JP5045763B2 - Biometric authentication device, biometric authentication method, blood vessel information storage device, and blood vessel information storage method - Google Patents

Description

The present invention relates to an image collation method and an image collation apparatus that collate two images such as a blood vessel image, a fingerprint image, a still image, and a moving image based on a linear component of the images.
More specifically, the present invention relates to a biometric authentication device, a biometric authentication method, a blood vessel information storage device, and a blood vessel information storage method.

  2. Description of the Related Art Conventionally, various image matching devices are known as devices that perform matching based on image information such as biometric authentication or blood vessel information processing. For example, a registered image and a comparison image to be compared are compared with each other in a predetermined positional relationship, a correlation value is calculated, and an image matching device or a correlation value is generated based on the correlation value. In this case, an image collation apparatus that generates a correlation value by calculation in units of pixels is known.

  However, in the above-described image matching device, even if the image has a small correlation, if there are many types of patterns in the image, for example, if there are many straight line components and there are many intersections between the registered image and the straight line components in the matching image, Since the intersections greatly contribute to the correlation value and the correlation value becomes large and sufficient collation accuracy cannot be obtained, improvement is desired.

  The present invention has been made in view of such circumstances, and an object thereof is to provide an image collation method and an image collation apparatus capable of collating images with high accuracy.

  More specifically, a biometric authentication device, a biometric authentication method, a blood vessel information storage device, and a blood vessel information storage method to which an image verification technology capable of performing image verification with high accuracy is applied.

According to the present invention, a blood vessel image acquiring means for acquiring a blood vessel image, the above blood vessel image acquired by the blood vessel image acquiring means by executing the Hough transform to generate a plurality of first linear components, those plurality of a linear component converting means for converting the first linear component to the first linear component information some of the plurality of indicating each of the line components, the plurality of first line component obtained as a result of the conversion in the straight line component conversion means A linear component extraction unit that extracts a plurality of second linear components that are greater than or equal to an extraction threshold value from information, wherein one conversion is performed on the plurality of first linear component information obtained as a result of conversion in the linear component conversion unit. a preset threshold value or more regions the degree of overlap of the pattern of the curves in the image a plurality extracted, extracting the plurality of second linear components, and the straight line component extracting means, the linear component extracted hand In based on the relative positional relationship of the extracted plurality of second linear components, by comparing the blood vessel information for collation Hough transform is being performed, the above blood vessel image acquired by the blood vessel image acquiring means A biometric authentication device is provided, comprising: an authentication unit that authenticates the correspondence with the blood vessel information for verification.

According to the present invention, a blood vessel image acquiring means for acquiring a blood vessel image, by performing a Hough transform to generate a plurality of first linear component in the blood vessel image acquired by the blood vessel image acquiring unit, those plural the first and the linear component conversion means for converting the first linear component information some of the plurality of indicating each of the line components for line component, the plurality of first straight line obtained as a result of the conversion in the straight line component conversion means A linear component extraction unit that extracts a plurality of second linear components that are greater than or equal to an extraction threshold value from component information, wherein one converted image is obtained for the first linear component information obtained as a result of conversion in the linear component conversion unit. a preset threshold value or more regions the degree of overlap of the patterns of curves of the inner and multiple extraction, extracting the plurality of second linear components, and the straight line component extraction unit, extracting the linear component The plurality of second linear component extracted by the stage, and a storage means for storing in the storage unit as the information used for authentication processing, vessel information storage device is provided.

According to the present invention, a blood vessel image acquisition step of acquiring a blood vessel image, by performing a Hough transform to generate a plurality of first linear component in the blood vessel images obtained in this blood vessel image acquisition step, of the plurality first and a linear component conversion step of converting the first linear component information some of the plurality of indicating each of the line components for one of linear components, the plurality of first linear component information obtained as a result of the conversion in the straight line component conversion process A linear component extraction step of extracting a plurality of linear components greater than or equal to the extraction threshold, wherein the first linear component information obtained as a result of the conversion in the linear component conversion step is a curve pattern in one converted image. the degree of overlap is a plurality extracting a preset threshold value or more regions of extracting the plurality of second linear components, and the straight line component extracting step, apply in this line component extraction step Based on the relative positional relationship of the extracted plurality of second linear components Te, by matching the matching vessel information Hough transform has been performed, the blood vessel image is obtained in the blood vessel image acquiring step, the There is provided a biometric authentication method comprising: an authentication step of authenticating the correspondence with the blood vessel information for verification.

Further, according to the present invention, a blood vessel image acquisition step of acquiring a blood vessel image, a Hough transform is performed on the blood vessel image acquired in the blood vessel image acquisition step, and a plurality of first linear components are generated. and a linear component conversion step of converting the first linear component to the first linear component information some of the plurality of indicating each of the line components, the plurality of first line component obtained as a result of the conversion in the straight line component conversion process A linear component extraction step for extracting a plurality of second linear components that are greater than or equal to the extraction threshold value from the information, wherein the first linear component information obtained as a result of the conversion in the linear component conversion step is within one converted image. of a preset threshold value or more regions the degree of overlap of the patterns of curves with multiple extraction, extracting the plurality of second linear components, and the straight line component extraction step, the linear component extracted The plurality of second linear component extracted in step, and a storage step of storing in the storage unit as the information used for authentication processing, vessel information storage method is provided.

According to the present invention, it is possible to provide a biometric authentication apparatus and a biometric authentication method that can perform biometric authentication with high accuracy.
Further, according to the present invention, a blood vessel information storage device and a blood vessel information storage method can be provided.

FIG. 1 is a hardware functional block diagram of a first embodiment of an image collating apparatus according to the present invention. FIG. 2 is a software functional block diagram of the image collating apparatus shown in FIG. FIG. 3 is a diagram for explaining the operation of the Hough transform unit shown in FIG. FIG. 4 is a diagram for explaining the operation of the Hough transform unit shown in FIG. FIG. 5 is a diagram for explaining the operation of the similarity generation unit shown in FIG. FIG. 6 is a flowchart for explaining the operation of the image collating apparatus according to the present embodiment shown in FIG. FIG. 7 is a diagram showing a position correction unit of the second embodiment of the image collating apparatus according to the present invention. FIG. 8 is a diagram for explaining the operation of the Fourier-Melin transform unit 211 shown in FIG. FIG. 9 is a diagram for explaining the difference between the autocorrelation method and the phase-only correlation method. FIG. 10 is a diagram for explaining the correlation intensity distribution when there is a parallel displacement between two images in the phase only correlation method. FIG. 11 is a diagram for explaining the correlation intensity distribution when there is a rotational shift between two images in the phase-only method. FIG. 12 is a diagram for explaining the correlation image data output by the phase only correlation unit 23. FIG. 13 is a diagram for explaining the correlation image data shown in FIG. FIG. 14 is a flowchart for explaining the operation of the image collating apparatus 1 shown in FIG. FIG. 15 is a functional block diagram of the third embodiment of the image collating apparatus according to the present invention. FIG. 16 is a diagram for explaining the operation of the position correction unit shown in FIG. FIG. 17 is a flowchart for explaining the operation of the third embodiment of the image collating apparatus according to the present invention. FIG. 18 is a functional block diagram of the fourth embodiment of the image collating apparatus according to the present invention. FIG. 19 is a diagram for explaining the operation of the image collating apparatus 1c shown in FIG. FIG. 19A shows a specific example of the image IM11. FIG. 19B is a diagram illustrating an image obtained by extracting a region equal to or greater than the first threshold value from the image IM11 illustrated in FIG. FIG. 19C is a diagram illustrating an image obtained by extracting a region greater than or equal to the second threshold value that is larger than the first threshold value from the image IM11 illustrated in FIG. FIG. 19D is a diagram showing a specific example of the image IM12. FIG. 19E is a diagram illustrating an image obtained by extracting a region having a first threshold value or more from the image IM12 illustrated in FIG. FIG. 19F is a diagram illustrating an image obtained by extracting a region equal to or larger than the second threshold value that is larger than the first threshold value from the image IM11 illustrated in FIG. FIG. 20 is a flowchart for explaining the operation of the image collating apparatus according to this embodiment. FIG. 21 is a flowchart for explaining the operation of the fifth embodiment of the image collating apparatus according to the present invention. FIG. 22 is a flowchart for explaining the operation of the image collating apparatus according to the sixth embodiment of the image collating apparatus according to the present invention.

First Embodiment FIG. 1 is a hardware functional block diagram of a first embodiment of an image collating apparatus according to the present invention.
As shown in FIG. 1, the image matching device 1 according to the present embodiment includes an image input unit 11, a memory 12, an FFT processing unit 13, a coordinate conversion unit 14, a Hough conversion unit 15, a CPU 16, and an operation processing unit. 17.
For example, the image input unit 11 is connected to the memory 12, and the memory 12, the FFT processing unit 13, the coordinate conversion unit 14, the Hough conversion unit 15, and the CPU 16 are connected by a bus BS. The operation processing unit 17 is connected to the CPU 16.

The image input unit 11 is an input unit for inputting an image from the outside. For example, the registered image AIM and an image to be compared with the registered image AIM (also referred to as a collation image RIM) are input to the image input unit 11.
The memory 12 stores, for example, an image input from the image input unit 11. Further, for example, as shown in FIG. 1, the memory 12 stores a registered image AIM, a collation image RIM, a program PRG, and the like.
The program PRG is executed by, for example, the CPU 16 and includes a procedure for realizing functions related to conversion processing, correlation processing, collation processing, and the like according to the present invention.

  For example, under the control of the CPU 16, the FFT processing unit 13 performs a two-dimensional Fourier transform process based on the image stored in the memory 12 and outputs the processing result to the coordinate conversion unit 14, the CPU 16, and the like.

The coordinate conversion unit 14 performs logarithmic-polar coordinate conversion based on the result of the two-dimensional Fourier transform process processed by the FFT processing unit 13 under the control of the CPU 16, for example, and outputs the coordinate conversion result to the CPU 16.
The operation processing unit 17 performs a predetermined process such as releasing the electronic key when the registered image AIM and the collation image RIM match, for example, based on the result of the processing of the CPU 16 described later.

  The Hough conversion unit 15 performs a Hough conversion process, which will be described later, under the control of the CPU 16 and outputs the processing result to the CPU 16. The Hough conversion unit 15 preferably uses a dedicated circuit configured by hardware in order to perform Hough conversion processing at high speed, for example.

  The CPU 16 performs collation processing according to the embodiment of the present invention based on, for example, the program PRG, the registered image AIM, and the collation image RIM stored in the memory 12. Further, the CPU 16 controls the image input unit 11, the memory 12, the FFT processing unit 13, the coordinate conversion unit 14, the Hough conversion unit 15, the operation processing unit 17, and the like to realize the processing according to the present embodiment.

FIG. 2 is a software functional block diagram of the image collating apparatus shown in FIG.
For example, when the CPU 16 executes the program PRG in the memory 12, the functions of the position correction unit 161, the Hough conversion unit 162, the extraction unit 163, and the determination unit 164 are realized as shown in FIG.
The position correction unit 161 corresponds to the position correction unit according to the present invention, the Hough conversion unit 162 corresponds to the conversion unit according to the present invention, the extraction unit 163 corresponds to the extraction unit according to the present invention, and the determination unit 164 This corresponds to the collating means according to the invention.

  For example, based on the image pattern in the registered image AIM and the collation image RIM stored in the memory 12, the position correction unit 161 shifts the position of each of the two images in the horizontal and vertical directions, the magnification, the rotation angle, and the like. , And the corrected image is output to the Hough transform unit 162. Specifically, for example, the position correction unit 161 performs position correction processing for each of the registered image AIM and the collation image RIM, and outputs the result to the Hough conversion unit 162 as signals S1611 and S1612.

For example, the Hough conversion unit 162 causes the Hough conversion unit 15 that performs a dedicated Hough conversion process in hardware to execute the Hough conversion process.
Specifically, for example, the Hough transform unit 162 performs a Hough transform process based on the signal S1611 that is the registered image AIM subjected to the position correction process, and outputs the processing result as a signal S1621. Further, the Hough transform unit 162 performs a Hough transform process based on the signal S1612 that is the collation image RIM subjected to the position correction process, and outputs the processing result as a signal S1622.

FIG. 3 is a diagram for explaining the operation of the Hough transform unit shown in FIG.
The Hough transform unit 162, for example, for each of the first image and the second image, the distance ρ0 from the reference position O to the shortest point P0 to the straight line L0 passing through the point in the image, and the reference position O and the shortest Based on the angle θ between the straight line n0 passing through the point P0 and the reference axis including the reference position O, the point in the image is converted into a curved pattern, and the linear component in the image is converted into a plurality of overlapping curved patterns. Image processing to be converted is performed to generate a first converted image and a second converted image.

In order to simplify the explanation, for example, as shown in FIG. 3A, on the xy plane, the straight line L0, the point P (x1, y1), the point P2 (x2, y2), the point P3 on the straight line L0. Assume that (x3, y3) exists.
Assuming that a straight line passing through the origin 0 and perpendicular to the straight line L0 is n0, for example, the straight line n0 and the x-axis as the reference axis have a relationship of an angle θ0, and there is a relationship of a distance | ρ0 | . Here, | ρ0 | indicates the absolute value of ρ0. The straight line L0 can be expressed by a parameter (ρ0, θ0).
The Hough transform for coordinates (x, y) on the xy plane is defined by, for example, Expression (1).

[Equation 1]
ρ = x · cos θ + y · sin θ
... (1)

For example, when the Hough transform shown in Formula (1) is performed for each of the points P1, P2, and P3, the points are converted into curves in the ρ-θ space as shown in FIG. Specifically, in the Hough transform, the point P1 (x1, y1) is on the curve PL1 (x1 · cos θ + y1 · sin θ), the point P2 (x2, y2) is on the curve PL2 (x2 · cos θ + y2 · sin θ), and the point P3 (x3, x3). y3) is converted into a curve PL3 (x3 · cos θ + y3 · sin θ).
The patterns of the curves PL1, PL2, and PL3 intersect at an intersection CP (ρ0, θ0) in the ρ-θ space. The intersection point P (ρ0, θ0) on the ρ-θ space corresponds to the linear component L0 on the xy space.
Conversely, as shown in FIG. 3A, the linear component L0 on the xy plane corresponds to the intersection CP of the patterns PL1, PL2, and PL3 in the ρ-θ space.

  As described above, the Hough transform process is performed on the binarized image, and depending on the degree of overlapping of the pattern of the curve in the ρ-θ space of the process result, what linear component is present on the xy plane before the conversion. It can be determined whether it is dominant.

FIG. 4 is a diagram for explaining the operation of the Hough transform unit shown in FIG.
For example, the Hough transform unit 162 performs a Hough transform process on the registered image AIM subjected to the position correction process illustrated in FIG. 4A to generate an image S1621 illustrated in FIG. 4C, and the position illustrated in FIG. The corrected collation image RIM is subjected to Hough transform processing to generate an image S1622.
Each pixel in the images S1621, S1622 is set to a value corresponding to the degree of curve pattern overlap. In the present embodiment, the image is displayed in white as the degree of overlap of the curved pattern is higher in the image indicated by the predetermined gradation.
As will be described later, the matching unit 1642 performs the matching process based on the degree of the overlap of the curve patterns, and therefore performs the matching process based on the linear component in the original xy space.

The extraction unit 163 extracts, for each of the first converted image and the second converted image, a region where the degree of overlapping of the curve patterns in one converted image is equal to or greater than a preset threshold. Specifically, for example, based on the signal S1621 as the first converted image shown in FIG. 4C, the extracting unit 163 has a degree of overlapping of the curve patterns in one converted image equal to or greater than a preset threshold value. Are extracted, and an image S1631 shown in FIG. 4E is generated and output to the collation unit 1642.
Further, for example, the extraction unit 163 has an area where the degree of overlapping of the pattern of curves in one converted image is equal to or more than a preset threshold based on the signal S1622 as the second converted image shown in FIG. Is extracted, and an image S1632 shown in FIG. 4F is generated and output to the collation unit 1642.
By performing this extraction process, for example, noise components different from the linear components in the xy space of the registered image AIM and the collation image RIM, for example, point components are removed.

The determination unit 164 determines the first conversion image based on the degree of overlapping of the patterns in the first converted image and the second converted image, and the matching and mismatching of the patterns in the first converted image and the second converted image, respectively. The second image and the second image are collated.
Specifically, the determination unit 164 performs collation based on, for example, the signal S1631 and the signal S1632, and outputs the collation result as the signal S164.

For example, as illustrated in FIG. 2, the determination unit 164 includes a similarity generation unit 1641 and a collation unit 1642.
For example, the similarity generation unit 1641 performs a comparison process for each of a plurality of different positional relationships in the first converted image and the second converted image, and generates a similarity as a correlation value based on the result of the comparison process To do.
Specifically, the similarity generation unit 1641 performs comparison processing for each of a plurality of different positional relationships in two images based on, for example, the signal S1631 and the signal S1632, and based on the result of the comparison processing, the similarity as a correlation value Generate degrees.

  For example, if the two images are assumed to be f1 (m, n) and f2 (m, n), for example, the similarity generation unit 1641 calculates, for example, the similarity Sim according to Equation (2), and outputs the calculation result as S1641.

…（２）

... (2)

FIG. 5 is a diagram for explaining the operation of the similarity generation unit shown in FIG.
For example, when the similarity generation unit 1641 generates the similarity between two images including the line components (also referred to as line shapes) illustrated in FIGS.
As shown in FIG. 5C, the similarity according to the number of intersection points CP of the two images is generated. Here, for the sake of simple explanation, the line component is indicated by a black pixel having a bit value “1”, and the other is indicated by a white pixel having a bit value “0”.

The collation unit 1642 collates the registered image AIM and the collation image RIM based on the signal S1641 indicating the similarity generated by the similarity generation unit 1641.
For example, the collation unit 1642 determines that the registered image AIM matches the collation image RIM when the similarity is greater than a predetermined value.

  FIG. 6 is a flowchart for explaining the operation of the image collating apparatus according to the present embodiment shown in FIG. With reference to FIG. 6, the operation of the image collating apparatus will be described focusing on the operation of the CPU.

For example, a registered image AIM and a collation image RIM are input by the image input unit 11, and each data is stored in the memory 12.
In step ST101, for example, as shown in FIG. 2, the position correction unit 161 performs position correction processing based on the image patterns in the registered image AIM and the collation image RIM stored in the memory 12, and in detail, The positional deviation in the horizontal and vertical directions of each image and the deviation such as the enlargement ratio and the rotation angle are corrected, and the corrected images are output to the Hough transforming unit 162 as signals S1611 and S1612.

In step ST102, the Hough transform unit 162, based on the signal S1611 that is the registered image AIM subjected to the position correction process shown in FIG. 4A, for example, displays the image from the reference position O as shown in FIG. Based on the distance ρ0 to the shortest point P0 to the straight line L0 passing through the point and the angle θ between the reference position O and the straight line n0 passing through the shortest point P0 and the x axis as the reference axis including the reference position O. Then, a point in the image is converted into a curve PL pattern, and a linear component L in the image is converted into a plurality of overlapping curve PL patterns. For example, as shown in FIG. A signal S1621 is generated as a converted image on the -θ space.
Further, the Hough transform unit 162 performs a similar Hough transform process based on the signal S1612 that is the collation image RIM subjected to the position correction process illustrated in FIG. 4B, for example, as illustrated in FIG. 4D. A signal S1622 is generated as a converted image in the ρ-θ space.

In step ST103, the extraction unit 163 extracts, for each of the converted image S1621 and the converted image S1622, an area where the degree of overlapping of the curve patterns in one converted image is equal to or greater than a preset threshold.
Specifically, as described above, each pixel in the images S1621, S1622 is set to a value corresponding to the degree of overlap of the curve patterns. The higher the degree of pattern overlap, the more white it is displayed.
For example, the extraction unit 163 extracts an area where the degree of overlapping of the curve patterns in the converted image S1621 shown in FIG. 4C is equal to or more than a preset threshold, and for example, an image S1631 shown in FIG. It is generated and output to the collation unit 1642.
For example, the extraction unit 163 extracts an area where the degree of overlapping of the curve patterns in the converted image S1622 shown in FIG. 4D is equal to or more than a preset threshold, and for example, an image S1632 shown in FIG. It is generated and output to the collation unit 1642.

  The collation unit 1642 collates the registered image AIM and the collation image RIM based on the overlapping degree of the patterns in the conversion image S1631 and the conversion image S1632, and the matching and mismatching of the patterns in each of the conversion image S1631 and the conversion image S1632. Do.

  Specifically, in step ST104, the similarity generation unit 1641 performs a comparison process for each of a plurality of different positional relationships in the converted image S1631 and the converted image S1632, and uses, for example, a mathematical expression (2) as a correlation value based on the result of the comparison process. ), The similarity Sim is generated and output as a signal S1641.

In step ST105, the matching unit 1642 performs a matching process on the registered image AIM and the matching image RIM based on the similarity Sim as the correlation value generated by the similarity generation unit 1641. Specifically, the collation unit 1642 determines that the registered image AIM and the collation image RIM match when the similarity Sim is higher than a preset threshold value, and does not match when the similarity image Sim is equal to or less than the threshold value. Determine.
For example, when the image matching device according to the present embodiment is applied to a vein pattern matching device in the security field, the operation processing unit 17 performs predetermined processing such as releasing the electronic lock.

  As described above, for each of the registered image AIM and the matching image RIM, the Hough transform process, specifically, the distance ρ from the reference position O to the shortest point P0 to the straight line L passing through the point in the image, and the reference position O And the straight line n0 passing through the shortest point P0 and the angle θ between the x axis as the reference axis including the reference position O, the points in the image are converted into the pattern of the curve PL, and a plurality of linear components in the image are converted. Hough transform unit 15 that performs image processing to convert the pattern PL into the overlapping curve PL and generates converted image S1621 and converted image S1622, and among the patterns in converted image S1621 and converted image S1622 generated by Hough transform unit 15 Registered image AIM based on the degree of overlap and the matching and mismatching of the patterns in each of converted image S1621 and converted image S1622. Since the discrimination unit 164 that collates the collation image RIM is provided, it is possible to collate an image including a linear component as a feature with high accuracy.

  That is, the Hough conversion unit 15 performs a Hough conversion process to generate a converted image in consideration of the linear component (also referred to as a line shape) in the registered image AIM and the collation image RIM, and the determination unit 164 converts the converted image. Since collation is performed based on the degree of overlapping of the patterns of the respective curves, it is possible to perform collation of images including linear components as features with high accuracy.

  Further, if the extraction unit 163 is provided following the Hough conversion unit 162, the degree of overlap of the pattern of the curve PL in one converted image of each of the converted images S1621, S1622 is set in advance by the extraction unit 163. An area equal to or greater than the threshold value is extracted, that is, the point component that is a noise component on the registered image AIM and the matching image RIM is removed, and only the straight line components are extracted to generate images S1631 and S1632, and the determination unit 164 extracts Since the registered image AIM and the collation image RIM are collated based on the matching and mismatching of the patterns in the set area, further improved high-precision collation processing can be performed.

  The discrimination unit 164 is provided with a similarity generation unit 1641 that performs a comparison process for each of a plurality of different positional relationships in the converted image and generates a similarity Sim as a correlation value based on the result of the comparison process. A collation unit 1642 that collates the registered image AIM and the collation image RIM based on the degree Sim, generates a similarity as a correlation value by simple calculation, and performs a collation process based on the similarity, Collation processing can be performed at high speed.

  In the present embodiment, the position correction process is performed in step ST101, but the present invention is not limited to this form. For example, when it is not necessary to perform position correction processing, the Hough conversion unit 15 may perform Hough conversion processing for each of the registered image AIM and the collation image RIM. By doing so, the processing load is reduced, and the collation process can be performed at high speed.

In the present embodiment, in step ST103, the extraction unit 163 extracts an area that is equal to or greater than a threshold set in advance according to the degree of overlap of the patterns of the curve PL. However, the present invention is not limited to this form.
For example, without the extraction unit 163 performing the extraction process, the determination unit 164 determines whether or not the patterns S1621 and S1622 overlap, and the patterns S1621 and S1622 match and do not match. The registered image AIM and the verification image RIM may be verified. By doing so, the processing load is reduced, and the collation process can be performed at high speed.

Second Embodiment FIG. 7 is a diagram showing a position correction unit of a second embodiment of the image collating apparatus according to the present invention.
In the image collating apparatus 1a according to the present embodiment, the hardware functional blocks have substantially the same configuration as the image collating apparatus 1 according to the first embodiment. For example, as illustrated in FIG. The memory 12 includes an FFT processing unit 13, a coordinate conversion unit 14, a Hough conversion unit 15, a CPU 16, and an operation processing unit 17.
The software functional block of the image collating apparatus 1a has substantially the same configuration as that of the image collating apparatus 1 according to the first embodiment. When the CPU 16 executes the program PRG, the position correcting unit 161, the Hough converting unit 162, The functions of the extraction unit 163 and the determination unit 164 are realized.
The same functions are denoted by the same reference numerals, description thereof is omitted, and only differences are described.

The difference is that the position correction unit 161 calculates the correlation value based on the phase component as a result of the rotation angle correction process or enlargement ratio correction process between the registered image AIM and the collation image RIM and the Fourier transform process as the position correction process. It is a point which produces | generates and performs the position correction process of the registration image AIM and the collation image RIM based on the produced | generated correlation value.
Specifically, in the position correction unit 161 according to the present embodiment, the CPU 16 executes the program PRG and controls the FFT processing unit 13, the coordinate conversion unit 14, and the like, for example, as shown in FIG. The functions of the rotation information unit 21, the correction unit 22, the phase only correlation unit 23, the deviation information generation unit 24, and the correction unit 25 are realized.

The magnification information-rotation information unit 21 and the correction unit 22 perform a rotation angle correction process or an enlargement ratio correction process on the registered image AIM and the collation image RIM.
The magnification information-rotation information unit 21 generates magnification information and / or rotation information based on the registered image AIM and the collation image RIM, and outputs the information to the correction unit 22 as a signal S21.
The magnification information includes information indicating the enlargement / reduction ratio of the registered image AIM and the collation image RIM. The rotation information includes information indicating the rotation angle of the registered image AIM and the collation image RIM.

Specifically, for example, the magnification information-rotation information unit 21 includes a Fourier-Melin transform unit 211, a phase only correlation unit 212, and a magnification information-rotation information generation unit 213.
The Fourier-Merlin transform unit 211 performs Fourier-Merlin transform, which will be described later, based on the respective image information, and outputs signals SA211 and SR211 indicating the respective transformation results to the phase-only correlation unit 212.

  Specifically, the Fourier-Melin transform unit 211 includes Fourier transform units 2111, 21112, logarithmic transform units 21121, 2112, and logarithmic-polar coordinate transform units 2111, 211132.

  For example, when the registered image AIM is an N × N image and the registered image AIM is f1 (m, n), the Fourier transform unit 21111 performs Fourier transform as shown in Equation (3), and Fourier image data F1 (U, v) is generated and output to the logarithmic conversion unit 21121. For example, when the collation image RIM is an N × N image and the collation image RIM is f2 (m, n), the Fourier transform unit 21112 performs Fourier transform as shown in Equation (4), and Fourier image data F2 (U, v) is generated and output to the logarithmic conversion unit 21122.

  The Fourier image data F1 (u, v) is composed of an amplitude spectrum A (u, v) and a phase spectrum Θ (u, v) as shown in Equation (3), and the Fourier image data F2 (u, v) is As shown in Equation (4), it is composed of an amplitude spectrum B (u, v) and a phase spectrum Φ (u, v).

…（３）

... (3)

…（４）

... (4)

  The logarithmic conversion units 21121, 2112 perform logarithmic processing based on the amplitude components of the Fourier image data F1 (u, v), F2 (u, v) generated by the Fourier transformation units 2111, 21112. This logarithmic processing of amplitude components emphasizes high frequency components including detailed feature information of image data.

  More specifically, the logarithmic conversion unit 21121 performs logarithmic processing based on the amplitude component A (u, v) as shown in Expression (5) to generate A ′ (u, v) and outputs it to the logarithmic-polar coordinate conversion unit 21131. To do. The logarithmic conversion unit 21122 performs logarithmic processing based on the amplitude component B (u, v) as shown in Equation (6) to generate B ′ (u, v) and outputs it to the logarithmic-polar coordinate conversion unit 21132.

…（５）

... (5)

…（６）

(6)

The logarithm-polar coordinate converters 21131, 21132 convert the logarithm-polar coordinate system (for example, log-r, Θ) based on the signals output from the logarithmic converters 21121, 2112.
In general, for example, when the point (x, y) is defined as shown in the equations (7) and (8), when r = eμ, μ = log (r), and any point (x, There is a unique (log (r), Θ) corresponding to y). Due to this property, the logarithm-polar coordinate conversion units 21131, 21132 perform coordinate conversion.

…（７）

... (7)

…（８）

(8)

  Specifically, the logarithmic-polar coordinate conversion units 21131, 21132 define a set (ri, θj) shown in Equation (9) and a function f (ri, θj) shown in Equation (10).

…（９）

... (9)

…（１０）

(10)

  The logarithm-polar coordinate converters 21131, 21132 use the set (ri, θj) and the function f (ri, θj) defined by the mathematical expressions (9), (10) to generate image data A ′ (u, v), B. '(U, v) is subjected to logarithmic-polar coordinate conversion as shown in equations (11) and (12) to generate pA (ri, θj) and pB (ri, θj), respectively, and signal SA211 It is output to the phase only correlation unit 212 as SR211.

…（１１）

... (11)

…（１２）

(12)

FIG. 8 is a diagram for explaining the operation of the Fourier-Melin transform unit 211 shown in FIG.
The image f1 (m, n) and the image f2 (m, n) include rectangular regions W1 and W2 having different predetermined angles with respect to the x and y axes, for example.
In the Fourier-Melin transform unit 211, for example, as shown in FIG. 8, the image f1 (m, n) is Fourier transformed by the Fourier transform unit 21111 to generate Fourier image data F1 (u, v), and logarithmic transformation is performed. Image data pA (r, θ) is generated by the unit 21121 and the log-polar coordinate conversion unit 21131.

  Similarly, the image f2 (m, n) is Fourier-transformed by the Fourier transform unit 21112 to generate Fourier image data F2 (u, v). The logarithmic transform unit 21122 and the logarithmic-polar coordinate transform unit 21132 produce image data. pB (r, θ) is generated.

As described above, the images f1 (m, n) and f2 (m, n) are transformed from Cartesian coordinates onto a logarithmic-polar coordinate system (also referred to as Fourier-Melin space) by Fourier transformation and logarithmic-polar transformation. .
In the Fourier-Melin space, the component moves along the log-r axis according to the scaling of the image, and moves along the θ axis according to the rotation angle of the image.
Using this property, the scaling (magnification information) and the rotation angle of the images f1 (m, n) and f2 (m, n) are changed to the amount of movement along the log-r axis in the Fourier-Merlin space, and θ It can be determined based on the amount of movement along the axis.

The phase-only correlation unit 212 converts the signal SA211 and the signal SR211 indicating the pattern data output from the Fourier-Melin transform unit 211 by a phase-only correlation method using, for example, a phase-only filter (SPOMF: Symmetric phase-only matched filter). Based on this, the amount of each parallel movement is obtained.
For example, as illustrated in FIG. 7, the phase only correlation unit 212 includes Fourier transform units 2120 and 2121, a synthesis unit 2122, a phase extraction unit 2123, and an inverse Fourier transform unit 2124.

  The Fourier transform units 2120 and 2121 are based on the signals SA211 (pA (m, n)) and SR211 (pB (m, n)) output from the logarithmic-polar coordinate conversion units 21131 and 21132, respectively. ) To perform a Fourier transform. Here, X (u, v) and Y (u, v) are Fourier coefficients. The Fourier coefficient X (u, v) is composed of an amplitude spectrum C (u, v) and a phase spectrum θ (u, v) as shown in Equation (13). The Fourier coefficient Y (u, v) is composed of an amplitude spectrum D (u, v) and a phase spectrum φ (u, v) as shown in Equation (14).

…（１３）

... (13)

…（１４）

... (14)

  The combining unit 2122 combines X (u, v) and Y (u, v) generated by the Fourier transform units 2120 and 2121 to obtain a correlation. For example, the synthesizing unit 2122 generates X (u, v) · Y * (u, v) and outputs it to the phase extracting unit 2123. Here, Y * (u, v) is a complex conjugate of Y (u, v).

The phase extraction unit 2123 extracts phase information by removing the amplitude component based on the combined signal output from the combining unit 2122.
For example, the phase extraction unit 2123, for example, based on X (u, v) Y * (u, v), the phase component Z (u, v) = ej (θ (u, v) −φ (u, v) ) Is extracted.

  The extraction of the phase information is not limited to the above-described form. For example, after extracting the phase information based on the outputs of the Fourier transform units 2120 and 2121 and the equations (15) and (16), only the phase component is synthesized as shown in the equation (17), and Z (u, v) May be generated.

…（１５）

... (15)

…（１６）

... (16)

…（１７）

... (17)

The inverse Fourier transform unit 2124 performs an inverse Fourier transform process based on the signal Z (u, v) having only phase information output from the phase extraction unit 2123 to generate a correlation strength image.
Specifically, the inverse Fourier transform unit 2124 performs an inverse Fourier transform process based on the signal Z (u, v), as shown in Equation (18), and generates a correlation strength image G (p, q).

…（１８）

... (18)

  The magnification information-rotation information generation unit 213 determines the shift amount from the image center of the peak position in the correlation strength image G (p, q) generated by the inverse Fourier transform unit 2124, that is, the registered image AIM and the collation image RIM. Since this is equivalent to the amount of translation between the pattern data obtained as a result of the Fourier-Melin transformation, the magnification information (enlargement / reduction ratio) of the matching image RIM with respect to the registered image AIM is detected by detecting this deviation amount. ) And correction information S21 including data indicating the rotation angle information is generated.

The correcting unit 22 corrects the collation image RIM based on the correction information S21 output from the magnification information-rotation information-rotation information unit 21 magnification information-rotation information generation unit 213. Specifically, the correction unit 22 performs an enlargement / reduction process on the collation image RIM based on the magnification information and the rotation angle information included in the correction information S21, performs the rotation process, and outputs the result to the phase-only correlation unit 23. By the correction process of the correction unit 22, the difference in scaling and rotation components between the registered image AIM and the collation image RIM is removed.
For this reason, only the translation component remains as a difference between the registered image AIM and the corrected collation image RIM.

The phase only correlation unit 23 detects a translation component between the above-described registered image AIM and the corrected collation image RIM, and a correlation value thereof. This detection is obtained, for example, by the phase only correlation method using the above-described phase only filter.
Specifically, the phase only correlation unit 23 includes Fourier transform units 2311 and 2312, a synthesis unit 232, a phase extraction unit 233, and an inverse Fourier transform unit 234.

  The Fourier transform units 2311 and 2312, the synthesis unit 232, the phase extraction unit 233, and the inverse Fourier transform unit 234 are respectively the Fourier transform units 2120 and 2121, the synthesis unit 2122, the phase extraction unit 2123, and the phase-only correlation unit 212 described above. Since it has the same function as each of the inverse Fourier transform units 2124, it will be briefly described.

The Fourier transform unit 2311 performs Fourier transform on the registered image AIM and outputs the result to the synthesis unit 232. At this time, the registered image AIM subjected to the Fourier transform process by the Fourier transform unit 21111 may be stored in the memory 12 and output to the combining unit 232. By doing so, the Fourier transform processing is not performed twice, so that the processing is reduced.
The Fourier transform unit 2312 performs Fourier transform on the image S22 that has been corrected by the correction unit 22 and outputs a Fourier image as a processing result to the synthesis unit 232.

The synthesizing unit 232 synthesizes the Fourier images S2311, S2312 output from the Fourier transform units 2311, 2312 and outputs the synthesized image S232 to the phase extraction unit 233.
The phase extraction unit 233 extracts the phase information as described above based on the composite image S232 and outputs the signal S233 to the inverse Fourier transform unit 234.
The inverse Fourier transform unit 234 performs inverse Fourier transform based on the signal S233 to generate a correlation strength image (correlation image data), and outputs the correlation strength image (correlation image data) to the shift information generation unit 24.

A detailed description of the parallel displacement detection process in the above-described phase-only correlation method will be given.
For example, the original image f1 (m, n), the original image f2 (m, n), and the image f3 (m, n) = f2 (m + α, n + β) obtained by translating the image f2 (m, n) are Fourier transformed. Processing is performed to generate Fourier coefficients F1 (u, v), F2 (u, v), and F3 (u, v) as shown in equations (19) to (21).

…（１９）

... (19)

…（２０）

... (20)

…（２１）

... (21)

  Based on the Fourier coefficients F1 (u, v) to F3 (u, v), as shown in the mathematical expressions (22) to (24), the phase images F′1 (u, v) to F′3 having only phase information. (U, v) is generated.

…（２２）

... (22)

…（２３）

... (23)

…（２４）

... (24)

  The correlation Z12 (u, v) of the phase image of the correlation between the phase image F′1 (u, v) and the phase image F′2 (u, v) is calculated as shown in the equation (25), and the phase image F ′ The correlation Z13 (u, v) of the phase image of the correlation between 1 (u, v) and the phase image F′3 (u, v) is calculated as shown in equation (26).

…（２５）

... (25)

…（２６）

... (26)

  The correlation strength image G12 (r, s) of the correlation Z12 (u, v) and the correlation strength image G13 (r, s) of the correlation Z13 (u, v) are calculated as shown in equations (27) and (28). To do.

…（２７）

... (27)

…（２８）

... (28)

As shown in equations (27) and (28), when the image f3 (m, n) is shifted by (+ α, + β) compared to the image 2 (m, n), the phase-only correlation method is used. In the strong correlation image, a correlation strength peak is generated at a position shifted by (−α, −β). Based on the position of the correlation strength position, the amount of parallel movement between the two images can be obtained.
In addition, by using this phase-only correlation method on the above-described Fourier-Melin space, the amount of parallel movement on the Fourier-Melin space can be detected. This parallel movement amount corresponds to magnification information and rotation angle information in the real space as described above.

FIG. 9 is a diagram for explaining the difference between the autocorrelation method and the phase only correlation method.
In the autocorrelation method, for example, as shown in FIGS. 9A and 9B, when the image IM1 and the image IM2 that is the same as the image IM1 are subjected to Fourier transform to generate the autocorrelation function SG1, the result shown in FIG. 9C is obtained. As described above, a correlation strength distribution having a peak having a high correlation strength and a small correlation strength at the periphery thereof is obtained. In FIG. 9C, the vertical axis (z-axis) indicates the correlation strength, and the x-axis and y-axis indicate the shift amount.

  On the other hand, in the above-described phase-only correlation method, as shown in FIGS. 9D and 9E, when the image IM1 and the image IM2 that is the same as the image IM1 are subjected to Fourier transform and only the phase information is correlated, FIG. ), A correlation strength distribution having a high correlation strength and only a sharp peak is obtained. Thus, the phase-only method can obtain clear information regarding the correlation as compared with the autocorrelation method. In FIG. 9F, the vertical axis (z-axis) indicates the correlation strength, and the x-axis and the y-axis indicate the shift amount.

FIG. 10 is a diagram for explaining a correlation intensity distribution when there is a parallel shift between two images in the phase only correlation method.
For example, the correlation intensity distribution in the phase limiting method of the image IM1 as shown in FIGS. 10A and 10B and the image IM3 moved several pixels parallel from the image IM1 is shown in FIG. 10C, for example. Thus, sharp peaks with high correlation strength are distributed at positions shifted from the peak positions in the correlation image data shown in FIG. 9F by a distance corresponding to the amount of parallel movement. However, the peak intensity shown in FIG. 10C is smaller than the peak intensity shown in FIG. This is because the matching pixel regions of the images IM1 and IM3 are smaller than those of the images IM1 and IM2.

FIG. 11 is a diagram for explaining the correlation strength distribution when there is a rotational shift between two images in the phase limiting method.
The correlation intensity distribution in the phase limiting method of the image IM1 as shown in FIGS. 11A and 11B and the image IM4 rotated several degrees from the image IM1 is, for example, as shown in FIG. A correlation strength distribution is obtained. If the phase-only correlation method is simply used, it is difficult to detect the correlation due to rotational deviation.

For this reason, in the image collation apparatus 1a according to the present embodiment, the registered image AIM and the collation image RIM are subjected to Fourier-Melin transform, and the phase-only correlation method is used in the Fourier-Melin space to obtain the translation amount. And detecting magnification information and rotation angle information of the registered image AIM and the collation image RIM. Based on the information, the magnification and rotation angle of the verification image are corrected.
A translation shift between the corrected collation image RIM and the registered image AIM is detected by the phase only correlation method, and at the same time, collation between the images is performed based on the correlation peak.

  FIG. 12 is a diagram for explaining the correlation image data output by the phase only correlation unit 23. 12A is a registered image AIM, FIG. 12B is a collation image RIM, and FIG. 12C is a diagram showing a specific example of correlation image data. FIG. 13 is a diagram for explaining the correlation image data shown in FIG.

  When the registered image AIM and collation image RIM include, for example, binarized and linear patterns such as handwritten characters and blood vessel patterns, for example, the registered image AIM and collation image RIM as shown in FIGS. In this case, the phase only correlation unit 23 outputs correlation image data as shown in FIG.

  In FIG. 12C, the z-axis direction indicates the correlation peak intensity in the correlation image data, and the correlation peak intensity corresponds to the correlation degree between the registered image AIM and the collation image RIM. As shown in FIGS. 12C and 13, the correlation peak position PP in the x-axis direction and the y-axis direction has a deviation amount from the center of the image, for example, corresponding to the parallel movement amount between the registered image AIM and the collation image RIM. To do.

  The shift information generation unit 24 generates a shift amount from the image center as shift information based on the signal S23 output from the 23 phase only correlation unit 23, and outputs the signal S24 to the correction unit 25.

  The correction unit 25 performs position correction processing of the collation image RIM based on the signal S24 and the collation image RIM as deviation information output from the deviation information generation unit 24, and outputs the result as a signal S1612.

  FIG. 14 is a flowchart for explaining the operation of the image collating apparatus 1 shown in FIG. Only the differences from the first embodiment will be described with reference to FIG. 14 for the operation of the image collating apparatus 1a having the configuration described above.

For example, the registered image AIM and the collation image RIM are input by the image input unit 11, and each image data is stored in the memory (ST1).
Here, in order to obtain the magnification (enlargement / reduction ratio) and rotation angle information of the collation image RIM with respect to the registration image AIM, the registration image AIM is read from the memory 12 (ST2), and the Fourier of the magnification information-rotation information unit 21 is read. The transform unit 21111 performs Fourier transform processing (ST3), and Fourier image data S21111 is stored and stored in the memory 12 (ST4).
The amplitude component in the Fourier image data S21111 is subjected to logarithmic processing by the logarithmic conversion unit 21121 and converted to a logarithmic-polar coordinate system by the logarithmic-polar coordinate conversion unit 21131 (ST5).

  The collation image RIM is read from the memory 12 (ST6), and similarly subjected to Fourier transform processing by the Fourier transform unit 21112 (ST7). The amplitude component in the Fourier image data S21112 is subjected to logarithmic processing by the logarithmic transform unit 21122. Then, the logarithmic-polar coordinate conversion unit 21132 converts the logarithmic-polar coordinate system (ST8).

  The image signals (also referred to as pattern data) SA211 and SR211 obtained as a result of performing the Fourier-Melin transform on the registered image AIM and the collation image RIM described above are Fourier transformed by the Fourier transform units 2120 and 2121 of the phase only correlation unit 212, respectively. It is transformed (ST9), synthesized by the synthesizer 2122 (ST10), the amplitude component is removed from the synthesized signal by the phase extractor 2123 (ST11), and the remaining phase component is inverse Fourier transformed by the inverse Fourier transformer 2124. (ST12) Based on the amount of deviation of the peak position of the obtained correlation image data from the image center, the magnification information-rotation information generation unit 213 generates (detects) correction information including the magnification information and the rotation information. (ST13).

  The correction unit 22 performs correction processing for enlargement / reduction of the collation image and rotation processing based on the correction information, and removes the scaling component and the rotation component between the images (ST14). The remaining difference is only the translation component and is detected using the phase only correlation method.

  The corrected image RIM subjected to the correction process is Fourier-transformed by the Fourier transform unit 2312 of the phase-only correlation unit 23 (ST15) to generate Fourier image data S2312, and the Fourier-transformed registered image AIM stored in the memory 12 Is read (ST16), and the combining unit 232 generates combined data S232 (ST17).

At this time, the registered image AIM may be Fourier-transformed by the Fourier transform unit 2311 to generate Fourier image data S2311 and input to the synthesis unit 232.
The amplitude information in the synthesized data S232 is removed by the phase extraction unit 233 (ST18), the remaining phase information is input to the inverse Fourier transform unit 234, and the signal S23 is output as correlation image data (ST19).
Based on the signal 23, the shift information generation unit 24 detects the shift amount of the peak position, and outputs a signal S24 as shift information (ST20).
The correction unit 25 performs position correction processing of the registered image AIM and the collation image RIM based on the signal S24 as the deviation information and the collation image RIM, and outputs a signal S1612. Also, the registered image AIM is output as signal S1611 (ST21).

Hereinafter, the operation is the same as that of the first embodiment, the Hough transform unit 162 performs the Hough transform process (ST22), the extractor 163 performs the extract process (ST23), and the similarity generation unit 1641 generates the similarity Sim. (ST24).
In step ST25, the collation unit 1642 determines whether or not the similarity Sim is greater than (or exceeds) a preset threshold value. If it is greater, it is determined that the registered image AIM and the collation image RIM match ( ST26) If it is smaller than the threshold, it is determined that there is a mismatch (ST27).

  As described above, in the present embodiment, the position correction unit 161 uses the phase component as a result of the rotation angle correction process or the enlargement ratio correction process between the registered image AIM and the matching image RIM and the Fourier transform process as the position correction process. Since the correlation value is generated based on the position and the position correction processing of the registered image AIM and the matching image RIM is performed based on the generated correlation value, the position correction processing can be performed with high accuracy. As a result, the matching processing can be performed with high accuracy. It can be carried out.

Third Embodiment FIG. 15 is a functional block diagram of a third embodiment of an image collating apparatus according to the present invention.
The image collating apparatus 1b according to the present embodiment has a hardware functional block substantially the same configuration as the image collating apparatus 1 according to the first embodiment. For example, as illustrated in FIG. The memory 12 includes an FFT processing unit 13, a coordinate conversion unit 14, a Hough conversion unit 15, a CPU 16, and an operation processing unit 17.
The software functional block of the image collating device 1b has substantially the same configuration as that of the image collating device 1 according to the second embodiment. When the CPU 16 executes the program PRG, the position correcting unit 161b, the Hough converting unit 162, The functions of the extraction unit 163 and the determination unit 164b are realized. In FIG. 15, the same functions are denoted by the same reference numerals, description thereof is omitted, and only differences are described.

The major difference is that the position correction unit 161b and the determination unit 164b are different.
In the present embodiment, the position correction unit 161b generates a plurality of correlation values indicating the correction position by correlation processing based on the registered image AIM and the collation image RIM, and the registered image AIM based on the generated plurality of correlation values. In addition, a plurality of position correction processes for the collation image RIM are performed.

FIG. 16 is a diagram for explaining the operation of the position correction unit shown in FIG. The numerical value indicates the correlation peak intensity of the correlation image data on the XY plane of the correlation image data.
For example, in the case of a registered image AIM and a collation image RIM including a binarized line component (line shape) pattern, correlation peak intensities (also referred to as correlation intensities) between images having a large correlation are shown in FIGS. The value is small as shown in
For example, the deviation information generation unit 24 of the position correction unit 161b can generate, based on the signal S23, the top N correlation strengths, for example, eight correlation values and correlation peak positions in this embodiment, as shown in FIG. Is specified as a candidate for a two-dimensional positional relationship between the registered image AIM and the collation image RIM.
The position correction unit 161b performs a plurality of position correction processes as necessary based on the plurality of correlation values and the corresponding correlation peak positions.

  The Hough transform unit 162 performs a Hough transform process on each of the registered image AIM and the collation image RIM as a result of the plurality of position correction processes, and generates converted images S1621, S1622. The determination unit 164b generates a correlation value based on the patterns in the two converted images, and performs a matching process on the registered image AIM and the matching image RIM based on the generated correlation value and a preset threshold value. Further, the determination unit 164b performs a matching process based on the sum of correlation values corresponding to different positions and a preset threshold value based on the results of the plurality of position correction processes.

Specifically, the determination unit 164b includes a similarity generation unit 1641, a collation unit 1642b, and an integration unit 1643.
For example, as described above, the similarity generation unit 1641 generates the similarity Sim based on Expression (2) based on the signals S1631 and S1632, and outputs the similarity Sim to the integration unit 1643 and the collation unit 1642b as the signal S1641.

The integration unit 1643 integrates the similarity Sim based on the signal S1641, and outputs the integration result to the collation unit 1642b as a signal S1643.
The matching unit 1642b performs a matching process on the registered image AIM and the matching image RIM based on the signal S1641 and the signal S1643.
The difference from the collation unit 1642 according to the first embodiment is that the collation unit 1642b differs from the registered image AIM and the collation image when the signal S1643, which is the accumulated value of the similarity Sim by the accumulation unit, is larger than a predetermined threshold. The point is that the RIM matches.

  FIG. 17 is a flowchart for explaining the operation of the third embodiment of the image collating apparatus according to the present invention. With reference to FIGS. 14 and 17, the operation of the image collating apparatus will be described only with respect to the operation of the CPU 16, and only the differences from the first embodiment and the second embodiment.

Steps ST1 to ST19 are the same as those in the second embodiment shown in FIG.
On the basis of the correlation image data S23 output from the phase-only correlation unit 23 in step ST19, for example, N candidates Pi (P0, P1,. P2,..., Pn-1) are selected (ST28).

In step ST29, the integrating unit 1643 initializes variables for integration. For example, the variable i is initialized to 0 and the integrated value S is initialized to 0.
The correcting unit 25 performs position correction processing of the registered image AIM and the collation image RIM based on, for example, each candidate (coordinate) Pi and the amount of deviation from the center of the corresponding correlation image data (ST30).

  Next, as described above, the Hough conversion unit 162 performs a Hough conversion process for each of the registered image AIM and the collation image RIM (ST31), the extraction unit 163 performs an extraction process (ST32), and the similarity generation unit 1641 performs a similar process. Degree Sim (i) is calculated and output to integrating section 1643 and collating section 1642b (ST33).

The matching unit 1642b compares the similarity Sim (i) with a preset first threshold th1. If the similarity Sim (i) is smaller than the first threshold (ST34), the integrating unit 1643 Then, the similarity Sim (i) is integrated, and in detail, it is integrated by the formula S = S + Sim (i) and output to the collation unit 1642b (ST35).
In step ST35, the collation unit 1642b compares the integrated value S with a preset second threshold th2 (ST36). If the integrated value S is smaller than the second threshold th2, the variable i and the value N -1 is compared (ST27), and if the variable i does not match N-1, 1 is added to the variable i (ST28), and the process returns to step ST30. In step ST27, if the variable i matches N-1, it is assumed that the images do not match (ST39).

  On the other hand, in the comparison process of step ST34, the collation unit 1642b determines that the images match when the similarity Sim (i) is equal to or greater than the first threshold value. In the comparison process of step ST36, In the matching unit 1642b, when the integrated value S is equal to or greater than the second threshold th2, it is assumed that the images match (ST40). For example, the image matching device according to the present embodiment is applied to the vein pattern matching device in the security field. When applied, the operation processing unit 17 performs processing such as releasing the electronic lock.

  As described above, in the present embodiment, the position correction unit 161 generates a plurality of correlation values indicating the correction position, and based on the generated plurality of correlation values, a plurality of registered images AIM and matching images RIM. Position correction processing is performed, and the Hough conversion unit 162 performs Hough conversion processing as image processing for each of the registered image AIM and the verification image RIM as a result of the plurality of position correction processing, and the verification unit 1642b performs pattern conversion in each converted image. Since the collation process is performed based on the integrated value of the similarity as the correlation value according to, for example, even when the correlation between the two pieces of image data to be compared is small, the positional relationship between each of a plurality of candidates By summing up the similarities calculated for each of them, it is possible to perform collation with higher accuracy than in the case where collation is performed by the similarity alone.

  In addition, when the similarity Sim is larger than the first threshold th1, it is determined that they match, so that the matching process can be performed at high speed.

Fourth Embodiment FIG. 18 is a functional block diagram of an image collating apparatus according to a fourth embodiment of the present invention.
The image collating apparatus 1c according to the present embodiment has a hardware functional block substantially the same configuration as the image collating apparatus according to the above-described embodiment. For example, as illustrated in FIG. 12, an FFT processing unit 13, a coordinate conversion unit 14, a Hough conversion unit 15, a CPU 16, and an operation processing unit 17.
The software functional block of the image collating apparatus 1c has substantially the same configuration as that of the image collating apparatus according to the above-described embodiment. When the CPU 16 executes the program PRG, the position correcting unit 161, the Hough converting unit 162, and the extraction are performed. The functions of the unit 163c and the determination unit 164 are realized. In FIG. 18, the same functions are denoted by the same reference numerals, description thereof is omitted, and only differences are described.

A major difference is that the extraction unit 163c is different.
In the present embodiment, the extraction unit 163c extracts, for each of the first converted image and the second converted image, a region where the degree of overlapping of the curved pattern in one converted image is equal to or greater than a preset threshold value, Based on the size of the extracted region, the threshold value is controlled so that the size of the extracted region is larger than the set value.
Further, the extraction unit 163c controls the threshold based on the size of the extracted region so that the size of the extracted region is within the setting range.

FIG. 19 is a diagram for explaining the operation of the image collating apparatus 1c shown in FIG.
FIG. 19A shows a specific example of the image IM11. FIG. 19B is a diagram illustrating an image obtained by extracting a region equal to or greater than the first threshold value from the image IM11 illustrated in FIG. FIG. 19C is a diagram illustrating an image obtained by extracting a region greater than or equal to the second threshold value that is larger than the first threshold value from the image IM11 illustrated in FIG.
FIG. 19D is a diagram showing a specific example of the image IM12. FIG. 19E is a diagram illustrating an image obtained by extracting a region having a first threshold value or more from the image IM12 illustrated in FIG. FIG. 19F is a diagram illustrating an image obtained by extracting a region equal to or larger than the second threshold value that is larger than the first threshold value from the image IM11 illustrated in FIG.

  For example, in the above-described embodiment, the threshold value is fixed when the feature value (parameter) is extracted from the converted image after the Hough transform process is performed on the registered image or the collation image. May not be extracted, or on the contrary, redundant feature amounts may be extracted.

  Specifically, for example, after performing a Hough transform process on the image IM11 illustrated in FIG. 19A, and extracting an area larger than the first threshold from the converted image, an image IM111 as illustrated in FIG. 19B is obtained. When an area that is generated and larger than the second threshold is extracted, an image IM112 as shown in FIG. 19C is generated.

  Further, for example, when a region larger than the first threshold is extracted from the converted image after the Hough conversion process is performed on the image IM12 illustrated in FIG. 19D, an image IM121 illustrated in FIG. 19E is generated. When an area larger than the second threshold is extracted, an image IM122 as shown in FIG. 19F is generated.

For example, the image IM112 and the image IM121 are a specific example having an appropriate data amount (feature amount) regarding the collation processing.
The image IM111 is a specific example having a redundant data amount (feature amount) regarding the collation processing.
The image IM122 is a specific example in which a sufficient data amount (feature amount) cannot be extracted with respect to the matching process.

For example, when the threshold value is fixed, the comparison / collation processing is performed between the image IM111 and the image IM121 or between the image IM112 and the image 122, so that sufficient determination accuracy may not be ensured.
For example, when the threshold value is changed and the comparison / collation processing is performed between the image IM112 and the image IM121 having a sufficient data amount (feature amount) regarding the collation processing, the determination accuracy is improved.

  Therefore, when the extraction unit 163c according to the present embodiment extracts a feature amount from the converted image after the Hough conversion process as described above, the feature amount extracted using a certain threshold value is in a range suitable for matching. If it is not within the range, the threshold value is changed and the feature quantity is extracted again, and control is performed so that the feature quantity falls within an appropriate range.

  FIG. 20 is a flowchart for explaining the operation of the image collating apparatus according to the present embodiment. With reference to FIG. 20, the operation of the image collating apparatus will be described focusing on the differences from the embodiment shown in FIG.

Since operations in steps ST1 to ST22 are the same, description thereof is omitted.
In step ST231, the extraction unit 163c extracts, as the feature amount, only the parameter portion exceeding the preset threshold value in each of the converted images S1612 and S1622 generated by the Hough conversion unit 162.

Based on the size of the extracted region, the extraction unit 163c controls the threshold so that the size of the extracted region is within the set range.
Specifically, based on the size of the extracted region, the extraction unit 163c controls the threshold so that the size of the extracted region is larger than the minimum value min, and the size of the extracted region is The threshold value is controlled to be smaller than a maximum value max that is larger than the minimum value min.

For example, in step ST232, the extraction unit 163c determines whether or not the size of the region extracted as the feature amount is larger than the minimum value min. If it is determined that the size is not larger, the extraction unit 163c decreases the threshold by a predetermined amount ( (Reduced) (ST233), the process returns to step STST231.
On the other hand, in step ST232, if the extraction unit 163c determines that the size of the region extracted as the feature amount is larger than the minimum value min, the process proceeds to step ST234.

  In step ST234, the extraction unit 163c determines whether or not the size of the region extracted as the feature amount is smaller than the maximum value max. If it is determined that the region is not small, the extraction unit 163c increases the threshold by a predetermined amount (larger). (ST233), the process returns to step STST231. On the other hand, in step ST234, if the extraction unit 163c determines that the size of the region extracted as the feature amount is smaller than the maximum value max, the extraction unit 163c proceeds to the process of step ST24.

  The processing after step ST24 is the same as the processing in FIG.

As described above, the extraction unit 163c controls the threshold value based on the size of the extracted region so that the size of the extracted region is within the set range. Further, it is possible to extract optimum data for the collation process.
Further, an image suitable for the collation process can be obtained, and the collation unit 164 can perform the collation process with higher accuracy.
Further, for example, even when the imaging system changes significantly and the information amount of the input image changes extremely, the comparison / collation process can be handled without any change.

Fifth Embodiment FIG. 21 is a flowchart for explaining the operation of the fifth embodiment of the image collating apparatus according to the present invention.
The image collation processing device according to the present embodiment has a hardware functional block that has substantially the same configuration as the image collation device according to the above-described embodiment. For example, as illustrated in FIG. 12, an FFT processing unit 13, a coordinate conversion unit 14, a Hough conversion unit 15, a CPU 16, and an operation processing unit 17.

For example, when the feature amount of the collation image or registered image is remarkably small, there may be a case where a feature amount sufficient for comparison and collation cannot be obtained and sufficient determination accuracy cannot be obtained.
In the image matching apparatus according to the present embodiment, an area where the degree of overlapping of the pattern of curves in one converted image is equal to or larger than a preset threshold is extracted, and the size of the extracted area is equal to or smaller than a set value. Discard the image.

  Hereinafter, with reference to FIG. 21, a description will be given of a case where certain image data is stored as a registered image for collation in applications such as a personal authentication device using a blood vessel pattern, with a focus on differences from the above-described embodiment. .

  In step ST <b> 101, the CPU 16 uses the image input unit 11 to capture image data to be saved as the registered image AIM, and stores the image data in the memory 12. The CPU 16 causes the Hough converter 15 to perform a Hough conversion process on the captured image data (ST102).

In step ST103, the CPU 16 extracts a region exceeding a preset threshold value as a feature amount (ST103).
In step ST104, the CPU 16 determines whether or not the size of the extracted area is larger than the minimum value min necessary for the collation process. If it is larger, the CPU 16 has a sufficient amount of information as the registered image AIM. Are stored in the memory 12 (ST105).

  On the other hand, if it is not larger than the minimum value min in step ST104, the input image is discarded (ST106), and the image data is displayed by displaying that the amount of information as the registered image AIM is insufficient. Is requested again (ST107). Hereinafter, the above-described collation process is performed.

  As described above, in the present embodiment, at the time of registration, an area where the degree of overlap of curve patterns in the registered conversion image is equal to or greater than a preset threshold is extracted, and the size of the extracted area is set to a set value ( Since the image is discarded when it is equal to or smaller than the minimum value (min), a registered image having a relatively large amount of features related to the matching process can be obtained with certainty. Further, the collation process can be performed with high accuracy.

Sixth Embodiment FIG. 22 is a flowchart for explaining the operation of an image collating apparatus according to a sixth embodiment of the image collating apparatus according to the present invention.
Next, the case where the collation image RIM input from the outside is collated with the registered image AIM stored in the memory 12 in advance will be described with reference to FIG.

  In step ST <b> 201, the CPU 16 uses the image input unit 11 to capture image data as a collation image RIM and stores the image data in the memory 12. The CPU 16 causes the Hough converter 15 to perform a Hough conversion process on the captured collation image RIM (ST202).

In step ST203, the CPU 16 extracts a region exceeding a preset threshold value as a feature amount (ST203).
In step ST204, the CPU 16 determines whether or not the size of the extracted area is larger than the minimum value min necessary for the collation process. If not, the CPU 16 has a sufficient amount of information as the collation image RIM. If not, the image is discarded (ST205), a message indicating that the amount of information is insufficient is displayed, the user is prompted to re-input the image data (ST206), and the collation process is terminated.

  On the other hand, if it is determined in step ST204 that the value is larger than the minimum value min, it is determined that the verification image RIM has a sufficient amount of information, and the subsequent verification processing in steps ST2 to ST27 is performed.

  Hereinafter, the operations in steps ST2 to ST27 are the same as the operations shown in FIG.

  As described above, in the present embodiment, at the time of matching, a region where the degree of overlapping of the curve pattern in the matching conversion image is equal to or greater than a preset threshold value is extracted, and the size of the extracted region is a set value ( Since the image is discarded when it is equal to or smaller than the minimum value (min), it is possible to reliably obtain a collation image having a relatively large amount of features related to the collation processing. Further, the collation process can be performed with high accuracy.

Note that the present invention is not limited to the present embodiment, and various suitable modifications can be made.
For example, in the present embodiment, the similarity generation unit calculates the similarity using Equation (2), but the present invention is not limited to this form. For example, the similarity generation unit only needs to perform a process of calculating a similarity suitable for the correlation of linear components (line-shaped patterns).

  In the present embodiment, Fourier transform processing, logarithmic transformation processing, and logarithm-station coordinate transformation are performed, and the magnification information and the rotation angle information are generated by calculating the parallel movement amount in the Fourier-Merlin space. It is not limited to form. For example, coordinate conversion may be performed in a space where magnification information and rotation angle information can be detected.

  In addition, the first threshold th1 and the second threshold th2 are fixed values, but the present invention is not limited to this form. For example, by making each of the threshold values variable depending on the image pattern, it is possible to perform more accurate collation.

  DESCRIPTION OF SYMBOLS 1, 1a, 1b ... Image collation apparatus, 11 ... Image input part, 12 ... Memory, 13 ... FFT processing part, 14 ... Coordinate conversion part, 15 ... Hough conversion part, 16 ... CPU, 17 ... Operation processing part, 21 ... Magnification information-rotation information part, 22 ... correction part, 23 ... parallel movement part, 24 ... deviation information generation part, 161, 161b ... position correction part, 162 ... Hough conversion part, 163 ... extraction part, 164,164b ... discrimination part 211 ... Fourier-Merin transform unit 212 ... Phase-only correlation unit 213 ... Magnification information-rotation information generation unit 232 ... Synthesis unit 233 ... Phase extraction unit 234 ... Inverse Fourier transform unit 241 ... Candidate identification unit, 242 ... Similarity generation unit, 243 ... integration unit, 244 ... collation unit, 1641 ... similarity generation unit, 1642, 1642b ... collation unit, 1643 ... integration unit, 2120, 2121 ... Fourier transform unit, 2122 Synthesis unit, 2123 ... phase extraction unit, 2124 ... inverse Fourier transform unit, 2311, 2312 ... Fourier transform unit, 2421 ... similarity correction unit, 211111, 21112 ... Fourier transform unit, 2111, 21212 ... logarithmic transform unit, 21131, 21132 ... Logarithm-polar coordinate converter.

Claims (4)

Blood vessel image acquisition means for acquiring a blood vessel image;
Run the Hough transform to generate a plurality of first linear component in the blood vessel image acquired by the blood vessel image acquiring means, a plurality of respectively show some of each line components for the first linear component of this plurality of first Linear component conversion means for converting into one linear component information;
Linear component extraction means for extracting a plurality of second linear components that are larger than an extraction threshold from the plurality of first linear component information obtained as a result of conversion in the linear component conversion means, the linear component conversion means For the plurality of first linear component information obtained as a result of the conversion, a plurality of regions where the degree of overlapping of the curve patterns in one converted image is greater than or equal to a preset threshold value are extracted, and the plurality of second line components are extracted. Linear component extraction means for extracting a linear component;
Based on the relative positional relationship between the second linear components extracted in said plurality in the straight line component extracting means, by comparing the blood vessel information for collation Hough transform it is executed, acquired by the blood vessel image acquiring means Authentication means for authenticating that the blood vessel image corresponding to the verification blood vessel information;
A biometric authentication device. Blood vessel image acquisition means for acquiring a blood vessel image;
Run the Hough transform to generate a plurality of first linear component in the blood vessel image acquired by the blood vessel image acquiring means, a plurality of respectively show some of each line components for the first linear component of this plurality of first Linear component conversion means for converting into one linear component information;
Linear component extraction means for extracting a plurality of second linear components that are larger than an extraction threshold from the plurality of first linear component information obtained as a result of conversion in the linear component conversion means, the linear component conversion means For the first linear component information obtained as a result of the conversion, a plurality of regions where the degree of overlapping of the curve patterns in one converted image is greater than or equal to a preset threshold are extracted, and the plurality of second linear components are extracted. Extracting the linear component, and
The linear component extraction means second linear components extracted in said plurality, a storage means for storing in the storage unit as the information used for the authentication process,
Comprising
Blood vessel information storage device. A blood vessel image acquisition step of acquiring a blood vessel image;
A plurality of first linear components are generated by performing Hough transform on the blood vessel image acquired in the blood vessel image acquiring step to generate a plurality of first linear components, and each of the plurality of first linear components indicates the degree of each linear component . A linear component conversion process for converting into linear component information of
A linear component extraction step for extracting a plurality of second linear components greater than an extraction threshold from the plurality of first linear component information obtained as a result of the conversion in the linear component conversion step, wherein the linear component conversion step For the first linear component information obtained as a result of the conversion, a plurality of regions where the degree of overlapping of the curve patterns in one converted image is greater than or equal to a preset threshold are extracted, and the plurality of second linear components are extracted. Extracting a linear component, and
Based on the relative positional relationship of the plurality of second linear component extracted in the straight line component extraction step, by matching the matching vessel information Hough transform has been performed, the acquired in the blood vessel image acquiring step An authentication step of authenticating that the blood vessel image corresponds to the blood vessel information for verification;
A biometric authentication method. A blood vessel image acquisition step of acquiring a blood vessel image;
A plurality of first linear components are generated by performing Hough transform on the blood vessel image acquired in the blood vessel image acquiring step to generate a plurality of first linear components, and each of the plurality of first linear components indicates the degree of each linear component . A linear component conversion process for converting into linear component information of
A linear component extraction step for extracting a plurality of second linear components greater than an extraction threshold from the plurality of first linear component information obtained as a result of the conversion in the linear component conversion step, wherein the linear component conversion step For the first linear component information obtained as a result of the conversion, a plurality of regions where the degree of overlapping of the curve patterns in one converted image is greater than or equal to a preset threshold are extracted, and the plurality of second linear components are extracted. Extracting a linear component, and
A second linear components extracted in the plural form in the straight line component extracting step, a storage step of storing in the storage unit as the information used for the authentication process,
A blood vessel information storage method comprising:

Images