JPH05233796A - Image processing method and image processor - Google Patents

Abstract

PURPOSE:To perform the collation of two binary pictures at high speed and to reduce the memory amount of registration information. CONSTITUTION:The set of more than one black picture element is treated as a line, the respective black picture element address of a first changed picture obtained by performing a processing thinning the line width of a first picture to the line width designated value which is less than an original line width is stored as registration information in a memory 6 and a second picture is stored in a picture memory 4. Next, in the collation process of the first picture and the second picture, the registration information of the first picture and the respective black picture element address of a second changed picture obtained by performing a processing thinning the second picture to the line width designated value are collated and the coincidence of the first picture and the second picture is decided by a central processing unit 5. In this process, the multistage controls of image information moving amount and the line width and the jumping reference of registration information, etc., are performed and the processing amount is reduced. The memory amount of registration information is reduced by preparing registration data by compressing the binary picture data for which a thinning processing is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to pattern recognition of digitized images (fingerprints, imprints, figures, characters, etc.) for image processing devices (electronic computers, electronic exchanges, communication control devices,
In the case of using an IC card, an image recognition device, an image collation device, hardware and / or software in an image inspection device, etc., image information for registration is stored in a storage device, and further, by collating two images The present invention relates to an image processing method and an image processing apparatus for determining whether a match or a mismatch occurs based on the matching (degree of matching) of images.

[0002]

2. Description of the Related Art As an example of an image targeted for pattern recognition, a case where the image is a fingerprint will be described. A fingerprint is a ridge pattern on a finger. Since the valley line (between the ridge lines) is determined by the ridge line, the pattern drawn by the valley line may be used as the fingerprint instead of the ridge line. The line treated as a fingerprint is called a fingerprint line. A fingerprint input device for personal identification is an imaging device (for example, C
As a method of inputting from a CD (charge coupled device) camera,
Prism method (eg, Shimizu et al., “Fingerprint Information Detection Method Using Prism—Comparison of Total Reflection Method and Optical Path Separation Method”),
IEICE Transactions, Vol.J68-D, No.3, pp.414-415
(1985)), and hologram method (for example, "Personal collation device using holographic fingerprint sensor" by Igaki et al., Technical Report of IEICE, PRU87-31,
pp.27-33, (1987)) and the like are known.

A fingerprint image of analog information input from the image pickup device is converted into a digital grayscale image of a fingerprint by an A / D (analog / digital) converter.
The grayscale image of the fingerprint is represented by the coordinates (X, Y) which are the pixel addresses of the image memory and the brightness of the pixels which are the constituent elements of each pixel address of the image memory. The method of setting the X axis and the Y axis is arbitrary. The unevenness of the fingerprint may be directly converted into a binary image to be used as a fingerprint image. The grayscale image of the fingerprint is
It is also possible to perform correction by smoothing or the direction of the ridge.

[0004] The feature points representing the features of the fingerprint are end points,
There are forks and intersections. The digitized feature image of the grayscale image of the fingerprint is binarized from the fingerprint image and further thinned to obtain a range of pixels representing the feature point (for example, a set of 3 × 3 pixels centered on the feature point). The same pattern as the pattern can be detected by being present in the thinned image (for example,
Sasakawa et al., "Fingerprint collation device for personal identification with enhanced ability to deal with low quality images", IEICE Transactions, Vol.J72-
D-II, No.5, pp.707-714 (1990)).

In the fingerprint collation, a fingerprint for which information for collation is registered in a memory in advance is referred to as a registered fingerprint, and a fingerprint for matching with the registered fingerprint is referred to as an inspection fingerprint. Known methods of collating the registered fingerprint and the inspection fingerprint include a method using a feature point, a method using the direction of a ridge, and a method using pattern matching between original images of the registered fingerprint and the inspection fingerprint. As an example of a matching method by pattern matching between thinned images, Japanese Patent Laid-Open No. 63-132386 discloses a matching method by superimposing a thinned image of a registered fingerprint and a thinned image of an inspection fingerprint. Smoothing is a process for reducing the influence of noise on a fingerprint image. For example, there is a locally weighted average filter that uses the values of neighboring pixels of an arbitrary pixel, Takagi and Shimoda (supervised) “Image Analysis Handbook”, pp. .538-548, Todai Press (1991), etc.

In thinning a binary image, the line width of most of the pixels of the type of line object (most part means half or more to all, and ideally all) is one pixel. The target type pixel is either a black pixel or a white pixel, and will be described as a black pixel hereinafter. As a method of binarizing a grayscale image and thinning the binary image, the black pixels on the outer side in the set of black pixels are sequentially retained while maintaining the connectivity (4 or 8 connections) of the black pixels. Hilditch's thinning method, etc. (see, for example, Tamura (supervised) "Introduction to Computer Image Processing", Soken Shuppan, pp.80-83 (1985), Tamura's "Multi-faceted image processing") And its software system ”, Research Report of Electrotechnical Laboratory, pp.25-64,
835 (February 1984) and Mori et al., "Basics of Image Recognition [I]", pp.65-71, Ohmsha (1986)). In addition, Kobayashi "Image thinning and feature point extraction", IEICE Technical Report, PRU90-
149, pp.33-38 (1991) also describes a thinning method for grayscale images or binary images.

A method of binarizing a grayscale image into a binary image is described, for example, in "Basics of Image Recognition [I]" by Mori et al., Pp.3.
7-47, Ohmsha (1986), etc. When the fingerprint is input, the inspection fingerprint and the registered fingerprint are misaligned (rotated and / or moved in parallel). Therefore, the verification of the inspection fingerprint and the registered fingerprint requires alignment of both fingerprints. As the alignment method (rotation and parallel movement), the method that uses the ridge direction, the method that uses the representative feature points and the peripheral feature points, and the trial and error for the range in which only parallel movement is possible Positioning method and the like are known. The formulas of coordinate transformation and geometric transformation required for alignment are described in, for example, Plastock et al., "Computer Graphics," translated by Koriyama, pp.84-88, McGraw-Hill Book (1987). There is.

It is useful to find an approximate center point of the fingerprint image in the alignment at the time of matching. Japanese Patent Sho 58
-55548 "Diagram center position determination method" discloses a method in which a direction in which a ridge has a steep slope is sequentially searched for a center point. In Ito et al. “Classification Method of Fingerprint Images Focusing on Central Point”, IEICE Technical Report, PRU89-79, pp.15-22 (1989), the number of intersections with parallel lines on each side of a rectangle is described. A method of sequentially approaching the center position by using it is described. "A Study on Extraction of Reference Points in Fingerprint Matching", 1987 IEICE National Conference on Information and Systems, No.125,
In, the number of ridges passing through each scanning line is counted to obtain the distribution of the number of lines.

Kobayashi, "Consideration of Fingerprint Image Matching Method," IEICE Technical Report, PRU91-45, IEICE, pp.25-30 (1991)
(July 2016), we proposed a method by template matching of black pixels acquired from the thinned image of the registered fingerprint (or the image subjected to the thinning process) and the binary image of the inspection fingerprint (or the image subjected to the thinning process). However, in that method, the processing amount and the memory amount are reduced as compared with the method of performing template matching between binary images.

[0010]

As described above, various techniques have been proposed for collating images such as fingerprints, but the processing device still requires a relatively large storage capacity and processing amount. As a method of storing a line figure with a small storage capacity, a method using Freeman's chain code (see, for example, "Image Information Processing" by Yasui In and Nakajima, pp.113-1).
14, Morikita Publishing (1991)) is known, but it is difficult to apply it to complicated cases such as fingerprints. Therefore, as a processing device for fingerprint collation, it is difficult to put it into practice in a general personal computer or the like, and it is necessary to use a high-speed computer or a dedicated hardware device, which is expensive. The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an image processing method and an image processing apparatus that can be implemented with a high processing speed and a small storage capacity.

[0011]

In order to solve the above-mentioned problems, in the structure according to claim 1, the black pixel and the white pixel are the constituent elements of the pixel of the image, and each pixel address has an X coordinate and a Y coordinate. For a binary image represented by coordinates, one of the pixel types of black pixels and white pixels is set as the target pixel type, the position is arbitrary, and a pixel set having a prescribed number of pixels and a shape is defined, The set is composed of representative pixels and non-representative pixels. The binary image is regarded as a set of pixel sets, and the Y coordinate and the X coordinate of the representative pixel of each pixel set on the Y coordinate are targeted to each pixel set. It is checked whether or not the pixel type exists, and only when the target pixel type exists, the X coordinate of the representative pixel, the classification bit indicating whether the representative pixel is a black pixel or a white pixel,
Further, the binary image is stored by the data of the division bit of the black pixel and the white pixel in each non-representative pixel.

Further, in the structure according to claim 2,
Two images of a binary image having black pixels and white pixels as constituent elements of the pixels of the image are referred to as a first image and a second image, and a matching processing unit that checks the matching after alignment of both images is provided, and one image is provided. Is moved by means of a pattern that has been moved by an arbitrary conversion, and the two images are collated,
For the first image, the pattern is sequentially moved in the first movement range with the first movement step size by an arbitrary conversion, and the black pixels of the pattern are checked by the interlaced search in which the first registration information search increase width is 2 or more. A first pattern moving means for determining a first moving position where the evaluation criterion is suboptimal is provided, and one or more of the first pattern moving means obtained by changing each parameter within a specified range including non-change It is characterized in that the movement position where the evaluation criterion is optimum is obtained by sequentially executing each pattern moving means.

Further, in the structure according to claim 3,
Black pixels and white pixels are constituent elements of an image pixel, one of the pixel types of black pixels and white pixels is a target pixel type, and the other is a non-target pixel type, and each pixel address is represented by an X coordinate and a Y coordinate. A binary image, two images of which are a first image and a second image, and a matching processing unit that checks the matching after alignment of both images is provided, and one of the images is a pattern after movement by arbitrary conversion By providing a collation processing unit for collating the two images, converting a part of the target pixel type of the first image into a non-target pixel type, and then creating registration data for the target pixel type. Is characterized by.

[0014]

According to the structure described in claim 1, it is checked whether or not the target pixel type exists in each pixel set, and only when the target pixel type exists, the X coordinate of the representative pixel and the representative pixel are black. A binary image is stored by the data of the division bit of the pixel or the white pixel and the division bit of the black pixel and the white pixel in each of the non-representative pixels. As a result, the information is stored only when there is a pixel of the target pixel type (for example, a black pixel) in the pixel set, and the division between the black pixel and the white pixel can be displayed in bits, so that the data compression can be performed in a relatively short processing time. Decompression is performed.

According to the second aspect of the present invention,
The two images of the binary image having the black pixels and the white pixels as the constituent elements of the pixels of the image are the first image and the second image, and the matching processing unit checks the matching after the alignment of the two images.
Further, when one of the images is moved by means of a pattern after movement by an arbitrary conversion and the two images are collated, the pattern of the first image is converted by the arbitrary movement into a first moving step width The first pattern moving is the first pattern moving position that is sequentially moved in one moving range, and the black pixel of the pattern is checked by the interlace search in which the increase width of the first registered information search is 2 or more Required by means. Then, for the first pattern moving means, for each of one or more pattern moving means obtained by changing each parameter in the specified range including non-change,
By performing the operations sequentially, the movement position where the evaluation criterion is optimum is obtained.

According to the third aspect of the present invention,
Since the registration data is created after the part of the first image is converted to the non-target pixel type, the required storage capacity for storing this registration data can be made small, and
The processing amount in the collation processing means becomes small.

[0017]

Example I. Configuration of Embodiment I. 1. Overall Configuration of Embodiments An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a fingerprint identification system according to an embodiment. In the figure, reference numeral 2 is an image input device, and a fingerprint (fingerprint image) is input through an image pickup device 7 provided here, and is supplied to the image processing device 1. The input fingerprint image is converted into digital information via the A / D converter 3 and stored in the image memory 4. Reference numeral 6 denotes a memory, which stores information such as programs, various data, files (a set of data), and the like. The memory 6 may be a mixture of storage devices having different characteristics (for example, a semiconductor memory and a magnetic disk). In this case, movement of information between them is performed by hardware or software as necessary. Be done.

Reference numeral 5 denotes a central processing unit, which is composed of at least one CPU unit and performs various controls on other components based on a program stored in the memory 6. That is, a plurality of fingerprints (hereinafter referred to as “registered fingerprints”) are stored in the memory 6, and the central processing unit 5 includes a fingerprint image (hereinafter referred to as “inspection fingerprint”) supplied from the image input device 2 and each fingerprint. The registered fingerprint is collated to determine whether the two match. It should be noted that the image memory 4 and the memory 6 are merely divisions based on stored information and can be realized by a common storage device.
Further, when the image pickup device 7 directly outputs a digital image, the A / D converter is unnecessary.

I. 2. Configuration of Each Part Hereinafter, details of each component in FIG. 1 will be described. Although the details of the image memory 4 will be described later, in the central processing unit 5, the digitized grayscale image of the fingerprint is converted into a binary image, and the line represented by the binary image is thinned. Processing is performed. The central processing unit 5 inputs / outputs image data to / from the image memory 4 for these processes. On this occasion,
The pixel address in the image memory 4 is represented as (X, Y) by the X coordinate and the Y coordinate. Hereafter, the pixel address (X,
Y) is represented as a pixel (X, Y) or simply (X, Y). A portion of the image memory 4 that stores one image is called a screen. The image memory 4 can store one or more screens.

Each screen of the image memory 4 is composed of a plurality of pixels, and the range of all pixel addresses is 0≤X≤.
And _{X h, 0 ≦ Y ≦ Y} h. By the way, in the present embodiment, not all the pixel data in such a range is subjected to the processing, but the processing range further specified according to the type of processing is the processing target. In addition, when a number below the decimal point occurs in the pixel address and the number of pixels due to the calculation including the pixel address and the number of pixels, the processing is performed by rounding down, rounding down, or rounding up. Pixel values are represented by brightness. Which part of the luminance becomes the ridge depends on the processing method of the image input device 2 and the image processing in the image processing device 1. In any case, the luminance characteristic corresponding to the ridge is determined by the image processing device 1. It is possible to process by setting in advance. A set of one or more pixels is called a pixel set.

In an image binarized into black pixels and white pixels, either a black pixel or a white pixel can be selected as the type of the target pixel as a fingerprint line (one of them is a ridge). It should correspond to each valley line).

In this embodiment, black pixels are treated as fingerprint lines. The thinning is a process in which most of the line width is set to 1 pixel, but in the present embodiment, the line width of the black pixel set is included in the image of the black pixels of the original binary image. Is called a thinning process, and an image obtained as a result of the thinning process is called a thinning image. Therefore, thinning is one form of thinning processing in this embodiment. The line width is defined as the minimum distance (the number of pixels) of reaching the other edge through the line when the point of the edge of the arbitrary line is determined (determined for each position of the edge of the line).

FIG. 2A shows the status of the image data stored in the image memory 4. The screen of image 10
An image (binary image or grayscale image) obtained from an image input from the image input device 2 and digitized is stored. For example, a past image is stored on the screen of the image 11 and can be used for processing the image 10. The image 11 may be unnecessary depending on the selection of the processing means. In that case, the image memory 4 may be the image 10 only. Figure 2 (b) shows
The situation of data and the like stored in the memory 6 is shown, the program and data 12 store the program and data for realizing the present embodiment, and the registration information 13 contains
Store the registration information of the registered fingerprint image in a file.

When the image is a grayscale image, the code value corresponding to the brightness is set. The binary image may be obtained by binarizing the grayscale image set in the image memory 4 or may be directly set in the image memory 4 depending on the image input method. Binary images are represented by black and white pixels only, but code values corresponding to the respective luminance values of the black and white pixels (for example, 1 for black pixels and 0 for white pixels).
Shall be specified. Whether the black pixel has a high brightness part or a low brightness part depends on the target image and the input method thereof, and may correspond to either case. The logical origin and coordinate axes of the image stored in the image memory 4 can be determined independently of the physical pixel position of the image memory 4. The X axis and Y axis can be set freely, but for convenience of explanation, the X direction is from left to right (that is,
The X direction is the horizontal direction, and the Y direction is the vertical direction from top to bottom (that is, the Y increasing direction).

A set of pixels is called a pixel set. FIG. 3A is an example of a pixel set of 3 × 3 pixels (that is, 3 × 3 pixels), FIG. 3B is an example of a pixel set of 4 × 4 pixels, and FIG. It is an example of a pixel set of 4 × 3 pixels. Next, the notation and definitions used in the following description are shown. [m]: Represents truncation after the decimal point of an arbitrary number m. f (X, Y): The luminance at the pixel address (X, Y). FA: Fingerprint area. Since the fingerprint area can be expressed only by the fingerprint boundary, it is stored in the memory 6 as fingerprint boundary information. The fingerprint boundary is treated as within the fingerprint area. (X _C , Y _C ): Address of the approximate center point of the fingerprint. (X _RC , Y _RC ): Address of the approximate center point of the registered fingerprint. (X _TC , Y _TC ): Address of the approximate center point of the inspection fingerprint.

Rth: Registered fingerprint binary image (first image) obtained from registered fingerprint image (grayscale image or binary image)
In addition, the registered fingerprint change image (that is, the first change image) obtained by performing at least the thinning process. Tth: An inspection fingerprint changed image (that is, a second changed image) obtained by performing at least a thinning process on a registered fingerprint binary image obtained from the inspection fingerprint image (grayscale image or binary image). RA: A set of addresses (X, Y) of all black pixels in the fingerprint area in the registered fingerprint change image Rth. R
A is the union of RT (0), RB (0), and the unused fingerprint area black pixels.

RT (0): RT (0) is a subset of RA. RT (0) can specify an area of one or more arbitrary number of separated pixel sets as a small area of RA. RT
(0) is called a sub template. The part of the image from which the sub-template is extracted may be called the sub-template part. The sub-template is used for image registration (see FIGS. 4 and 5). RB (0): This is a set of addresses of black pixels in a portion excluding RT (0) from RA. RB (0) is a subset of RA. Any number of one or more separate regions of the pixel set can be specified. RB (0) is called a non-subtemplate. The part of the image from which the non-subtemplate is extracted may be called the non-subtemplate part. Non-subtemplates are not used for image registration (see FIGS. 4 and 5).

RT (S): A set of black pixel addresses (X, Y) of the sub-template RT (0) when the coordinate axis of the registered fingerprint change image is rotated S degrees about an arbitrarily determined point. is there. RT (S, H, V): S around the arbitrary axis (for example, the approximate center point of the registered fingerprint) on the coordinate axis of the registered fingerprint change image
This is a set of addresses of the sub-template RT (0) on the coordinate axis after the horizontal movement H and the vertical movement V after the rotation. RT (0,0,0) = RT (0), RT (S, 0,0) = RT
(S). (A set of S, H, and V values that specify the movement position determines one pattern in which the registered fingerprint change image is moved by coordinate conversion.)

RB (S, H, V): The coordinate axis of the registered fingerprint change image is rotated S degrees about an arbitrary point (for example, the approximate center point of the registered fingerprint), and then horizontal movement H and vertical movement V are performed. It is a set of addresses of the non-sub-template RB (0) on the coordinate axis after the execution. RB (0,0,0) = RB (0), RB
(S, 0,0) = RB (S). N1m: The number of sub-template matching black pixels, which represents the number of matching black pixels between the sub-template black pixel of the registered fingerprint and the inspection fingerprint black pixel. N1c: Sub template total black pixel number, which represents the total black pixel number of the sub template of the registered fingerprint.

N2m: The number of non-sub-template matching black pixels, which represents the number of matching black pixels between the non-sub-template black pixels of the registered fingerprint and the inspection fingerprint black pixel. N2c: The total number of black pixels in the non-sub-template, which represents the total number of black pixels in the non-sub-template of the registered fingerprint. Nm: represents N1m or N2m. The counter Nm is the value of the counter for the calculation of Nm. (Nm = N1m + N2m
In some cases. ) Nc: represents N1c or N2c. The counter Nc is the counter value for the calculation of Nc. (Nc = N1c + N2c
In some cases. )

FIGS. 4 and 5 show a fingerprint area defined by a fingerprint boundary in the image 10 in the image memory 4,
An example of the relationship between the part for extracting the sub-template RT (0) and the part for extracting the non-sub-template RB (0) is shown. Examples 1 to 4 are examples of the sub-template portion and the non-sub-template portion.

II. Operation of Embodiment II. 1. Overall Operation of Embodiment Next, the operation of this embodiment will be described. First, the operation of the present embodiment includes a registration process for storing the registered fingerprints in the memory 6 and a matching process for determining the matching relationship between each registered fingerprint and the inspection fingerprint supplied from the image input device 2. The outline of both processes will be described with reference to FIG.

II. 1.1. When the process is started in the schematic diagram of the registration process , in step ZA1, a fingerprint image to be registered is input from the image input device 2, and the image data is stored in the image memory 4. Here, the input fingerprint image contains various noises. Therefore, in step ZA2, a smoothing process is performed to reduce noise. Various methods are known as the smoothing process, and for example, it is conceivable to use a local weighted average filter that uses the values of neighboring pixels of an arbitrary pixel. The image pickup device 7 or A /
When the binary image can be directly input to the image memory 4 by the configuration of the D converter 3 or the like, the smoothing process may be omitted. Further, when the binary image is smoothed, the image becomes a grayscale image, so it is necessary to perform the binarization described below again.

The smoothed fingerprint image has several densities, but in step ZA3, the fingerprint image is binary with "black" and "white" for subsequent processing. It is converted into information. The binarized fingerprint image is composed of pixels having addresses (X, Y) in the range of “0 ≦ X ≦ X _h , 0 ≦ Y ≦ Y _h ”. The portion other than the fingerprint portion (background portion) is unnecessary for processing. Therefore, the background separation processing for removing this background portion is performed. Next, when the processing proceeds to step ZA4, the approximate center point (X _C , Y _C ) of the fingerprint area is obtained.

Next, in the case of registration processing, the processing proceeds to step ZR1. In step ZR1, thinning processing is performed. That is, the fingerprint line in the fingerprint image obtained in steps ZA1 to ZA4 has a certain thickness, and this is converted into a thinner line image. In this embodiment, most of the line width is converted into one pixel width,
The registered fingerprint change image (first change image) is obtained. next,
When the process proceeds to step ZR2, the image subjected to the thinning process is recorded in the memory 6. Here, it is conceivable to store the image obtained in step ZR1 as it is,
In this embodiment, the following processing is further performed.

Partial conversion processing of the target pixel of the registered image The black pixel (that is, the target pixel) of the fingerprint image is partially converted into a white pixel within a range that does not affect the accuracy of matching. As a result, the data amount of the registered fingerprint becomes small. Here, the part that remains without being converted to white pixels is a "sub template".
And “non-sub-template” (details will be described later). Registration process of registration information of registered image Sub-templates and non-sub-templates are extracted from the registered fingerprint change image and stored in respective files. Here, if the data is compressed when the data is stored in the file, the required storage capacity of the memory 6 can be further reduced. Therefore, data compression is also performed.

II. 1.2. Outline of Collation Process In the collation process, first, the processes of steps ZA1 to ZA4 are performed on the inspection fingerprint. Next, when the process proceeds to step ZC1, the inspection fingerprint is subjected to the thinning process as in step ZR1. Next, when the process proceeds to step ZC2, this inspection fingerprint is compared with each registered fingerprint to determine whether or not it matches any of the registered fingerprints.

II. 2. Detailed Operation of the Embodiment Of the processes shown in FIG. 11, steps ZA1 and ZA2 are known techniques, and therefore the details of the processes other than these steps will be described below. If there is no description of the process to be performed next in a step in each processing procedure, the process immediately proceeds to the next step. The constants used in each means are
Considering various conditions in the image processing device 1 and the image input device 2, an appropriate value is statically or dynamically set. Of course, the realization of each step can be modified as long as the contents of the processing are the same.

II. 2.1. Binarization and background separation
The process (step ZA3) binarization is a process of converting a grayscale image into a binary image.
Background separation is a process of clarifying the effective range of the fingerprint image of the image 10 in the image memory 4. An example of a procedure for performing binarization and background separation will be described in procedure B. The input information of the procedure B is input image information. The output information of the procedure B is the binary image of the output image and the fingerprint boundary information.

(Procedure B) Assumption First, the range of the pixel address (X, Y) is 0 ≦ X ≦ X _h , 0.
Since ≦ Y ≦ Y _h , the length of the image in the X direction is “X _h +1” when one pixel is one unit. Here, the image is divided into a predetermined number of partial areas, and the length of the partial areas in the X direction (=
The length of the partial area in the Y direction) is represented by a constant K. Here, the constant K is the ratio L of the length of the image and the partial area in the X direction.
Select (L = (X _h +1) / K) to be an integer. The length of the partial area may be variable for each partial area.

Here, the address of each partial area is represented by a partial area address J (M, N). The value M is a number that increases from the origin in the X-axis direction by “0”, “1”, “2”, ..., Similarly, the value N is “0” in the Y-axis direction from the origin,
The number increases as "1", "2", .... Therefore, the address J (M, N) of the partial area to which an arbitrary pixel address (X, Y) belongs is obtained by M = [X / K] and N = [Y / K]. In addition, the maximum possible value K _{max of the} values M and N
Becomes K _max = ((X _h +1) / K) −1.

Next, the partial area address J (M, N) for uniquely identifying the partial area forming the image 10 is made to correspond to the leading pixel address of each partial area. Therefore, the pixel address (X,
Y) is obtained by X = K · M, (M = 0, 1, 2, ..., K _max ) Y = K · N, (N = 0, 1, 2, ..., K _max ). .. Based on the above assumptions, in step ZA3, the processes of the following steps B1 to B6 are performed.

[0043] Step B1 (binary conversion): In this step, M = 0,1,2, ···, K max N = 0,1,2, ···, the K _max, the following process To do. First, each partial area J
For each (M, N), the average brightness value Bav (M, N) of all pixels in each partial area is calculated as Bav (M, N) = (sum of brightness of pixels in partial area) / (partial area) The number of pixels inside). next,

(M, N) in X = K · M (M = 0 to K _max ) and Y = K · N (N = 0 to K _max ) covers the range of the image 10. That is, the range of the image, 0 when _{≦ X ≦ X h 0 ≦ Y} ≦ Y h, directed to a range of _{M = 0~ [X h / K} ] N = 0~ [Y h / K]. At this time,

Total number of partial areas of image 10 = ([X _h / K]
+1) · ([Y _h / K] +1). For each partial area, for the brightness f (X, Y) of the pixel (X, Y) of the partial area for which Bav (X, Y) has been calculated, the threshold value T for binarization is set as follows: T = average brightness of partial area + D (D is an integer (positive, negative, or zero) constant), and (X,
Y), that is, for (X, Y) in the range of K · M ≦ X ≦ K · (M + 1) −1 K · N ≦ Y ≦ K · (N + 1) −1, f (X, Y) ≧ T At this time, (X, Y) is set to a white pixel (however, when inverting black and white, it is set to a black pixel).

When f (X, Y) <T, (X, Y) is set to a black pixel (however, when black and white is inverted, it is set to a white pixel). Here, _BL and _BH are constants for distinguishing the effective partial area and the invalid partial area.

Step B2 (Determination of Effective Partial Area): The partial area luminance average value is again obtained for the binary image obtained as a result of step B1. Next, the partial area address I _MN = ([X / K], [Y / K]) (the partial area address is the start address of each partial area) X = K · M, (M = 0, 1, 2, ..., K _max ) Y = K · N, (N = 0,1,2, ..., K _max ), the average brightness value B _av (X, According to Y), the validity / invalidity is determined for each partial area. That is, when _BL ≤ B _av (X, Y) ≤ B _H , it is determined that the partial area is an effective partial area, and effective display is set in the effective partial area table G (M, N).

[0048] The invalid partial area of the image 10 is set to a white pixel.

Step B3 (inspection of the number of effective partial areas): By G (M, N), YT = the number of effective partial areas is counted, and when YT ≧ Y _C (Y _C is a constant), The process proceeds to step B4. If YT <Y _C , the procedure returns with an error due to an error of the number of effective partial areas.

[0050] Step B4 (leftmost fingerprint boundary of a partial region unit): the partial region fingerprint boundary information {(N _T, M _L,
M _R ), N _T = 0 to K _max }, and for each value of N = N _T , M _L ≦ M ≦ M _R is a fingerprint region in a partial region unit.
Here, the left end M _{L of the} fingerprint boundary is obtained for each partial area. Starting from N = 0, N = N _T , (N _T = 0 to
The following processing is performed for K _max ). In the increasing direction of M from the left end (M = 0), 1 (effective partial area display) of the element G (M, N) of the effective partial area table G is sequentially searched, and 1 is KC or more (KC is a partial Let the first M value M _L of the continuous portion for determining the fingerprint boundary of the area be a left edge of the fingerprint area for N at that time. When M (0 to K _max) and 1 (effective area display) cannot be found consecutively K C or more, the value of N is a non-fingerprint area for all M, and N _{T of the} non-fingerprint area is M _{Let L} = _MR = -1.

Step B5 (right edge of fingerprint boundary for each partial area): Here, the right edge M of the fingerprint boundary is set for each partial area.
_{Find R.} Starting from N = 0, N = N _T , (N _T = 0
.About.K _max ), skip N for which the left end is not set in step B4 (that is, N when M _L = -1), and sequentially enable the decreasing direction of M from the right end (M = K _max ). explore the elements G of the partial region table G (M, N) of 1 (valid partial region display), 1 K _C or more (K _C is a constant) then the value M _R of the first M of successive parts Let N be the right edge of the fingerprint area. From the above, the partial area fingerprint boundary information {(N _T , M _L , M _R ), N _T = 0 to K _max }
Is required.

Step B6 (fingerprint boundary information for each pixel): partial area fingerprint boundary information {(N _T , M _L , M _R ), N _T
From 0 to K _max }, the fingerprint boundary information for each pixel is obtained. The fingerprint boundary information for each pixel is represented by {(Y _T , _XL , X _R ),
Expressed as Y _T = 0~Y _h -1}, which is the value of each _{Y T, X L ≦ X ≦} X R is the meaning of the fingerprint area. Depending on the partial area fingerprint boundary information, K · N at N = 0 to K _max
≦ about Y ≦ K · N + K _max of _{Y = Y T, X L =} K · M L X R = K · M R + K max as fingerprint boundary information of each pixel {(Y _T, X _L,
X _R ), Y _T = 0 to Y _h −1}. (End of procedure B) Note that, hereinafter, when simply referred to as fingerprint boundary information, fingerprint boundary information for each pixel {(Y _T , _XL , X _R ), Y _T = 0 to
Means Y _h -1}. FIG. 12 is an example of a fingerprint effective area table.

II. 2.2. Find approximate center point of fingerprint image
(Step ZA4) The procedure for obtaining the approximate center point (X _C , Y _C ) of the fingerprint area from the binary image of the fingerprint in the image 10 is the following procedure Q.
Shown in. The input information of the procedure Q is input image information and fingerprint boundary information. The output information of the procedure Q is the approximate center point (X _C , Y _C ). FIG. 6 shows the relationship between each constant, the fingerprint area, and the approximate center point.

(Procedure Q) Step Q1 (Process for obtaining Y _C ): V _L ≦ X ≦ V
_For the range of _R , the horizontal line Y = KA _i , (i = 1,2, ...,
n; H _D ≤ Y ≤ H _U ) and the number of intersecting lines KH _i , (i = 1, 2, ..., N) between a black pixel line which is a fingerprint line). Here, the line segment width of the black pixel in each intersecting portion is set to a certain value (for example, one pixel) or more. _{H D, H U, V L} , V R is, V _L ≦ X ≦
The section of V _R and H _D ≦ Y ≦ H _U is within the fingerprint area,
It is also a constant selected so that the range of the approximate center point can be set. KA _i , (i = 1, 2, ..., N) is a constant that determines the scanning position. Next, regarding the number of intersecting lines KH _i , (i = 1, 2, ..., N) with respect to the horizontal line Y = KA _i , KH _i , (i = 1,
2, ···, n) is determined to i = i _m which maximizes. Generally i
_There is at least one _m (T _y pieces). Approximate center point Y
The coordinate Y _C is calculated by Y _C = (Σ _{i = im} KA _i ) / T _y . Here, Σ _{i = im} KA _i represents the sum of KA _i for each i = i _m.

Step Q2 (Process for obtaining X _C ): Vertical line X = KB in the range of H _D ≦ Y ≦ H _U
i , (i = 1,2, ..., n; V _L ≦ X ≦ V _R ) and the number of intersections KV _i of the black pixel line, which is a fingerprint line, (i = 1,2, ... , n).
Here, the line segment width of the black pixel in the intersecting portion is set to a certain value (for example, one pixel) or more. KB _i , (i = 1, 2, ..., N) is a constant. Next, the number of intersecting lines K with respect to the vertical line X = KB _i
For V _i , (i = 1,2, ..., N), KV _i , (i = 1,2, ..., N
Find i = i _s that maximizes n). In general, there are one or more i _s (T _x pieces). X _C , which is the X coordinate of the approximate center point, is calculated by X _C = (Σ _{i = is} KB _i ) / T _x . Here, Σ _{i = is} KB _i represents the sum of KB _i for each i = i _s . Step Q3 : Let (X _C , Y _C ) be the approximate center point of the fingerprint area. (End of procedure Q)

A limit is set on the range of the approximate center point. The means are as follows. The range of the approximate center point can be set to any shape, and is set to a small range that can be dealt with by the matching process. For example, it may be in the range of X _CL ≤X ≤X _CH and Y _CL ≤Y≤Y _CH . Outside this range (X _C , Y _C )
When is obtained, the approximate center is set to the closest value in the approximate center range. That is, when X _C <X _CL , X _C
= X _CL, the X _C> X when the _{X CH C = X CH, Y} C < When the Y _CL Y _C = Y _CL, when Y _C> Y _CH and Y _C = Y _CH.
As an alternative to the procedure Q, the center of the image 10 ([X
_h / 2], it is [Y _h / 2]) or approximate center point a point close (X _C, Y _C) and regarded means.

II. 2.3. Narrowing process (step ZR1) In this process, a set of one or more black pixels is treated as an image line. The method of holding the line width designation value is arbitrary, and it is possible to use it as the input information of the thinning processing means or to hold it in the thinning processing means. One or more line width designation values may be held in one thinning processing unit. In the present embodiment, the thinning means can be selectively used according to the line width designation value. However, the image processing apparatus 1 only needs to include thinning processing means for applying the line width designation value applied to the first image and thinning processing means applied to the second image. An example of the thinning processing means depending on the specified line width value is shown below.

Thinning processing for making most of the line width 1 pixel (line width specified value is 1 pixel) Means for thinning the line width of a binary image (or means for binarizing a grayscale image and narrowing the line width) It is possible to use a known method (for example, a thinning method capable of making most of the line width one pixel). There is also a method of directly binarizing and thinning a grayscale image. Narrowing processing for thinning most of the lines to a value equal to or less than the designated value of the line width (the designated value of the line width is an arbitrary value) This can be realized based on a known method. For example, the following a, b
Or there is a means like c.

A) The processing for one screen, in which line elements are deleted pixel by pixel from the outside of a line (corresponding to a black pixel which is a fingerprint line in the present embodiment) forming an image, most of the line width is 1 In the case of the thinning method that iterates to pixels, the thinning process The number of iterations should be set so that

B) In the case of a thinning method in which a black pixel is left from the center of the element of the line which constitutes the image, the center of the line segment is included for the black pixel on the line segment where the line of the image intersects the line of the image. Pixels less than the line width specified value are left (if the line segment width is greater than or equal to the line width specified value, black pixels for the line width specified value are left centered on the midpoint of the line segment and the line segment width is less than the line width specified value This can be achieved by leaving all black pixels on the line segment. c) When thinning the line width to one pixel, the line width can be increased after the thinning so that the line width becomes the specified value.

In the present embodiment, the line width designation value when performing the thinning processing can be arbitrarily designated for the registered fingerprint image and the inspection fingerprint image. It is necessary to determine it from the performance with respect to the designated value, the performance required for the image processing apparatus 1, and the like. Increasing the difference between the line width designation values in the thinning processing for the registered fingerprint image and the inspection fingerprint image will strengthen the positional deviation between the two images, but if the difference is too large, the matching accuracy will decrease. There is. The smaller the line width of the registered fingerprint change image is, the smaller the number of black pixels is, so that the memory amount required for the registration information can be reduced. Considering these matters, for example, one of the following specifications may be valid.

In the thinning process of the registered fingerprint image, the designated line width value is set to 1 pixel, and in the thinning process of the inspection fingerprint image, the designated line width value is set to an appropriate value of 2 pixels or more (for example, 3 pixels). The line width designation value for the thinning process of the registered fingerprint image is appropriately selected under the condition that it is smaller than the line width designation value for the thinning process of the inspection fingerprint image. In the following examples, the case will be mainly described.

II. 2.4. Partial target pixel of registered image
Conversion Process (Step ZR2) By partially converting the black pixel (that is, the target pixel) of the registered image into the white pixel within the range that does not affect the accuracy of matching, the following effect is produced. By reducing the black pixels in the sub-template part, the amount of matching processing is reduced,
And the file size of registration information decreases. By reducing the black pixels in the non-sub-template portion, the file amount of registration information is reduced. This means can be omitted in the collation of fingerprint images, and can be used especially when it is necessary to reduce the file amount and the processing amount. Procedure D is an example of means for partially converting the black pixels of the registered image into white pixels. The input information of the procedure D is the registered fingerprint thin image, and the output information is the converted registered fingerprint thin image.

(Procedure D) Step D1 (Process in X direction): Y = a · j, (0 ≦ X ≦
X _h ; 0 ≦ Y ≦ Y _h ; a is a constant (2 ≦ a <X _h ); j =
0,1,2, ...,) Convert the black pixels above into white pixels. Step D2 (processing in Y direction): X = b · j, (0 ≦ X ≦
X _h ; 0 ≦ Y ≦ Y _h ; b is a constant (2 ≦ b <Y _h ); j =
0,1,2, ...,) Convert the black pixels above into white pixels.
(End of procedure D)

By the procedure D, the partial conversion rate = the number of black pixels of the registration information after the partial conversion / the number of black pixels of the original binary image ε≈ ((a-1) / a) ((b-1 ) / B). Range that does not affect the matching accuracy (for example,
By converting a black pixel into a white pixel by several percent to several tens of percent), the data amount of registration information is reduced by the amount of the converted black pixel in the template, and the processing amount of matching is reduced in the sub-template portion. The amount of processing for the converted black pixels is reduced.

II. 2.5. Registration of registration information for registered images
Processing (step ZR2) In the fingerprint information registration processing, the registered fingerprint change image Rth in the image 10 which is the result of being input to the image 10 of the image memory 4 as the registered fingerprint and being subjected to the processing up to the thinning processing is changed to the sub-template RT. (0) and non-subtemplate RB (0)
Is extracted and stored in each file.
A procedure R is a procedure for performing fingerprint information registration processing. The input information of the procedure R is the registered fingerprint sub-template and non-sub-template file names, the registered fingerprint change image Rth,
The fingerprint boundary information of the registered fingerprint and the approximate center point (X _RC , Y _RC ) of the registered fingerprint. The output information of the procedure R is the file of the sub template RT (0) and the file of the non-sub template RB (0).

(Procedure R) Step R1 (Creation of Sub Template RT (0)): From the registered fingerprint change image Rth, a black pixel address within the fingerprint area and within the range of sub template RT (0) is extracted. Then, a file of the sub template RT (0) is created. The registration fingerprint approximate center point (X _RC , Y _RC ) is also stored in the storage file of RT (0).

Step R2 (Non-sub-template RB (0)
Creation): A black pixel address outside the sub-template RT (0) and within the fingerprint area FA is extracted from the registered fingerprint change image Rth to create a file of the non-sub-template RB (0).
(End of procedure R) The storage format of the data of each file of the sub-template and the non-sub-template is arbitrary. For example, the data may be compressed and stored in a file, and the data may be expanded at the time of use.

II. 2.6. Image data compression processing (step
In the procedure R, when the address (X, Y) of each black pixel is stored in the file as it is, the required storage amount is as follows. Storage storage amount = black pixel count ((X coordinate unit storage amount + Y coordinate unit storage amount)) The registered fingerprint image needs to be stored in a file for each of the sub-template and the non-sub-template. The storage capacity of the memory 6 can be saved more by compressing and storing it in a file than by directly storing the black pixel address of the binary image data. An example of effective processing when the binary image is a thin image is shown in M1, M2, and M3 below.

Method M1 (Example of Claim 1): A case where the pixel set is 3 × 3 pixels will be described. A 3 × 3 pixel set is shown in FIG. FIG. 7 shows an example of a part of the image data. The representative pixel and P4, to display the classification of black pixels and white pixels of each pixel in a pixel group in _{bits, Q = P 8 ‖P 7 ‖P} 6 ‖P 5 ‖P 3 ‖P 2 ‖P 1 ‖ By P ₀ (‖ represents concatenation. For example, 2‖3 = 23. The representative pixel P ₄ is not included in Q). , (In hexadecimal
00 to FF) can be displayed. P ₄ is a margin bit of the X coordinate address part, and displays the division of the white pixel and the black pixel (for example, when the X coordinate address is 1 byte, the X coordinate address / 3 is represented by 7 bits, and other bits are displayed). 1-bit representative pixel P ₄
Display the white pixel and black pixel section).

Next, the pixel address (X, Y) according to the relative position with the representative pixel P ₄ = (X ₄ , Y ₄ ) is as follows. P ₀ : X = X ₄ +1 and Y = Y ₄ +1 P ₁ : X = X ₄ +2, Y = Y ₄ +1 P ₂ : X = X ₄ +2, Y = Y ₄ P ₃ : X = X ₄ +1 Y = Y ₄ P ₄ : X = X ₄ , Y = Y ₄ P ₅ : X = X ₄ , Y = Y ₄ +1 P ₆ : X = X ₄ , Y = Y ₄ +2 P ₇ : X = X ₄ +1 _{, Y = Y 4 +2 P 8} : X = X 4 + 2, Y = Y 4 +2

Next, the compression processing will be described. At the time of compression, the (X, Y) coordinates of the image data are converted into the following format. In the fingerprint effective area, every 3rd Y coordinate, every 3rd X coordinate is set as a representative pixel, and it is checked whether or not there is a black pixel in the pixel set of 3 × 3 pixels (starting point is , Depending on the pixel set settings), only when there are black pixels,
The representative pixel X coordinate and the pixel set code are stored. (Decompression can be performed in reverse.) The storage format of the storage medium is as follows.

Y = 3j: Number of stored X coordinate sets {(P ₄ black / white division bit (first 1 bit) + (representative pixel a
(X coordinate of _j / 3)), pixel set code), (black and white division bit of P ₄ (first 1 bit) + (X coordinate of representative pixel b _j /
3)), pixel set code), ...}, where j = 0,1,2 ,.

Instead of the X coordinate set number of black pixels, an end symbol of the X coordinate value group for each Y coordinate value may be added. Here, the Y-coordinates in which the number of stored X-coordinate sets is 0 are packed without being set. A 3 × 3 pixel set in which all pixels are white pixels is skipped.
The case of the pixel set is 3 × 3 pixels, but it can be set arbitrarily (for example, 4 × 4 pixels, 5 × 3 pixels, etc.). The position of the representative pixel in each pixel set can be set arbitrarily. The black-and-white pixel section is added with a display for one of them in order to determine whether the representative pixel is a white pixel or a black pixel. The head boundary and length (RT (0) and RB (0) in this embodiment) of the target image data are set to be a multiple of 3.

Next, an example of the data reduction rate in the above method M1 will be shown. The X and Y coordinates of the original image are each represented by 1 byte, and the brightness is 1 byte (since it is a binary image, the brightness can be displayed with 1 bit, but at that time, the access processing amount increases). Compare with the case of storing in. Focusing on one pixel set range, data reduction rate = post-compression data amount / original data amount = [(3 x
Probability that there is at least one black pixel in the 3 pixel set) * {(X coordinate storage amount) + (black and white pixel section storage amount) + (peripheral pixel code storage amount)}] / {(number of pixels in pixel set)・ (Pixel unit memory capacity)}

If the probability that there is at least one black pixel in the pixel set is λ, and the storage amount of the black and white pixel section of the representative pixel is 1 byte (this may be 1 bit, but 1 byte for the area), the data reduction rate = 2λ / 9. (The data reduction rate is, for example, 0.02 when λ = 0.1.
It becomes 0.0111 when 22 and λ = 0.05. )

Method M2 (another embodiment of claim 1): The case where the pixel set is 4 × 4 pixels will be described. Arbitrary representative pixel P
_{For 0} , a 4 × 4 pixel set is shown in FIG. Black and white pixels of each pixel are displayed in bits, and Q = P ₁₅ ‖P ₁₄ ‖P ₁₃ ‖ ・ ・ ・ ‖P ₇ ‖P ₆ ‖P ₅ ‖P ₄ ‖P ₃ ‖P ₂ ‖ By P ₁ ‖P ₀ (‖ represents concatenation), the state of the black pixel and the white pixel of each peripheral pixel is represented by a 2-byte pixel set code Q, (0000 in hexadecimal).
~ FFFF) can be displayed. Next, the representative pixel P ₀ = (X ₀ ,
The pixel address (X, Y) by the relative position with respect to Y ₀ ) is
It looks like this:

P ₁ : X = X ₀ -1, Y = Y ₀ P ₂ : X = X ₀ -2, Y = Y ₀ P ₃ : X = X ₀ -3, Y = Y ₀ P ₄ : X = X ₀ , Y = Y ₀ -1 P ₅ : X = X ₀ -1, Y = Y ₀ -1 P ₆ : X = X _0-2 , Y = Y ₀ -1 P ₇ : X = X ₀ -3 , Y = Y ₀ -1 P ₈ : X = X ₀ , Y = Y _0-2 P 9: X = X ₀ -1, Y = Y _0-2 P ₁₀ : X = X _0-2 , Y = Y _{0 -2 P 11: X = X 0} -3, Y = Y 0 -2 P 12: X = X 0, Y = Y 0 -3 P 13: X = X 0 -1, Y = Y 0 -3 P 14 _{: X = X 0 -2, Y} = Y 0 -3 P 15: X = X 0 -3, Y = Y 0 -3

[0079] compression of the registered fingerprint image data, (X, Y) coordinates of the registered fingerprint image _{data, (X = 0~X h, Y} = 0~
Y _h ) is converted into the following format. In the fingerprint effective area, every 4th Y coordinate is set as every 4th X coordinate as a representative pixel, and it is checked whether or not there is a black pixel in the 4 × 4 pixel set range (starting point is (Depending on the setting of the pixel set), the representative pixel X only when there is a black pixel
Store coordinates and pixel set code. That is, the file storage format is

Y = 3 + 4j: number of X coordinate sets of black pixels {(X coordinate of representative pixel a _j , pixel set code), (X coordinate of representative pixel b _j , pixel set code), ...} .. (j = 0,1,2 ...)

Instead of the number of X coordinate sets of black pixels, an end symbol of the X coordinate value group for each Y coordinate value may be added. Y
The coordinates are only for the case where there is a black pixel for the X coordinate,
Store only once (that is, when the number of stored X coordinate sets is 0, the Y coordinate is not set and packed). 4 × 4 pixels, which are all white pixels, are excluded. In the decompression process for the compressed registered fingerprint image data, decompression to the (X, Y) format (reverse process of compression) is performed when the file is read. The fingerprint boundary information is obtained from the left end and the right end of the X coordinate value for each Y coordinate value in the compressed file storage format of RB (0), for example.

Method M3 (further embodiment of claim 1):
When the pixel set has only one pixel, the non-representative pixel is an empty set, and the storage format in this case is: {Y coordinate value Y = j, black pixel X standard number; black pixel X coordinate j ₁ , black pixel The X coordinate is j ₂ , ...} (j = 0, 1, 2, ...). Here, the Y coordinate value is X
Only when the black pixel of the coordinate exists, the data is stored only once. Note that instead of the number of black pixel X coordinate values, an end symbol of the X coordinate value group for each Y coordinate value may be added.

II. 2.7. Matching process between registered image and inspection image
Processing (step ZC2) The matching process in step ZC2 is input to the image 10 of the image memory 4 as an inspection fingerprint, and the black pixels of each of the black pixel sets of the inspection fingerprint changed image that is the result of the processing up to the thinning process are performed. And a process of checking the matching with each black pixel of the black pixel set stored in the memory 6 as the registration information regarding the registered fingerprint. Coordinate conversion by rotation and parallel movement of the coordinate axes for aligning the registered fingerprint and the inspection fingerprint may be performed on either one of the images, but in the present embodiment, the registered fingerprint is narrower within the fingerprint area. Assuming that the number of black pixels is small because the specified line width value for processing is small, the registered fingerprint is moved to match the inspection fingerprint. The outline of the matching process will be described below.

Sub Template Matching (Primary Matching) The sub template matching is performed at the position where the black pixel of the registered fingerprint change image and the black pixel of the inspection fingerprint change image are the best match for the registered fingerprint sub template RT (0). Is a process for obtaining. That is, first, for the sub-template RT (0) of the registered fingerprint change image, the sub-template RT (0, H, V) when the approximate center point of the registered fingerprint is made to coincide with the approximate center point of the inspection fingerprint, Obtained by translating the coordinate axis of the registered fingerprint. Next, in the vicinity of the center, when the coordinate axes of the sub-template RT (0, H, V) are rotated and moved in parallel in the vertical and horizontal directions, the inspection fingerprint change image and the registered fingerprint when the black pixels most match The conversion angle S and the parallel movement amount (horizontal movement amount H, vertical movement amount V) of the sub-template RT (S, H, V) are calculated (S, H, V are integers).

Collation of Non-Sub Template (Secondary Collation) Collation of the non-sub template is performed by using the registered fingerprints RT (S, H, V) S, H, V of the registered fingerprint obtained by the collation of the sub template. A process of calculating the black pixel address by performing coordinate conversion of the black pixel address of RB (0), checking the match with the black pixel address of the inspection fingerprint change image, and outputting information regarding the match between the registered fingerprint and the inspection fingerprint. Is. That is, first, using the angular rotation amount S, the horizontal movement amount H, and the vertical movement amount V of the coordinate axis of the sub-template of the registered fingerprint obtained by the sub-template matching, the black of the non-sub-template RB (0) of the registered fingerprint is black. RB by converting pixel coordinates
(S, H, V) is obtained. Next, the RB of the registered fingerprint change image
The coincidence between the black pixel of (S, H, V) and the black pixel of the inspection fingerprint change image is checked.

The number of black pixels in the mismatched portion is checked. As a final determination, the matching of all the fingerprint areas of the registered fingerprint and the inspection fingerprint is determined based on the matching of the sub-template, the matching of the non-sub-template, and the check result of the mismatched portion. Based on the outline of the collation process described above, the procedure for performing the collation process is shown in Procedure C. The input information of the procedure C is the file name of the sub-template and non-sub-template of the registered fingerprint, the inspection fingerprint change image, and the approximate center point of the inspection fingerprint. The output information of the procedure C is the collation result.

(Procedure C) Step C1 ( Comparison of Sub Template): Step C1a : The sub template RT (0) is stored in the memory 6 from a file. Next, sub template RT
(0), S = S _{min to} S _max , (increase step width K of S), H
= H _{min to} H _max , (incremental step width K of H), and V = V
_{Using min to} V _max , (increasing step width Kv of V), the image matching check assisting procedure (procedure W) described later is executed by the registration information black pixel search increasing step width Kr = Kra, (Kra ≧ 1). As a result, S, H, and V are respectively S _min ~
S _max , H _{min to} H _max , and V _{min to} V _max are changed in increments of increments Ks, Kh, Kv to obtain suboptimal S, H, V
Then, the values Sa, Ha, Va are calculated.

Step C1b : S, H and V are respectively S: (Sa-Dsb) to (Sa + Dsb), increment step width Ksb H: (Ha-Dhb) to (Ha + Dhb), increment step width Khb V: (Va −Dvb) to (Va + Dvb), using the increment step Kvb, the registration information black pixel search increment step Kr = Krb,
The procedure W is executed by (Krb ≧ 1), and the suboptimal {S,
{Sb, Hb, Vb} which is the value of H, V} is obtained. Here, Dsb, Dhb, and Dvb are constants for determining the moving range.

Here, the above-mentioned increased step size (Ksb, Khb, K
vb) is the increment step size Ks, Kh,
It is set to a value smaller than Kv. That is, in step C1a, a relatively large range is checked with the step size as a coarse value, and {S, H, V} obtained in step C1a is checked.
The relatively small range including the sub-optimal value of is investigated by the fine increments (Ksb, Khb, Kvb) in step C1b. This makes it possible to reduce the amount of processing when the range of movement in alignment is increased, as compared with the case where fine increments are used for all.

Step C1c : S, H and V are respectively S: (Sb-Dsc) to (Sb + Dsc), increment step width Ksb H: (Hb-Dhc) to (Hb + Dhc), increment step width Khb V: (Vb) −Dvc) to (Vb + Dvc), using the increment step Kvb, the registration information black pixel search increment step Kr = Krc,
The procedure W is executed by (Krc ≧ 1), and the suboptimal {S,
H, V} values {Sc, Hc, Vc} are obtained. Here, Dsc, Dhc, and Dvc are constants for determining the moving range.

Step C1d : S, H and V are respectively S = Sc, Dsd = 0, increment step width Ksd = 0 H = Hc, Dh _d = 0, increment step width KH _D = 0 V = Vc, Dvd = 0, increment step Kvd = 0, registration information black pixel search increment step Kr = Krd,
The procedure W is executed by (Krd = 1), and the optimum {S, H,
The value of V} {Sm, Hm, Vm} is obtained. here,
Dsd, Dh _d, Dvd are constants for defining the displacement range. As a result, optimum values of {S, H, V} at which the sub-template matching rate T1 is maximum and the sub-template matching rate T1 = N1m / N1c are obtained. Next, regarding the predetermined constant Tk1, if T1 ≧ Tk1, it is determined that the registered fingerprint and the inspection fingerprint match by sub-template matching, and the process proceeds to step C2. If T1 <Tk1, the registered fingerprint and the inspection fingerprint are It is determined that they do not match, and the procedure C is ended.

Step C2 (collation of non-sub-template): non-sub-template RB (0) and step C1
By using the optimum S, H, and V obtained by the above as input information, the image consistency check assisting procedure (procedure W) is executed. As a result, a non-subtemplate matching rate T2 = N2m / N2c is obtained. Next, regarding the predetermined constant TK2, if T2 ≧ Tk2, it is determined that the registered fingerprint and the inspection fingerprint match, and the process proceeds to step C3. If T2 <Tk2, it is determined that the registered fingerprint and the inspection fingerprint do not match. , The procedure C is completed.

Step C3 (matching of mismatched portion): It is necessary to check the mismatch between the black pixel of the registered fingerprint and the black pixel of the inspection fingerprint, and exclude the case where there are too many black pixels in the mismatched portion of the inspection fingerprint change image. .. Therefore, the following process is performed in order to approximately obtain and determine the ratio of the black pixels in the non-matching portion when the line width of the inspection fingerprint change image is matched with the line width of the registered fingerprint change image. Step C3a: From the range of RT (0) and RB (0),
An approximate range of the matching target area of the registered fingerprint change image after the conversion of {S, H, V} is obtained. The coordinates (X ′, Y ′) in the range after conversion from the coordinates (X, Y) are calculated by the following equations, as in the procedure W. X '= (X-X _RC ) ・ cos (S) + (Y−Y _RC ) · sin (S) + X _TC −HY ′ = − (X−X _RC ) · sin (S) + (Y−Y _RC ) · cos (S) + Y _TC −V where cos (·) and sin (·) represent trigonometric functions.

Step C3b : The number of black pixels Tnw of the inspection fingerprint change image of the collation target area (that is, the union of RT (S, H, V) and RB (S, H, V)) after coordinate conversion is counted. . That is, Tnw = total number of black pixels in the collation target area after coordinate conversion of the inspection fingerprint changed image. At this time, assuming that the line width of the inspection fingerprint change image is w, the total number of black pixels Tnc when the line width (line width λ) of the registered fingerprint change image is approximately Tnc = Tnw / (w / Λ). (Note that the registered fingerprint change image can be partially deleted.
When (Procedure D) is executed, Tnc = (partial conversion rate) · Tnw / (w / λ). )

Nm = number of coincident black pixels (= N1m + N2) in the collation target area after coordinate conversion of the registered fingerprint change image and the inspection fingerprint change image
m) Nc = The total number of black pixels (= N1c + N2c) in the matching target area after coordinate conversion of the registered fingerprint change image has already been obtained.

At this time, as the degree of mismatch of the black pixels, for example, the mismatch portion ratio is set as Tz = | Tnc-Nm | / Nc (= | Tnc- (N1m + N2m) | / (N1c + N2c)), and Tz≤Tkc In this case, as the final collation determination, it is determined that the registered fingerprint and the inspection fingerprint match, and if not, it is determined that they do not match.
Here, Tkc, (0 ≤ Tkc ≤ 1) is a constant indicating the allowable ratio of the mismatched black pixels of the inspection fingerprint image, and the smaller the value, the more severe the condition. (End of procedure C)

FIG. 8 is a schematic flowchart of the matching process based on the procedure C. FIG. 9 is a conceptual diagram of parameter relationships in step C3. FIG. 6 shows the relationship between the black pixel set of the registered fingerprint change image and the matching portion and the non-matching portion when the line width of the inspection fingerprint change image matches the line width of the registered fingerprint change image in the newly registered fingerprint area. Explaining.

The optimum value of the moving increment step size and the characteristics of the number of steps will be described. As a precondition, the step size of the last step is 1, the number of steps is up to 3, and other than the last step, the black pixel interlacing process with the registered information black pixel search increment step width Kr of 2 or more is possible. Also, the search movement range in the first stage affects the matching accuracy and is therefore defined. As a property, the larger the increment of movement, the smaller the number of searches,
The smaller the search range, the smaller the number of searches. The first line width should be thicker (reason: the step size can be increased). The Nth step in the middle should be smaller, but the step size before and after Can be arbitrarily determined in consideration of the step size of
The search range after the second step is determined by the step width of the previous step. The step width after the second step must be smaller than the previous step (1/2 or less) to have the effect of increasing the number of steps, and absorb line distortion. For this purpose, the line width should be thick, and the last line width should be as thin as possible to reduce false recognition.

Step C1 of procedure C (multi-stepping in three steps)
In step 1 and 2, the increment of each movement value can be made larger than 1 (the value depends on the line width of the inspection fingerprint thinned image), and in steps 2 and 3, the previous step With the suboptimal values of S, H, and V determined in step 1 as the base points, matching may be performed within the range determined by the line width of the previous stage, and in the first and second steps, the black pixel of the registration information is changed to the registration information black pixel. Since the number of black pixels of the registered fingerprint can be limited by skipping by using a certain number of jumps by setting the search increment step width Kr to 2 or more, the processing amount of collation (almost proportional to the number of alignment search times) can be reduced. Moving range of the first stage (S _min ~ S _max
Etc.) is defined according to the maximum allowable range for fingerprint input.

This processing has the following properties.
The larger the moving range, the larger the number of searches. The smaller the increment size, the larger the number of searches. The step size increase in the middle stage is determined by considering the step size increase immediately before and after each step. Interlaced search is possible except at the final stage. If the setting of the moving range, the increment step, and the interlace search is inappropriate, erroneous recognition easily occurs.

II. 2.7. Image Consistency Check Aid Procedure
(Continuing from Step ZC2) Next, the image consistency check assisting procedure (procedure W) used everywhere in the procedure C will be described. The outline of the processing in this procedure W is as follows. First, for each pixel address (X _R , Y _R ) of the sub-template RT (0) or non-sub-template RB (0) for the registered fingerprint, the approximate center point (X _RC of the registered fingerprint (X _R , Y _R ). , Y
The _RC), test fingerprint (X _T, approximate center point of Y _T) (X _TC,
YTC ) and move in parallel to match. Next, rotate the coordinate axis of the registered fingerprint to convert the black pixel address (X
_R @, Y _R @) checks whether the pixel is a black pixel in the fingerprint area of the inspection fingerprint change image, and also performs parallel movement. RT (S, H, V)
In the case of S, H, V and T1, N1m, N1c at which the matching rate T1 at each value of S, H, V becomes maximum are obtained. For RB (S, H, V), there is only one S, H, V, and the value at that time is T2, which is the maximum.
give. In addition, T1 and T2, N1m and N2m, N1c and N2c
In this procedure, since both primary verification and secondary verification are common,
Called T, Nm, Nc.

The input information of the procedure W is the set of black pixel addresses of the designated portion (either RT (0) or RB (0)) of the registered fingerprint change image, the angle conversion amount S (minimum value, maximum value) of the coordinate axis. ， Incremental step size）, Horizontal movement H of coordinate axis of registered fingerprint
(Minimum value, maximum value, increment step width), vertical movement amount V (minimum value, maximum value, increment step width) of coordinate axis of registered fingerprint, inspection fingerprint change image, and search increment step width K of registered fingerprint black pixel search _r etc. The registration information black pixel search increment step size Kr designates an increment for searching the registration fingerprint black pixel when the registration fingerprint black pixel and the inspection fingerprint black pixel are collated. For example, when Kr = 1, all Registered fingerprint black pixels are searched, and when Kr = 2, every other registered fingerprint black pixels are searched.

The output information of the procedure W is the optimum coordinate axis rotation angle S of the registered fingerprint, the optimum coordinate axis horizontal movement amount H, the optimum coordinate axis vertical movement amount V, and the designated area (either RT (0) or RB (0)). Of the registered fingerprint change image and the inspection fingerprint change image, the total number of black pixels Nm of the registered fingerprint change image of the designated area, and the matching rate T. In this procedure, up to a certain number of black pixels are checked, and if the matching rate is less than or equal to the specified value for censoring, the processing for the pattern at that time is terminated. Further, when the matching rate is equal to or higher than the specified value for determination by the input information specification, {S, H,
V} is determined to be the optimum value (or sub-optimal value), and the subsequent matching process can be omitted. The processing procedure of procedure W is shown below. FIG. 10 shows a schematic flowchart of the procedure W.

(Procedure W) Step W1 (Selection of Angle S): The angle S is sequentially selected from the minimum value to the maximum value of S in increments of S according to the input information (that is, the minimum value of S is S _min , the maximum value is S _max , and the increment is Ks, S = S _min , S _min +
Ks, ..., _Smax ), and the process proceeds to step W2.

Step W2 (coordinate conversion by angle S):
Input registered fingerprint thin image black pixel (RT (0) or R
B (0)), with respect to the target black pixel address (X _R , Y _R ) searched by the registration information black pixel search increment step width Kr, when S = 0, X _R @ = X _R- X _RC + X _TC Y _R @ = Y _R -Y _RC + Y _TC

When S ≠ 0 For all black pixel addresses (X _R , Y _R ) for the black pixel set (either RT (0) or RB (0)) of the registered fingerprint change image input Then, the approximate center point (X _RC , Y _RC ) of the registered fingerprint is set to the approximate center point (X _TC ,
After translation to match the Y _TC), performs coordinate axis rotation of angle S around the (X _TC, Y _TC). This _{is, X R @ = (X R} -X RC) · cos (S) + (Y R -Y RC) · sin (S) + X TC Y R @ = - (X R -X RC) · sin ( It can be realized by the calculation of S) + (Y _R −Y _RC ) · cos (S) + Y _TC . As a result, H = V = 0
All black pixel addresses (X
_R @, Y _R @), that is, the registered fingerprint approximate center (X
All black pixel addresses (X _R @, Y _R @) of the newly registered fingerprint when the registered fingerprint coordinate axis is rotated S degrees around _RC , Y _RC ) and the horizontal movement amount H = vertical movement amount V = 0 The set of is obtained.

Here, the reason why (X _R @, Y _R @) is calculated by the above equation will be described. First, the address (X _R , Y _R )
For all of, the approximate center point of the enrolled fingerprint (X _RC , Y
The new address after the translation for matching ( _RC ) to the approximate center point (X _TC , Y _TC ) of the test fingerprint is X _R # = X _R − (X _RC −X _TC ) Y _R # = Y _R − ( a Y _RC -Y _TC), a _{(X R #, Y R #} ) is the new address. next,
Rotate the coordinate axis around the angle (X _TC , Y _TC ) by the angle S. This _{is, X R @ = (X R} # -X TC) · cos (S) + (Y R # -Y TC) · sin (S) + X TC = (X R -X RC) · cos (S) _{+ (Y R -Y RC) ·} sin (S) + X TC Y R @ = - (X R # -X TC) · sin (S) + (Y R # -Y TC) · cos (S) + Y TC = It can be seen that it can be obtained by − (X _R −X _RC ) · sin (S) + (Y _R −Y _RC ) · cos (S) + Y _TC . In addition, trigonometric functions sin (・), co
As the value of s (·), a value in the range of fluctuation of the angle S may be held in advance.

Step W3 (calculation of matching rate T): Step W3a : The matching black pixel number counter Nm and the registered fingerprint total black pixel number counter Nc are initialized to 0 respectively. Step W3b : The inspection fingerprint thin image is examined for each address of the set (X _R @, Y _R @), and if it is a black pixel in the fingerprint area, 1 is added to the coincident black pixel number counter Nm, and the registration is performed. 1 is also added to the fingerprint black pixel number counter Nc. If it is a white pixel in the fingerprint area or outside the fingerprint area (not treated as a black pixel or a white pixel), 1 is added to the registered fingerprint black pixel number counter Nc.

Here, in the matching process of the sub-template, it is checked whether or not it is possible to abandon the matching of {S, H, V} at this time. That is, the constant {Nci, T
ci; i = 1,2, ..., k}, when counter Nc = Nci (Nci and k are constants), then S,
Assuming that Nm at this time is Nmi, (i = 1,2, ..., k), which is the degree of matching up to the middle of the pattern determined by the values of H and V, then Nmi / Nci <Tci, (i = 1 , 2, ..., k) (Tci and k are constants), it is unlikely to perform further checks, so S, H, and V at that time are abandoned on the way and the next S , H, and V, the process proceeds to step W4.

The value of the constant Tci is 0≤Tci≤1.
However, this value is determined as follows, for example. If the total number of black pixels in the registered fingerprint change image is Nc, Nci / Nc, (i =
1, 2, ..., K) represent the progress status of the processing, and the possible range is 0 ≦ Nci / Nc ≦ 1. As Nci increases and approaches Nc, Nmi / Nci approaches Nm / Nc, which is the concordance rate for {S, H, V} examined at this time, so Tci increases as Nci increases. Can be set. The larger Tci, the wider the range of abandonment,
Although the effect of reducing the processing amount becomes large, on the other hand, erroneous recognition is also likely to occur, so it is necessary to set an appropriate value. The maximum number of calculations is determined by the set value of k (if abandonment is performed once during collation, then the {S,
It is not necessary to calculate whether the value of H, V} can be abandoned later. The specific value of Tci needs to be determined depending on the characteristics of the target image. Further, the conditional expression that determines the range of abandonment during collation is not limited to the example shown in procedure W as long as it determines the degree of coincidence halfway.

Furthermore, in the matching process of the sub template,
When there is an intermediate decision (for example, it can be configured to be specified by the input information of the procedure W), the following processing is performed. If Nmi / Nci> Td (Ncd and Td are constants) when the counter Nc = Ncd, the step W3c is executed because the matching rate is sufficient without any further checking. {S, H, V} of output information {S, H, V
V}.

The values of the constants Ncd and Td are set as follows, for example. The value of Ncd is Nc or less with respect to the total number of black pixels Nc of the registered fingerprint change image, and is a value close to Nc. If Nc is not known in advance, the estimated value may be used. The Td value is smaller than 1 and close to 1. If the value of Td is made small, the effect of reducing the processing becomes large, but erroneous recognition easily occurs. For example, {S, H,
When the step size increment of V} is performed in two steps, it can be used for reducing the processing time of the first step.

Step W3c : For all addresses in the set of (X _R @, Y _R @), when step W3b is completed, T = Nm / Nc is calculated. Then, Nm, Nc, T are stored for {S, H, V} at this time.

Step W4 (translation by H and V):
Newly registered fingerprint black pixel address set (X = H = V = 0)
@, Y @), H and V set in the H and V storage areas are sequentially selected (when H = V = 0, already calculated in step 3), and from the minimum value to the maximum value of H , and the minimum value of V to a maximum value, is sequentially when changing at each incremental value (i.e., when the minimum value of H is H _mi n, the maximum value H _max, incremental value is Kh, H = H _min , H _min +
The maximum value is changed to H _max by Kh, .... The minimum value of V is V _min, the maximum value V _max, when incremental value is _{Kv, V = V min, V} min + Kv, maximum V _max · · ·, by
Change for each {S, H, V},
Since (X @ -H, Y @ -V) becomes the newly registered fingerprint black pixel address set after translation, the same processing as step W3 is performed for each combination of {S, H, V}.
The matching rate T is stored.

Step W5 (check unprocessed S):
When there is an unprocessed S value, the process proceeds to step W1. When there is no unprocessed S value, the process proceeds to step W6. Step W6 (determination of maximum matching rate): each value of S and H
= H _{min to} H _max , V = V _{min to} V _max , for each {S, H, V}, S, H, V, and T, Nm, at which T = Nm / Nc is maximum. Nc is obtained and used as output information. When only one S, H, and V is input, the matching rate at that time is output information.
(End of procedure W)

III. Modifications It is needless to say that the present invention is not limited to the embodiments described above and can be applied to, for example, the following expansions or modifications. (1) Modified Example Including Correction Process of Binary Image In the above embodiment, step ZA3 (see FIG. 11)
The binary image obtained in 1. was directly used for the subsequent processing. However, the image obtained in this step may include, for example, black pixels isolated in white pixels, white pixels isolated in black pixels, and the like. Therefore, in such a case, it is preferable to perform simple image correction such as removal of the isolated black pixel set (changed to white pixels), filling of the isolated white pixel set (changed to black pixels). It should be noted that how much the image correction is performed may be appropriately determined according to the quality of the image, and may be omitted or enhanced as necessary.

(2) Other Modifications The above-described image input method, smoothing processing, binarization processing, background separation processing, correction processing, processing for obtaining an approximate center point, thinning processing, and matching processing. The calculation formulas for the concordance rate and the non-coincidence rate in (3) do not limit the present invention.
That is, such a technique can be appropriately modified, expanded, or partially omitted by another method (for example, a known method). The method of setting the X coordinate and the Y coordinate is arbitrary. When the rotational displacement is so small that it can be ignored, the alignment may be determined by the positional displacement of only possible parallel movement, and the determination may be made by the matching rate when the matching rate is the best. In step W2 of procedure W, X _R @ and Y _R @
The equation for obtaining can be used as long as it is a transformation that allows rotation and parallel movement, and is not limited to this embodiment.

[0118] For _{example, X R @ = X R ·} cos (S) + Y R · sin (S) Y R @ = - X R · sin (S) + Y R · cos (S) can also be used. Also, the use of coordinate transformation or geometric transformation is free. By adding the value obtained by rotating or translating the sub-template as registration information, the processing amount for collation can be reduced (in this case, the memory amount increases).

In the present embodiment, the case where the image is a fingerprint has been described, but the present invention can be applied when the image can be regarded as being composed of lines. In the embodiment, the collation of one inspection fingerprint and one registered fingerprint is described, but the present embodiment is also performed when searching for the best matching registered fingerprint from two or more registered fingerprints for one inspection fingerprint. It can be applied by repeating the collation of examples. The sub-templates and non-sub-templates can be divided freely, and there is an extension such as not dividing them or providing more divisions.

As described above, claims 1 and 3
According to the method described in (1), the required storage capacity for processing can be reduced, and in the configurations described in (2) and (3), the processing amount required for image collation processing can be reduced. it can.

[Brief description of drawings]

FIG. 1 is a block diagram of a fingerprint identification system according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of an image memory and use of the memory.

FIG. 3 is a diagram showing an example of a pixel set.

FIG. 4 is a diagram showing an example of divisions between sub-templates and non-sub-templates.

FIG. 5 is a diagram showing an example of divisions between sub-templates and non-sub-templates.

FIG. 6 is an operation explanatory diagram of processing for obtaining an approximate center point of a fingerprint image.

FIG. 7 is an operation explanatory diagram of processing of dividing a registered image into a pixel set of 3 × 3 pixels.

FIG. 8 is a schematic flowchart of a matching process.

FIG. 9 is an explanatory diagram of checking a mismatched portion.

FIG. 10 is a schematic flowchart of an image consistency check assisting procedure (procedure W).

FIG. 11 is a block diagram showing the overall operation of one embodiment.

FIG. 12 is a diagram showing an example of a fingerprint effective partial area table.

[Explanation of symbols]

5 Central processing unit (collation processing means, first pattern moving means) 6 Memory (storage means)

Claims (3)

[Claims] 1. A black pixel and a white pixel are constituent elements of an image pixel, and each pixel address is one of a black pixel and a white pixel for a binary image represented by an X coordinate and a Y coordinate. Is a target pixel type, a position is arbitrary, and a pixel set having a specified number of pixels and a shape is defined. The pixel set is composed of representative pixels and non-representative pixels. Regarded as the body,
Regarding the Y coordinate and the X coordinate of the representative pixel of each pixel set on the Y coordinate, it is checked whether or not the target pixel type exists in each pixel set, and only when the target pixel type exists, the X coordinate of the representative pixel, An image processing method characterized in that a binary image is stored according to data of a classification bit indicating whether a representative pixel is a black pixel or a white pixel and data concerning a classification bit of a black pixel and a white pixel in each non-representative pixel. 2. Two images of a binary image having black pixels and white pixels as constituent elements of image pixels are referred to as a first image and a second image,
In the case where a matching processing unit for checking the coincidence of both images is provided and one of the images is moved by means of a pattern that has been moved by arbitrary conversion and the two images are matched, The pattern is sequentially moved in the first movement range by the first movement step size by an arbitrary conversion with respect to the image, and the black pixel of the pattern is checked by the interlaced search in which the increase width of the first registration information search is 2 or more, and the evaluation criterion is set. A first pattern moving means for determining a sub-optimal first movement position is provided, and one or more pattern movements obtained by changing each parameter of the first pattern moving means within a specified range including non-change. An image processing apparatus characterized in that a moving position at which an evaluation criterion is optimum is obtained by sequentially executing the means. 3. A black pixel and a white pixel are constituent elements of an image pixel, one of the pixel types of the black pixel and the white pixel is a target pixel type, and the other is a non-target pixel type, and each pixel address has an X coordinate. The two images of the binary image represented by
The image and the second image are provided with a collation processing means for checking the coincidence after the both images are aligned, and one of the images is moved by means of a pattern that has been moved by arbitrary conversion to collate the two images. An image processing apparatus comprising: collation processing means for performing; converting a part of the target pixel type of the first image into a non-target pixel type, and then creating registration data for the target pixel type.