus having a protection function of limiting system usage rights, and an input means for inputting an image of a specific portion of a system user, and the input. It is characterized by having a matching means for matching the user from the input image input from the means.

[0006]

Embodiments of the present invention will now be described in detail with reference to the drawings.

<Embodiment 1> FIG. 1 shows Embodiments 1 to 1 shown below.
4 is a block diagram of the entire information processing apparatus according to 4, wherein 1 is a system bus, 2 is a CPU that controls the entire apparatus, 3 is a main memory used as a work area for storing programs, and 4 is a CPU without a CPU. A DMA control circuit (Direct M) that transfers data between the memory and the I / O.
Below emoryAccessController D
(Denoted as MAC), 10 is an interface with LAN 9, 11 is I / O such as ROM, SRAM, RS232C, etc.
, 13 is a hard disk, 14 is a floppy disk, 12 is a hard disk 13 or a floppy disk 1
4 interface, 16 high resolution printer,
Reference numeral 15 is a printer interface, 17 is a keyboard or mouse interface, 18 is a keyboard, 19 is a mouse as a pointing device, 9 is a local area network, 5 is a CRT control circuit including VRAM, 6 is a CRT monitor for displaying display data, 8 is a still video camera for capturing face images, 7 is a video camera 8
Image processing board that processes face image information captured by
Reference numeral 20 denotes a DSP (Di) which is a numerical operation processor for exclusively performing numerical operation when processing input information such as a face image.
(Digital Signal Processor), 2
Reference numeral 4 denotes a remote shutter for capturing a face image.

Referring to FIG. 1, after the CPU 2 is powered on,
After performing a start-up process such as a system check according to the program in the ROM of the I / O 11, the program such as the OS stored in the hard disk 13 is brought into the main memory. The application program operates according to an instruction from the user's keyboard 18 or mouse 19. The screen display on the CRT 6 is performed by the CRT control circuit 5
This is performed by the CPU 2 directly writing data to the internal video memory.

A specific implementation method of the present invention will be described with reference to the block diagram of the face image processing board 7 shown in FIG. Further, here, an example of a document processing apparatus that can protect a personal file will be described. With the still video camera 8 for reading the face image of the system user, the user uses the self-portrait for causing the device to recognize his / her face in order to open his / her file after starting the system. Take in. This will be described with reference to FIG. First, the user 27 sits down at a designated position and peels the front of the tube surface of the CRT 6, and accurate recognition images cannot be collected if the positions are different at this time. A measurement distance to 27 is set in advance, and a face image having a focus and a positional relationship is photographed by an infrared distance measuring sensor incorporated in the still video camera 8. In this way, the face image information captured by the still video camera 8 is captured by the main body 26 of the information processing apparatus through the 25 cable lines.

Returning to the description of FIG. 2, the captured face image is converted into digital information by the A / D converter 21 in the image processing board 7. This is then sent to the feature extraction circuit 22, where the digitally converted face image information is converted into parameters that best represent the individual's features. Although there are various means for extracting the personal information parameter, a method of using the contour direction as the feature of the face image pattern will be described as an example with reference to FIG. 4 in order to easily recognize the features of the face different for each person.

First, the digital information input to the 22 feature extraction circuit is converted into original image data 31 of 1024 × 1024 dots in S30. Next, in S32, in order to finely determine the contour of the face image, the original image data 31 is subjected to contour extraction processing (feature extraction means) often performed in image processing to reduce the amount of information and obtain the illustration image 33. From this illustration image 33, the characteristics of the user are best represented 32.
A dot pattern 36 of × 32 is created. There are various algorithms, but here, by performing the feature tracking process focusing on the features such as the eyes, nose, ears, eyebrow shape, mouth, and face outlines from the illustration image 33 in S34,
We will create a user feature image. Then, the dot pattern 36 extracted in S35 is used as a characteristic parameter that best represents the personal information. The feature parameter thus obtained by the personal information feature extraction circuit 22 is transferred to the personal information feature matching circuit 23. Here, circuit 2
The parameter as the personal information obtained in 2 is collated with the parameter of the specific user registered in advance. When a flag indicating that the parameter of the personal information is created in the circuit 22 is set, the CPU 2 reads the personal information data (standard pattern) registered in advance in the hard disk 13 and transfers it to the matching circuit 23. Here, the feature extraction circuit 2
The parameter created in 2 and the characteristic parameter of the specific user registered in advance are compared (pattern matching). Again, there are various matching methods. For example, when the user's characteristic parameter is input as an unknown pattern, the unknown pattern and the characteristic parameter (standard pattern) of the specific user registered in advance are respectively referred to as Euclidean distances. If the absolute distance is within a set threshold value, the unknown pattern is determined as the specific user for the registrant whose distance is determined to be the shortest. Here, the final determination that the user is the original person is made if there is a match between the administrator of the file that the user wants to open and the registrant determined by the feature matching circuit 23. The series of steps described above may be performed by software or a dedicated hardware circuit may be used. The CPU 2 may be used for these calculations, but the DSP 20 which is a dedicated numerical calculation processor may be used to reduce the load on the CPU 2 and increase the processing speed.

Next, the operation of the embodiment will be described with reference to the flow chart. First, in the flow of password processing, as shown in FIG. 5, the system user makes a request to open his or her file in S101. In S102, it is confirmed whether or not the face image password is registered. If not, the process proceeds to S103 to open the file. If registered, the process proceeds to S104. Here, the face image is input using the still video camera 8. In S105, the process described in the description of FIG. 2 is performed, and if the verification of the user is successful, the password process ends, and S
Proceed to 103. If the person cannot be collated in S105, this process is repeated. However, the number of face image inputs is counted and stored in S106, and the number of inputs equal to or more than the count number (here, three times) in S107 is counted. If the person cannot be verified even if he / she goes, the face image password processing cannot be performed. All the password processing is completed by the above procedure.

<Second Embodiment> Here, the matching circuit 23 of FIG.
Another embodiment in which the matching method in is replaced with a neural network will be described. First, an example using a hierarchical structure called a perceptron for the neural network will be described with reference to FIG. The perceptron has a structure imitating a neuron in a human brain cell, and has an input layer 109 and a first intermediate layer 11
0, a second intermediate layer 111, and an output layer 112, each of which is a unit 117 which is a neuron.
Consists of. These units are connected to each other between different layers. A weight is given to the coupling between the layers, and the way of coupling the units can be changed by changing the weight. Here, this weight changing method will be briefly described by using a learning method called a back propagation method (error back propagation method). In the input layer 109 of FIG. 6, 32 × 32 of the face image described in the first embodiment is displayed.
The binary pattern 108 of (1024 units) is input as a characteristic parameter. The network structure of the input layer is
In order to efficiently capture the two-dimensional characteristic pattern of the face image, each band of 32 vertical patterns and 32 horizontal patterns of the input layer such as 109 is input to the intermediate layer.
The characteristic pattern of this signal is analyzed through the first intermediate layer 110 and the second intermediate layer 111, and the result is output to the output layer 112. The number N of units in the output layer corresponds to the number of users (2 ^N ) who can be registered. The output signal 113 is output as a recognition user signal, and the learning error is calculated by comparing this with the teacher signal 114 as a correct answer signal. The weight correction 116 is performed each time so that this error becomes smaller. In this way, the characteristic parameters for the number of people to be registered are input many times and repeated until the error becomes small (until the learning converges).

With respect to the neural network system in which learning has converged in this way, the system user inputs a face image from the still video camera 8 shown in FIG. This face image is input as an unknown pattern to the image processing board 7 (here, it corresponds to a neural network system), A / D converted by the A / D converter 21, and then converted into a feature pattern of the face image by the feature extraction circuit 22. To be converted. The user identification process is performed by inputting the feature pattern of the face image to the neural network to which the above learning has converged. When the user thus identified matches the administrator of the file to be opened, the face image password has been verified, and the file can be opened.

As described above, by using the neural network for the person matching by the face image, even if a person who is similar to the person's face image is used, it is possible to obtain the area of the face image pattern by the learning of the neural network. Also, an effect peculiar to this embodiment that the identification ability of this system is improved can be obtained.

<Third Embodiment> Next, an embodiment in which the conventional character password input and the face image input in the first and second embodiments of the present invention are duplicated will be described with reference to the flowchart of FIG. In S118, the system user makes a request to open his or her file. In S119, it is confirmed whether or not the character password is registered, and if there is, the process proceeds to S120 and the personal identification code determined by the user is input from the keyboard 18. If the collation cannot be obtained in S121, the process returns to S119 and is repeated. If the collation is obtained, the process proceeds to S123 and the file is opened. If the character password is not registered in S119 or the password is duplicated in S122, the process proceeds to S124. Here, the face image is input using the still video camera 8. If the process described in the explanation of FIG. 2 or the process described in the second embodiment is performed in S125 and the verification of the user is successful, the password process ends and the file is opened in S123. If the person cannot be collated in S125, this process is repeated. However, the number of face image inputs is counted and set in S126, and the image input is performed at S127 or more (here, three times). If the person cannot be verified even after performing, the face image password processing cannot be performed, and the process returns to S119. The duplicated password input process is performed as described above.

As described above, by duplicating the password processing, it is possible to obtain the effect peculiar to this embodiment that the confidentiality can be enhanced more than when one password is input.

<Embodiment 4> The biggest problem with the face image password is that the human face (personal appearance) is not always constant. In other words, when you are fat and round, when you are thin and thin, when your hair is growing, when you have just cut your hair, when you are wearing glasses and when you are not. Even if the database is made to recognize different states for a certain period of time, the recognition rate cannot be improved. Therefore, in the present embodiment, a system for updating the collation database every certain period will be described with reference to FIG. In FIG. 8, the flow from inputting the face image to matching is the same as that described in the previous embodiment. However, in this embodiment, if it is determined in S127 that the matching cannot be performed three times or more, the system It is determined that the recognition rate has decreased due to the change in the face image of the user due to the time difference. At this time, it is determined in S130 whether or not the database is updated. That is, the collation database is originally in a format including time information such as the registration date. The system user compares this information with the current date every time he logs in, and performs processing to update the database when the time difference becomes larger than a certain threshold value. The period for updating the database changes depending on the lifestyle of the user (it differs depending on the group with a lot of people who don't change their appearances and the group that doesn't change it very much). In this way, when the database change is determined in S130, the process goes to S131, where the face image and the registration date are re-registered, a new database for the next password process is created, and the process ends. To do. As described above, by observing the timing of re-change of the database, it is possible to obtain the effect peculiar to the embodiment, that is, whether the password processing is speedy due to the improvement of the recognition rate.

Further, when it is considered that the system user has changed within a predetermined database update period, for example, cutting hair, thinning or thickening his / her appearance, causing a decrease in recognition rate, or a recognition rate. Does not rise much,
When it is decided that the database should be re-registered, it is possible to re-register the database with the permission of the system administrator. At this time, the system user can enter the predetermined character password provided by the system administrator, and if there is a match, perform the re-registration process in S131 so that the database can be re-registered. it can. This reduces the reduction in recognition rate due to changes in the appearance of the user,
The operability can be further improved.

<Embodiment 5> FIG. 9 is a block diagram of the entire information processing apparatus according to each of the following embodiments, and is the same as FIG. 1 except for the optical reader 80 and the image processing board 70.

A specific processing method of this embodiment will be described with reference to the block diagram of the image processing board 70 shown in FIG. Further, here, an example of a document processing apparatus that can protect a personal file will be described. Reference numeral 80 is an optical reader that captures the palm of the system user. After the user starts the system, the user first opens the file in the designated direction on the glass surface of the optical reader 80. Put on. At this time, it is necessary to align the position where the palm is placed, and for this, an optical reading device as shown in FIG. 15 is proposed. Two
00 is the main body of the apparatus, 201 is a glass surface, 202 is a palm positioning frame, and 204 is a core mark that determines the center position on the glass surface 201. On this core mark 204,
Align the center of gravity (core) of your palm like 206.

Next, the optical reader 80 scans and reads the shape of the wrinkles (ridges) of the palm.
In this way, the hand phase captured by the optical reading device 80 is converted into digital information by the A / D converter 21 in the image processing board 70. Then 22
A is a feature extraction circuit, in which the digitally converted palm information is converted into parameters that best represent individual features. Here, there are various means for extracting the personal information parameter, but since a complicated pattern feature can be easily taken, a method of using the ridge direction as the feature of the palm pattern will be described as an example with reference to FIG. First, the digital information input to the personal information feature extraction circuit of 22A is S300.
Is converted into the original image data 350 of 256 × 256 dots. Next, in order to finely determine the flow of palm ridges, the region of 8 × 8 dots is divided as one processing region in S310, and the projection with respect to nine kinds of direction axes is used to determine the optimal ridge direction. Select one. Where 8
Within the unit area of × 8 dots, each pixel has 9 kinds of direction codes, the 1st to 8th correspond to 8 directions between 0 and π, and the 9th is a code with no ridge direction. did. In this way, the direction data (360) of 32 × 32 dots is obtained in S320. Next, a ridge tracking process S330 is performed in order to track the ridge-shaped features that have been turned into direction data without impairing them and to extract a palm feature pattern. In the process here, the peculiarities of the direction data are eliminated to some extent, and macro-tracking is performed so as not to give too much importance to the features of the fine ridge shape. First, a core showing the center of gravity of the palm, which has an arc shape or a rectangular shape, is obtained as a reference point for palm matching. Next, this core is set as a tracking target point, and a ridge line is tracked according to a predetermined rule in relation to the direction code around this target point. In this way, the palm pattern 32 × 32 dot direction pattern 360
Is S in the binary characteristic pattern 370 of 16 × 16 dots.
Converted at 340. This is a characteristic parameter that expresses the characteristics well. The feature parameter thus obtained by the feature extraction circuit 22A is transferred to the feature matching circuit 23A. Here, the parameter as the characteristic obtained by the circuit 22A is checked against the parameter of the specific user registered in advance. When a flag indicating that the characteristic parameter has been extracted by the circuit 22A is set, the CPU 2 reads the personal information data (standard pattern) registered in advance in the hard disk 13 and transfers it to the characteristic matching circuit 23A. Here, the parameter created by the circuit 22A and the characteristic parameter of the specific user registered in advance are compared (pattern matching). Again, there are various matching methods. For example, when the user's characteristic parameter is input as an unknown pattern, the unknown pattern and the characteristic parameter (standard pattern) of a specific user registered in advance are respectively referred to as Euclidean distances. If the absolute distance is within a certain set threshold for a registrant who is calculated to have the shortest distance among them,
The registrant is judged as the user to be judged. Here, the final determination that the user is the original person is made if there is a match between the administrator of the file that the user wants to open and the registrant determined by the characteristic matching circuit 23A. The series of steps described above may be performed by software or a dedicated hardware circuit may be used. The CPU 2 may be used for these calculations, but the DSP 20 which is a dedicated numerical calculation processor may be used to reduce the load on the CPU 2 and increase the processing speed.

Next, the operation of the present embodiment will be described with reference to the flow chart shown in FIG. 12, except that the manual phase input is performed by using the optical reader 80 in S104A.
Is the same as.

<Embodiment 6> An embodiment in which the matching method in the personal information matching circuit 23 of FIG. 10 is replaced with a neural network will be described with reference to FIG. 16 × 16
It is the same as FIG. 6 except that it is a binary pattern 108A of (256 units). The network structure of the input layer is designed to efficiently capture the two-dimensional fingerprint feature pattern.
16 vertical patterns of the input layer such as 109, and 16 horizontal patterns
Each band of the pattern is as an input to the hidden layer. The characteristic pattern of this signal is analyzed through the first intermediate layer 110 and the second intermediate layer 111, and the result is output to the output layer 112. The number N of units in the output layer corresponds to the number of users (2 ^N ) who can be registered. The output signal 113 is output as a recognition user signal, and the learning error is calculated by comparing this with the teacher signal 114 as a correct answer signal. The weight correction 116 is performed each time so that this error becomes smaller. In this way, the characteristic parameters for the number of people to be registered are input many times and repeated until the error becomes small (until the learning converges).

In the neural network system in which the learning is converged in this way, the system user inputs a manual phase from the optical reader 8. This phase is input to the image processing board 7 (here, it corresponds to a neural network system) as an unknown pattern, A / D converted by the A / D converter 21, and then converted into a phase characteristic pattern of the phase by the feature extraction circuit 22. It The user identification process is performed by inputting this palm feature pattern into the neural network to which the learning has converged. When the user thus identified matches the administrator of the file to be opened, it means that the personal identification of the palm password has been performed, and the file can be opened.

As described above, by using the neural network for the personal identification by the palm, even if a person who is similar to the palm of the principal uses it, the system can be obtained by the area dividing ability of the palm pattern by learning of the neural network. An effect peculiar to this embodiment is also obtained in that the discrimination ability of is improved.

<Embodiment 7> Next, an embodiment in which the conventional character password input and the palm input in Embodiments 5 and 6 of the present invention are duplicated will be described with reference to the flowchart of FIG. 7 is the same as that of FIG. 7 except the manual input by the optical reader 80 in S124A.

As described above, by duplicating the password processing, it is possible to obtain the effect peculiar to this embodiment that the confidentiality can be enhanced more than when one password is input.

<Embodiment 8> Although palms have been used as personal information in the above-mentioned embodiments, the number of registrants using the system may increase the possibility of individual verification. Therefore, in the present embodiment, in order to further improve the collation rate, an example will be given in which fingerprint information, which is personal information that is immovable to all people, is incorporated and personal information duplicated with palm information is packed as one database.

A brief description will be given of a method of collecting fingerprints, which will be described below with reference to the glass surface 2 of the optical reader 200 shown in FIG.
Fingerprints are attached to the glass surface by pressing a finger on which the stamp ink is attached to 01. Here, the fingerprint is positioned, but the core as the center of the fingerprint is aligned with the core 204 in the fingerprint positioning frame 203 as in the case of the palm. Here, the reading of the fingerprint is started by lowering the cover of the optical reading device and turning on the switch.
Fingerprint input is performed as described above. Also, the glass surface that is dirty due to fingerprints should be wiped clean for the next person.

In this fingerprint collation method, a 16 × 16 characteristic pattern is created by using the same algorithm as the above-mentioned palm phase collation. By creating a database in which the feature pattern of the fingerprint and the feature pattern of the palm are combined, the distinction becomes easy and the matching rate is improved even when the discrimination of the personal feature is difficult only by the palm. This method also has an effect peculiar to this embodiment, which is particularly advantageous when the number of system users is large.

<Embodiment 9> Although palm has been used as personal information in the above-mentioned embodiment, personal verification may become difficult as the number of registrants using the system increases. Here, in the present embodiment, in order to further improve the collation rate, an example in which the size of the user's hand is measured and personal information in which the size of the hand is duplicated is packed as one database will be given.

Next, how to recognize the size of the hand will be described with reference to FIG. The image read by the optical reading device looks like 215, and black is formed in the portion where the hand is not placed, and black and white in the other portions. Here, the size of the hand is defined by the distances 210, 211, 212, 213 and 214 from the palm core 216 (which matches the core 204 in FIG. 15) to the five fingers. That is, it can be seen that the size of the hand can be determined by calculating the distance from the core to the white dot and extracting the largest 5 points. These five distances are used as a characteristic pattern depending on the size of the hand.

By creating a database in which the feature pattern according to the size of the hand and the feature pattern according to the palm are combined, it becomes easy to distinguish the personal feature even when the palm alone is difficult, and the collation rate is improved. To do.
This method also has an effect peculiar to this embodiment, which is particularly advantageous when the number of system users is large.

<Embodiment 10> Here, since the pattern of the palm input described so far is simpler than that of a fingerprint or the like, by placing a copy of the palm on the optical reading device, a false password can be input by another person. There is a risk that it will be performed. Therefore, in this embodiment, it is confirmed that the actual hand is on this problem,
If not, the function is added to prevent the file from being opened.

Specifically, the temperature / pulse sensor 205 is attached to the apparatus 200 as shown in FIG. The sensor 205 is brought into contact with the wrist like 207 at the time of palm input so that the body temperature of the user and the pulse rate for one minute can be sensed. Before entering the palm-matching mode, the system user's body temperature / pulse data that has been registered in advance is matched, and the palm-matching mode is entered only when it falls within a predetermined range. By adding such a function, it is possible to prevent another person from opening the file using the copied palm and improve the confidentiality of the file.

[0037]

As described above, the following effects can be obtained by inputting a password using an image of a specific portion of the human body such as a face or a palm.

1. Even if the system user forgets the personal identification code decided by himself, opening his own file,
Or you can log in to the system. Also,
You don't have to remember it.

2. Even if another person knows his / her own personal identification code, the confidentiality can be maintained without being unprotected by another person.

3. For password input, the troublesome work of keyboard input is released, and processing can be performed smoothly.

4. Faces or hands have the characteristics that no humans have the same face or palm (everyone is the same), and they do not change for the rest of their lives (invariant throughout life) unless they are damaged. Since it can be stable information that is not affected, the reliability of user verification is improved.

[Brief description of drawings]

FIG. 1 is a block diagram of an information processing apparatus according to first to fourth embodiments of the present invention.

FIG. 2 is a block diagram of an image processing board of the embodiment.

FIG. 3 is a diagram showing system settings.

FIG. 4 is an operation flowchart of password processing.

FIG. 5 is a flowchart of a feature pattern extraction process for a face image.

FIG. 6 is a structural diagram of a neural net (perceptron structure).

FIG. 7 is an operation flowchart of password duplication processing.

FIG. 8 is a flowchart of a database change process.

FIG. 9 is a block diagram of an information processing apparatus according to examples 5 to 10 of the present invention.

FIG. 10 is a block diagram of the image processing board of the embodiment.

FIG. 11 is an operation flowchart of password processing.

FIG. 12 is a flowchart of a palm feature pattern extraction process.

FIG. 13 is a structural diagram of a neural net (perceptron structure).

FIG. 14 is an operation flowchart of password duplication processing.

FIG. 15 is an image diagram of an optical reading device for inputting a manual phase password.

FIG. 16 is an image diagram of a method for measuring the size of a hand.

[Explanation of symbols]

 1 System Bus 2 CPU 3 Main Memory 4 DMAC 5 CRT Interface 6 CRT Display 7 Image Processing Board 8 Still Video Camera 9 LAN 10 LAN Interface 11 I / O 12 Disk Interface 13 Hard Disk 14 Floppy Disk 15 Printer Interface 16 Printer 17 Keyboard and Mouse Interface 18 Keyboard 19 Mouse 20 Digital signal processor 21 A / D converter 22 Feature extraction circuit 23 Feature matching circuit

Claims (11)

[Claims] 1. An information processing apparatus having a protection function of limiting system usage rights, input means for inputting an image of a specific portion of a system user, and collation for collating the user from an input image input from the input means. An information processing apparatus comprising means. 2. The matching means according to claim 1, wherein the matching means extracts personal characteristics from the input image, registration means in which personal characteristics of a plurality of individuals are registered in advance, and the extracted personal characteristics. An information processing apparatus, comprising: a comparison unit that compares the personal characteristics with the personal characteristics registered in advance. 3. The matching means according to claim 1, further comprising means for extracting a personal characteristic from an input image and identification means having a learning function, wherein the collating means is based on the extracted personal characteristic. An information processing apparatus characterized in that a user is given a system use right by an identification means. 4. The apparatus according to claim 1, further comprising means for inputting a character password, and means for determining whether or not to give the user system utilization right by duplicating the input character password and the input image. An information processing device characterized by: 5. The method according to claim 2, wherein a database of personal characteristics registered in advance by the registration means is provided with registration date information, and the registration date information is compared with the date and time at the time of login to make a fixed decision. An information processing apparatus having means for updating a database for each period. 6. The information processing apparatus according to claim 2, wherein the registration means can update a database of personal characteristics registered in advance. 7. The information processing apparatus according to claim 1, wherein the specific location of the user is a face. 8. The information processing apparatus according to claim 1, wherein the specific location of the user is a hand. 9. The information processing apparatus according to claim 8, wherein the input unit includes a palm input unit and a fingerprint input unit. 10. The information processing apparatus according to claim 8, wherein the input unit includes a palm input unit and a hand size determination unit. 11. The method according to claim 8, wherein the collating means has a means for sensing a body by a temperature and pulse sensor,
An information processing apparatus, wherein the sensing result of the sensing means is suitable for user verification.