JPH08110938A - Method for inputting finger print - Google Patents

Abstract

PURPOSE: To surely judge whether an inputted finger print picture is suitable for collation e.g. or not, by dividing the whole input picture into plural blocks and judging the suitableness of collation or the like based upon the rate of the number of proper picture blocks to the total number of blocks. CONSTITUTION: An input picture is binarized. The whole binary picture is divided into plural blocks and the area of each divided block is found out. The area is found out by counting up the number of picture elements in each block. Blocks consisting of only a background and their adjacent blocks are removed. The number of blocks respectively having a prescribed number of picture elements or more and the number of blocks having a prescribed number of picture elements or less out of remaining blocks are found out and the rates of the obtained numbers of blocks to the number of remaining blocks are respectively calculated. When each rate is less than a prescribed value, the input picture is regarded as a picture suitable for collation and the binary picture is taken in. When the rate is more than the prescribed value, many defective picture blocks caused by excess or lacking finger pressure are included, the input picture is judged as an unsuitable picture and a reinput instruction is displayed.

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 What kind of fingerprint image is taken in when performing fingerprint matching is a very important issue in terms of matching accuracy. As a well-known fingerprint input method, there is a method of determining whether or not to capture the input image as an image suitable for matching based on whether or not the binary image area of the fingerprint image is within a predetermined range. However, since this method makes judgments based on the area of the entire input fingerprint image, for example,
Alternatively, even an image that is not suitable for collation due to the presence of finger pressure unevenness at a portion of the finger such as the left and right portions, may be captured as an appropriate image if the area of the entire image is within a predetermined range.

In this conventional method, the area is obtained by counting the number of pixels. However, for example, when the amount of holding down of the finger is weak and feature points such as ridges and branch points of the fingerprint are not sufficiently input. On the other hand, when the input image is binarized, such as when the finger is strongly pressed and the ridges of the fingerprint are crushed, the number of pixels in the portion where the finger pressure is insufficient decreases compared to the proper finger pressure. The number of pixels in the portion where the finger pressure is excessive increases.
FIG. 4 shows an example of a binary image of insufficient acupressure and FIG. 5 shows a conceptual diagram of an example of a binary image of excessive acupressure.

Therefore, in the method of judging whether or not the input image is proper based on the number of pixels of the whole image, for example, even if the right part of the finger is insufficient in the finger pressure and the number of pixels is small, the left part of the finger is excessive in the finger pressure and the pixels are excessive. If the number of images is large, even if the image partially includes an improper image, it may be captured as an appropriate image if the number of pixels of the entire image falls within a predetermined range.
Naturally, the matching rate based on these images was extremely poor.

[0005]

According to the present invention, since it is determined based on the entire fingerprint image whether or not the image for collation is taken in, if the finger pressure unevenness occurs at a portion of the finger, the collation accuracy is deteriorated. The present invention has been devised to solve the above problem, and an object thereof is to provide a fingerprint input method for more surely determining whether or not an input fingerprint image is appropriate for collation or the like.

[0006]

A fingerprint input method according to the present invention first binarizes an input image, divides this binary image into a plurality of blocks, and obtains the area of each block. The background and boundary blocks that are not suitable for determining the quality of the inner image of this block are removed. Among the blocks remaining after the removal, the number of blocks having an area of a predetermined value or more and a predetermined value or less when the finger pressure is proper is determined, and an image is captured based on the ratio of the number of blocks to the number of blocks remaining after the removal. It is to determine whether or not.

[0007]

According to the fingerprint input method of the present invention, the fingerprint input image is divided into a plurality of blocks, and it is determined whether the area of each block is within a predetermined value or not. It may be determined whether there is acupressure, or even shiatsu.

[0008]

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. 6 is a hardware configuration example for executing the fingerprint input method of the present invention. The fingerprint image captured from the optical system input device 1 is written in the image memory 5 via the optical system interface 2 in the image processing device 3. The CPU 4 manages this series of processing. Further, in the image processing apparatus 3, the CPU 4 has the image memory 5
It has a plurality of image memories 5 for performing binarization processing and the like on the image taken in and outputting the result.

Next, one embodiment of the present invention will be described with reference to the flowchart of FIG. First, a fingerprint image is input by the optical system input device 1 such as a camera (step 1-1), and this input image is binarized. (Step 1-2). Next, the entire binary image in step 1-2 is divided into a plurality of blocks as shown in FIG. 2 (step 1-3), and the area of each divided block is obtained (step 1-4). The area is obtained by counting the number of pixels in each block.

Next, a block which does not include fingerprint pixels, that is, a block which is composed only of the background is erased. These blocks do not include fingerprint pixels, and thus can be excluded from the image quality judgment targets. Further, blocks adjacent to these, in other words, blocks containing both background pixels and fingerprint pixels are excluded. This is because these blocks include both background pixels and fingerprint pixels at an indefinite ratio, so that the number of fingerprint pixels is not constant even if input is made with proper finger pressure, which is not preferable for determining image quality. . This process is performed in step 1-5. The number of blocks remaining after erasing these blocks is RB. That is, RB is the number of blocks of the image remaining after erasing the blocks consisting only of the background having the number of pixels of 0 and the blocks adjacent thereto as shown in FIG.

Next, from the remaining blocks, the number of pixels is M in order to detect a block including a defective image of excessive acupressure.
The number UB of blocks having an ax value or more is counted and registered (step 1-6). Next, in order to detect a block having a defective image with insufficient finger pressure, the number LB of blocks whose pixel number is the Min value or less is counted and registered (step 1-
7). The Max value and the Min value here are values registered in advance by adding the upper limit value and the lower limit value, respectively, as an allowable range to the number of pixels in one block in the case of proper acupressure that has been empirically known. .

That is, the Max value and the Min value setting are
It is calculated from the ideal number of pixels when the finger pressure is captured in a proper and uniform manner in advance. In order to keep the finger pressure within the range determined as a good image in the fingerprint matching algorithm that employs the fingerprint input method using the present invention, the amount of excessive finger pressure is In this case, the number of pixels in the maximum allowable range is the Max value, and the number of pixels in the minimum allowable range when the finger pressure is insufficient is the Min value. UB indicates the number of blocks equal to or greater than the Max value, FIG. 5 illustrates an example of blocks equal to or greater than the Max value, and LB indicates the number of blocks equal to or less than the Min value, and FIG. 4 illustrates a conceptual diagram of an example of blocks equal to or less than the Min value. .

Next, step 1-6 and step 1-7
The ratio of the total number of blocks UB and LB detected in step -5 to the number of remaining blocks RB in step 1-5 is calculated, and it is determined whether this ratio is less than or equal to a predetermined value (step 1- 8). If the ratio is lower than the predetermined value range, that is, if this ratio is low, it is determined that even if some defective images are included, it does not hinder the matching, and the input image is regarded as an image suitable for matching. Assumed, the process goes to step 1-9 to fetch the binary image. On the other hand, if it is equal to or more than the predetermined value, that is, if this ratio is high, a large number of defective images such as excessive finger pressure and insufficient finger pressure are included, and it is determined to be inappropriate for collation, and the process proceeds to step 1-10.

In step 1-10, the re-input instruction with the reason for uneven pressing and the place is displayed, and the input person can take an appropriate response. The input person again inputs the fingerprint and repeats the processing from step 1-1.

[0015]

As described above, according to the fingerprint input method of the present invention, the entire fingerprint image is divided into a plurality of blocks,
It can be determined whether or not each block is a proper fingerprint image, and whether or not the fingerprint image is appropriate for collation can be determined based on the ratio of the proper number of image blocks to the total number of blocks.

Therefore, in contrast to the case where the image quality unfavorable is simply determined based on the area of the entire fingerprint image, an image unsuitable for collation is also taken in to deteriorate the collation rate. Since only the image is captured, the matching accuracy increases. In other words, in the conventional method, the collation rate is reduced when there is unevenness in the finger pressure at a portion of the finger, but in the present invention,
It is possible to deal with uneven shiatsu and the collation rate is improved. Further, when the reason is not appropriate for the collation, the reason and place of the reason are displayed to the input person, so that the input person can be prompted to input more surely and properly.

[Brief description of drawings]

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

FIG. 2 is a conceptual diagram in which a binary image is divided into a plurality of blocks.

FIG. 3 is a conceptual diagram in which a block having 0 pixels and an adjacent block are erased.

FIG. 4 is a conceptual diagram showing an example of an image of insufficient acupressure and a block example below a Min value.

FIG. 5 is a conceptual diagram showing an image of excessive acupressure and a block example of a Max value or more.

FIG. 6 is a block diagram of hardware for implementing the present invention.

[Explanation of symbols]

 1 Optical System Input Device 2 Optical System Input Device Interface 3 Image Processing Device 4 CPU 5 Image Memory

Claims (1)

[Claims] 1. A fingerprint image is input and binarized, the binary image is divided into a plurality of blocks, and the area of each block is calculated. A boundary block including both pixels and fingerprint pixels is removed, and the number of blocks having an area of a predetermined value or more and a predetermined value or less is calculated from the remaining blocks, and the remaining blocks of the number of blocks A fingerprint input method characterized by determining whether or not to input an input image based on a ratio to the number.