JPS6190641A - Indivisual identification apparatus - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a device for identifying a corresponding individual using a fingerprint image, and in particular to an apparatus for identifying a corresponding individual using a fingerprint image, and in particular, it is possible to improve image memory by setting a plurality of window areas on a fingerprint image. This is to reduce capacity and shorten processing time.

[Conventional technology]

FIG. 6 is a block diagram showing an example of a conventional personal identification device for identifying an individual using a fingerprint image, in which 1 is a finger as a separate body for returning to work, and the fingerprint of this finger 1 is input as an image. The data is taken into the device 2 and converted into image data. 3 is a feature data extraction device that extracts feature data included in all image data by taking in all the image data output from the image input device 2; , a switch for switching between registration and verification, and an input device 5 which manually inputs the switching signal 6 of the switch 4 and the registration number signal 7't. Reference numeral 8 denotes a data storage device, which stores feature data extracted by the feature data oil field device 3 at the time of registration in advance in correspondence with the registration number signal 7 supplied from the input device 5. Reference numeral 9 indicates return-to-work feature data 10 extracted by the feature data storage device 3 during verification and registered feature data 11 stored in the feature data storage device 8 in correspondence with the registration number signal 7.
This is a comparison/determination device that compares and verifies the results and outputs the determination result signal 12.

In the individual identification device configured in this way, when a registration number is input into the input device 5 with the finger 1 placed on a predetermined position of the image input device 2 during registration, the registration number signal 7 is displayed as a characteristic. The switch 4 is set to the registered state by a switching signal 6 which is supplied to the data storage device 8 and which is generated from the input device 5. Next, from all the image data output from the image input device 2, only feature data is extracted in the feature data extraction device 3. The feature data extracted by this feature data extraction device 3 is
'It is stored in the feature data storage device 8 via the switch 4 in correspondence with the registration number signal 7.

Next, at the time of verification, when the registration number is inputted from the input device 5, the registration number signal 7 is outputted, the switching signal 6 is outputted, and the switch 4 is set to the verification state.

On the other hand, the reinstatement-specific feature data 10 extracted from the feature data extracting device 3 is supplied to the matching determination device 9. Further, registered feature data 11 stored in advance in the feature data storage device 8 in correspondence with the registration number signal 7 is supplied to the verification determination device 9 via the switch 4. Therefore, the verification determination device 9
compares and collates the reinstatement-specific characteristic data 10 and the registered characteristic data 11, and outputs the determination result as a determination result signal 12.

[Problem that the invention seeks to solve]

However, in the personal identification device having the above configuration, since feature data is extracted from all images, the scale of the device becomes large and the processing speed becomes slow. In addition, since all the position coordinates of points called feature points where the ridges in the fingerprint image are broken or branched as feature data are used, when the entire screen is targeted, It has various drawbacks such as high cost due to the need for complicated processing.

[Means for solving problems]

Therefore, the individual identification device according to the present invention extracts feature data within a plurality of predetermined window regions, and also extracts feature data such as directions of ridges and valleys or spatial frequencies within the window region. It uses feature data.

[Effect]

Since the individual identification device configured in this way processes characteristic data within a specific window area, the processing is simple and high-speed processing can be performed easily and at low cost. This is the result.

[Embodiment] FIG. 1 is a block diagram showing an embodiment of the personal identification device according to the present invention, and the same parts as in FIG. 6 are indicated using the same symbols. In the figure, reference numeral 13 denotes a window area setting screen provided between the image input device 2 and the feature data extraction 1fl device 3. Set the window area. The difference from FIG. 6 is that this window area setting mechanism 13 is provided.

In the personal identification device configured in this manner, the fingerprint image of the finger 1 is converted into an image signal by the image input device 1.

For the image signal output from the image input device 1, a plurality of window areas are set at predetermined positions in the window area setting mechanism 13. In other words, in the fingerprint image 14 shown in FIG.
Nine window areas 15 consisting of columns are designated. Then, only the image data within each window area 15 is provided to the feature data extraction device 3. Therefore, in this case, the amount of input image signals to the feature data extraction device 3 is significantly reduced compared to the conventional method, so the amount of data processing in the feature data extraction device 3 is significantly reduced. , the processing time will be reduced. Furthermore, since the amount of feature data to be extracted is also reduced, the amount of storage in the feature data storage device 8 is reduced and the capacity is reduced, resulting in cost reduction. Note that the feature data registration and verification process in this case is performed in the same manner as in the case described above.

FIG. 3 is a diagram for explaining another implementation sequence of the personal identification device according to the present invention, in particular when extracting the statistically dominant direction of the ridge bundle or valley bundle in each window area as feature data. FIG. In the figure, reference numeral 16 is a window area, within which a ridge bundle or valley line bundle 17 of the fingerprint is displayed.

Since this ridge bundle or valley line bundle 1T has a dominant direction that is statistically uniquely determined, this direction line 18
is the sign 2 formed by the reference direction 19 common to each window area.
The angle θ shown as 0 is assumed to be all feature data. Note that if the dominant direction of the ridge bundle or valley bundle 1T cannot be uniquely determined at the center of the spiral pattern, etc., it may be used as characteristic data by giving a quantity θ that satisfies this intention.

FIG. 4 shows the spatial frequency '13! due to the statistical control between ridge bundles or valley bundles in each wind region. , is used as a t-trf characteristic, and the normal direction 2 to the direction 18 of the ridge bundle or valley bundle 17 in the window region 16 is shown.
1, the spatial period 22 of ridges or valleys that cross direction 1 is
Similar to 8, it is determined by statistical uniqueness. Then, the spatial frequency, which is the reciprocal of this spatial period 22, can be used as feature data.

FIG. 5 shows a method for extracting feature data when the direction 18 of the ridge bundle or valley bundle 1T is not uniquely determined at the center of the spiral pattern. In this case, an appropriate reference direction 19 is determined, and a spatial period 22 crossed by this M quasi-direction 19 is determined and used as the spatial frequency total feature data. Further, by using a combination of the direction of the ridge or valley line and the spatial frequency as feature data, highly accurate identification becomes possible.

In each of the above embodiments, a case has been described in which the signal sounds of each window region are processed in common, but it goes without saying that an independent image sensor and feature data extraction device IT may be provided for each window region. In this case, high-speed processing due to parallelization becomes possible.

〔Effect of the invention〕

As explained above, since the personal identification device according to the present invention extracts feature data only from within a plurality of predetermined window areas, the data processing speed is increased. At the same time, even a small capacity image processing memory becomes sufficient. Furthermore, by using the direction or spatial frequency of the ridges or valleys of the fingerprint within the window area as feature data, the algorithm for data processing becomes simpler and the processing time is shortened. It has various excellent effects such as cost reduction.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of a personal identification device according to the present invention, FIG. 2 is a plan view showing the relationship between a fingerprint image and a window area, and FIGS. 3 to 5 are other embodiments of feature data extraction. FIG. 6 is a plan view showing the PI, and a block diagram showing a conventional array of personal identification devices. 1 is a finger, 2 is an image input device, 3 is a feature data extraction device,
4 is a switch, 5 is an input device, 8 is a feature data storage device, 9 is a collation determination device, and 13 is a window area setting machine model 6. In the drawings, the same reference numerals indicate the same or corresponding parts. Agent: Masuo Oiwa (2 others) Figure 2 Figure 3 Procedural amendment (self-motivation j 19 1, case description Patent Application No. 1982-211927 2,
Title of the invention Personal identification device 3, person making the amendment 5, subject of amendment (1) Claims column of the specification (2) Detailed explanation of the invention column 6 of the specification, contents of the amendment (1) Amend the claims of the specification as in the attached sheet. (2) In the 10th line of page 3 of the specification cylinder, the phrase "separate body for reinstatement" is corrected to read "identified body." (3) On page 4, lines 2 and 199, the words ``correspondence'' are corrected to ``correspondence.'' (4) Page 4, lines 3 to 4 “Characteristic data by reinstatement”
The text is corrected to read "characteristic data to be identified." (5) On page 4, line 8, the phrase "individual identification device" is amended to read "personal identification device." (6) Correct the phrase "switching signal" on the second line of page 5 to read "switching signal." (7) “Special data by reinstatement” on page 5, lines 4 and 9.
Correct each text to be "identified feature data." (8) On page 6, lines 3 and 9, ``Is the term ``individual identification device J'' replaced with ``personal identification device''? Incorrect. (9) On page 9, line 1, the word "matters" is amended to read "data." θω Do the same 9th harvest 4 Maru. ” should be corrected to “-Determined at will.” 7. Attached documents (11? Document stating the entire text of the claims after IO 1 document stating the entire text of the claims after amendment 2. Claims (1) Capturing fingerprints as an image an image input device that converts the image signal into an image signal and outputs it; a window area setting mechanism that sets a plurality of window areas in the image signal output from the image input device; A feature data extraction device that extracts feature data, a feature data storage device that stores the extracted feature data, and a comparison comparison between the identification data of the fingerprint to be verified and the feature data stored in the feature data storage device. and a verification device that outputs the determination result signal, a switch that switches the flow of signals at the time of registration and verification, and an input device that receives the registration number signal and the switching signal of the switch. A personal identification device characterized by: (2) An individual according to claim 1, characterized in that an image input device, a window area setting mechanism, and a feature data extraction device are provided independently according to each window area. Identification device. (3) A direction based on statistical control with respect to a reference direction common to each wind region of a ridge bundle or a valley bundle taken into each wind region is used as feature data. or the personal identification device according to claim 2. (4) Claim 4, characterized in that spatial frequencies based on statistical control between ridge bundles or valley bundles taken into each window area are used as feature data. Personal identification device according to paragraph 1 or paragraph 2. (5) Direction and distance between ridge bundles based on statistical control with respect to a reference direction common to each window of ridge bundles or valley bundles taken into each window area as feature data. The personal identification device according to claim 1 or 2, characterized in that it uses a combination of empty 17JJ frequencies based on statistical control between Aroibani line bundles.

Claims (5)

[Claims] (1) An image input device that captures a fingerprint as an image, converts it into an image signal, and outputs it; a window area setting mechanism that sets multiple window areas for the image signal output from this image input device; a feature data extraction device that extracts feature data from image data in a window area; a feature data storage device that stores the extracted feature data; A comparison/judgment device that compares and verifies the characteristic data that is stored and outputs a judgment result signal, a switch that switches the flow of signals at the time of registration and verification, and an input that receives the registration number signal and the switching signal of the switch. A personal identification device comprising: (2) The personal identification device according to claim 1, characterized in that an image input device, a window area setting mechanism, and a feature data extraction device are provided independently according to each window area. (3) Claims 1 or 2 characterized in that a direction based on statistical control with respect to a reference direction common to each window area of a ridge bundle or a valley line bundle taken into each window area is used as the feature data. Personal identification devices as described in Section. (4) The personal identification device according to claim 1 or 2, characterized in that the spatial frequency based on statistical control between ridge bundles or valley bundles taken into each window area is used as the feature data. . (5) Combining the statistically dominated direction and the statistically dominated spatial frequency between the ridge bundles or valley bundles with respect to the reference direction common to each window of the ridge bundles or valley bundles imported into each window area as feature data. The personal identification device according to claim 1 or 2, characterized in that it is used as a personal identification device.