JPH08161490A - Fingerprint input method - Google Patents

Abstract

PURPOSE: To correct an angle of a received fingerprint image with a simple method by extracting a tangential line of two fingers and obtaining a straight line between picture elements having a maximum value and a minimum value at XY coordinates of a thinned image of the tangential line so as to correct an angle of a fingerprint input image so that the straight line is set vertical. CONSTITUTION: A fingerprint image including two fingers adjacent with each other is entered, an input image is binarized by an image processing section IPU, a tangential line of the two fingers is extracted by contracting/expanding the binary image, and the extracted tangential line is subject to thinning processing. A central processing unit(CPU) calculates and extracts picture elements having a maximum value of the Y coordinate or a minimum value of the X coordinate, and a minimum value of the Y coordinate, or a maximum value of the X coordinate in the image subject to thinning processing. Then a straight line between the two picture elements is obtained and a cross angle where the straight line is in parallel with the Y axis and at a right angle with the X axis is obtained and all picture elements received at first and registered are rotated by the cross angle to correct the angle of the received image.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fingerprint input method for fingerprint collation or the like.

[0002]

2. Description of the Related Art In fingerprint collation, a pass / fail judgment is made by matching with a fingerprint registered in advance. The pre-registered fingerprint is
Although it is registered in the horizontal and vertical directions at the correct input angle, it is rare that the fingerprint image is input at the correct angle when the fingerprint data is input. Therefore, when performing fingerprint matching, it is necessary to correct the angle of the fingerprint input, but there is no fingerprint input method that electronically easily corrects the input angle, and conventionally, using a mechanical finger guide, etc. The installation angle itself was regulated.

[0003]

SUMMARY OF THE INVENTION The present invention was devised to solve such conventional problems, and provides a fingerprint input method for correcting the angle of an input fingerprint image by a simple method. The purpose is to

[0004]

In order to achieve the above object, the fingerprint input method according to the present invention inputs a fingerprint image including two fingers which are in contact with each other, and binarizes, contracts, and expands the fingerprint image to obtain two fingers. The tangent line of is extracted, the tangent line is thinned, the straight line between the pixels having the maximum value and the minimum value at the Y coordinate or the X coordinate of the thinned image is obtained, and the angle of the fingerprint input image so that the straight line becomes vertical. Is configured to correct.

[0005]

According to the fingerprint input method of the present invention, a tangent line of a finger can be extracted by inputting a fingerprint image including two fingers in contact and binarizing, contracting and expanding the fingerprint image, and after thinning the tangent line. The tangent line can be obtained as a straight line from the maximum value and the minimum value of the Y coordinate of.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, one embodiment of a fingerprint input method according to the present invention will be described with reference to the drawings. FIG. 1 is a flow chart for implementing the present invention. FIG. 5 is a block diagram of an apparatus for carrying out the present invention. The apparatus is composed of, for example, a central processing unit CPU, an image processing unit IPU and an image memory IM.
The central processing unit CPU, the image processing unit IPU and the image memory IM are connected via a system bus SB, and the image processing unit and the image memory IM are connected by a local bus LB.

First, a fingerprint image including two fingers touching each other is input from an input device (not shown) (step 1-1),
It is registered in the image memory IM (step 1-2). next,
An input pixel is input from the image memory IM via the local bus LB by an instruction from the central processing unit CPU.
Send to In this image processing unit IPU, the input image is binarized (step 1-3) as shown in FIG. 2, and then the binary image is subjected to contraction processing (step 1-4). Since the tangent line of the two fingers is thicker than the joint line and the fingerprint, the pixels of the tangent line of the two fingers remain until the end even if the contraction process is performed. Next, the pixels remaining until the end are subjected to expansion processing a predetermined number of times (step 1-
5) Extract only the tangents of two fingers.

Next, as shown in FIG. 3, the image subjected to the expansion process is subjected to a thinning process (step 1-6). In this way, it is possible to obtain the approximate center line of the tangent line. The pixel (Ymax) having the maximum value of the Y coordinate in this thinned image
Alternatively, the pixel having the minimum value of the X coordinate (Xmin) and the pixel having the minimum value (Ymin) or the pixel having the maximum value (Xmax) are calculated and extracted by the central processing unit CPU (step 1-7). That is, these two pixels are the uppermost or leftmost part and the lowermost or rightmost part of the tangent line of two fingers.

Next, as shown in FIG. 4, a straight line L between these two pixels is obtained (step 1-8), and this straight line is made parallel to the Y axis or at right angles to the X axis. For example, the central processing unit CPU calculates an angle for rotating the fingerprint image about a pixel having the maximum value of the Y axis (minimum value of the X axis) in the thinned figure, that is, the intersection angle θ between the straight line L and the Y axis. (Step 1-9). Next, step 1
The angle of the input image is corrected by rotating the coordinates of all pixels of the image registered in -2 by θ (step 1-10). This ensures a proper fingerprint input image.

[0010]

As described above, according to the present invention, the contact 2
Input a fingerprint image containing two fingers, binarize, contract and expand to extract the tangents of the two fingers, thin the tangents,
A straight line between pixels having the maximum value (minimum value) and the minimum value (maximum value) at the Y coordinate or the X coordinate of this thinned image is obtained, and the angle of the fingerprint input image is set so that this straight line has a predetermined angle state. Since the correction is performed, the angle of the input fingerprint image can be corrected by a simple method, and for example, it has an effect of being able to assist in highly accurate fingerprint collation.

[Brief description of drawings]

FIG. 1 is a flowchart showing an embodiment of the present invention.

FIG. 2 is a conceptual diagram of a binary image.

FIG. 3 is a conceptual diagram of a thinned image.

FIG. 4 is a conceptual diagram for explaining angle correction by rotation.

FIG. 5 is a block diagram of an apparatus for implementing the present invention.

[Explanation of symbols]

 CPU Central processing unit IPU Image processing unit IM Image memory SB System bus LB Local bus X X axis Y Y axis Ymax Pixel having the maximum value of Y coordinate Ymin Pixel having the minimum value of Y coordinate Xmax Pixel having the maximum value of X coordinate Xmin pixel having the minimum value of X coordinate L straight line θ intersection angle

Claims (1)

[Claims] 1. A fingerprint image containing two touching fingers is input, the input image is binarized, and the binary image is contracted and expanded to extract the tangent lines of the two fingers, and the extraction is performed. The tangent line created is thinned, and the Y coordinate or X of this thinned image
A fingerprint input method characterized in that a straight line between pixels having a maximum value and a minimum value in coordinates is obtained, and the angle of fingerprint input is corrected based on this straight line.