The present invention relates to an image matching method for performing a matching images to linear components in a first image and a second image and apparatus thereof. TECHNICAL FIELD [0001] The present invention relates to an image matching method and an image matching apparatus for matching two images such as blood vessel images, fingerprint images, still images or moving images, based on linear components within the images, and a program for the same. BACKGROUND ART [0002] Conventionally, there is known various image matching apparatuses as an apparatus for performing a matching images to an image information. For example, there is known an image matching apparatus in which a registered image and an image to be matched are compared based on a predetermined positional relationship, correlation values thereof are calculated, and the matching of the registered image and the image to be matched are carried out based on the correlation values. Otherwise, in the case where the correlation values are generated, there is also known an image matching apparatus in which the correlation values are generated by an operation in every pixel unit. DISCLOSURE OF THE INVENTION PROBLEM TO BE SOLVED BY THE INVENTION [0003] However, in the above image matching apparatus, if, there are pattern-grades such as many linear components in the images even if a correlation of the two images is small, and then there are many crossed points of the linear components in the registered image and the image to be matched, the crossed points greatly contribute to the correlation value to increase the correlation values, and, as a result, a sufficient matching accuracy may not be obtain. So there is demanded to improve the apparatus. [0004] The present invention was made in consideration of the above discussions, and, an object of the present invention is to provide an image matching method, an image matching apparatus, and a program for achieving an image matching at a high accuracy. MEANS FOR SOLVING THE PROGRAM [0005] To achieve the above object, according to a first aspect of the present invention, there is provided an image matching method for performing a matching images to linear components in a first image and a second image, the method having: a first step of performing an image processing for transforming points in each image of the first image and the second image to a curved pattern and transforming the linear component in the image to a plurality of overlapped curved-patterns, based on a distance from a reference position to a shortest point in a straight line passing through a point in the image and an angle between a straight line passing though the reference position and the shortest point and a reference axis including the reference position, and generating a first transformed image and a second transformed image, and a second step of performing a matching of the first image and the second image based on a degree of an overlap of the patterns in the first transformed image and the second transformed image generated in the first step and a matching or mismatching of the patterns in the first and second transformed images. [0006] Further, to achieve the above object, according to a second aspect of the present invention, there is provided an image matching method for performing a matching images to linear components in a first image and a second image, the method having: a first step of performing a Hough transform processing to the first image and the second image to generate a first transform image and a second transform image, and a second step of performing a matching of the first image and the second image based on a degree of an overlap of patterns in the first transformed image and the second transformed image generated in the first step and a matching or mismatching of the patterns in the first and second transformed images. [0007] Further, to achieve the above object, according to a third aspect of the present invention, there is provided an image matching apparatus performing a matching images to linear components in a first image and a second image, the apparatus having: a transform means for performing an image processing to the first image and the second image, by which points in each image are transformed to a curved pattern and the linear components in each image are transformed to a plurality of overlapped curved-patterns based on a distance from a reference position to a shortest point in a straight line passing through a point in the image and an angle between a straight line passing though the reference position and the shortest point and a reference axis including the reference position, and generating a first transformed image and a second transformed image, and a matching means for performing a matching of the first image and the second image based on a degree of an overlap of the patterns in the first transformed image and the second transformed image generated by the transform means and a matching or mismatching of the patterns in the first and second transformed images. [0008] Further, to achieve the above object, according to a fourth aspect of the present invention, there is provided an image matching apparatus performing a matching images to linear components in a first image and a second image, the apparatus having: a transform means for performing a Hough transform processing to the first image and the second image to generate a first transformed image and a second transformed image, and a matching means for performing a matching of the first image and the second image based on a degree of an overlap of patterns in the first transformed image and the second transformed image generated by the transform means and a matching or mismatching of the patterns in the first and second transformed images. [0009] Further, to achieve the above object, according to a fifth aspect of the present invention, there is provided a program that causes an information processing device to perform a matching images to linear components in a first image and a second image, the program having: a first routine for performing an image processing to the first image and the second image, by which points in each image are transformed to a curved pattern and linear components in each image are transformed to a plurality of overlapped curved-patterns based on a distance from a reference position to a shortest point through a point in the image toward a straight line and an angle between a straight line passing though the reference position and the shortest point and a reference axis including the reference position, and generating a first transformed image and a second transformed image, and a second routine for performing a matching of the first image and the second image based on a degree of an overlap of the patterns in the first transformed image and the second transformed image generated in the first routine and a matching or mismatching of the patterns in the same. [0010] Further, to achieve the above object, according to a sixth aspect of the present invention, there is provided a program that causes an information processing device to perform a matching images to linear components in a first image and a second image, the program having: a first routine for performing a Hough transform processing to the first image and the second image to generate a first transformed image and a second transformed image, and a second routine for performing a matching of the first image and the second image based on a degree of an overlap of patterns in the first transformed image and the second transformed image generated in the first routine and a matching or mismatching of the patterns in the first and second transformed images. [0011] According to the present invention, in the first step and in the first routine, a step transform means performs an image processing for transforming points in each image of the first image and the second image to a curved pattern and transforming linear components in each image to a plurality of overlapped curved-patterns, based on a distance from a reference position to a shortest point in a straight line passing through a position in the image and an angle between a straight line through the reference position and the shortest position and a reference axis including the reference.position, and generating the first transformed image and the second transformed image. In the second step and in the second routine, the matching means matches the first image and the second image, based on the degree of the overlap of the patterns in the first transformed image and the second transformed image generated by the transform means and a matching or mismatching of the patterns in the first and second transformed images. EFFECT OF THE INVENTION [0012] According of the present invention, there is provided an image matching method, an image matching apparatus, and a program, enabling the image matching at a high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS [0013] [FIG. 1] FIG. 1 is a functional block diagram illustrated in a form of hardware of an image matching apparatus of a first embodiment according to the present invention. [FIG. 2] FIG. 2 is a functional block diagram illustrated in a form of software of the image matching apparatus shown in FIG. 1. [FIG. 3] FIGs. 3A and 3B are views for explaining an operation of a Hough transform unit shown in FIG. 2. [FIG. 4] FIGs. 4A to 4F are views for explaining the operation of the Hough transform unit shown in FIG. 2. [FIG. 5] FIGs. 5A to 5C are views for explaining an operation of a similarity generation unit shown in FIG. 2. [FIG. 6] FIG. 6 is a flow chart for explaining an operation of the image matching apparatus according to the present embodiment shown in FIG. 1. [FIG. 7] FIG. 7 is a view showing a position correction unit of an image matching apparatus of a second embodiment according to the present invention. [FIG. 8] FIG. 8 is a view for explaining an operation of a Fourier and Mellin transform unit 211 shown in FIG. 7. [FIG. 9] FIGs. 9A to 9F are views for explaining different points between a self-correlation method and a phase-only correlation method. [FIG. 10] FIGs. 10A to IOC are views for explaining correlation intensity distributions in the phase-only correlation method in the case where a parallel movement shift exists between two images. [FIG. 11] FIGs. 11A to 11C are views for explaining correlation intensity distributions in the phase-only method in the case where a rotation shift exists between two images. [FIG. 12] FIGs. 12A to 12C are views for explaining correlation image data output by a phase-only correlation unit 23. [FIG. 13] FIG. 13 is a view for explaining the correlation image data shown in FIG. 12C. [FIG. 14] FIG. 14 is a flow chart for explaining an operation of the image matching apparatus 1 shown in FIG. 1. [FIG. 15] FIG. 15 is a functional block diagram of an image matching apparatus of a third embodiment according to the present invention. [FIG. 16] FIGs. 16A and 16B are views for explaining an operation of a position correction unit shown in FIG. 15. [FIG. 17] FIG. 17 is a flow chart for explaining an operation of the image matching apparatus of the third embodiment according to the present invention. [FIG. 18] FIG. 18 is a functional block diagram of an image matching apparatus of a fourth embodiment according to the present invention. [FIG. 19] FIGs. 19A to 19F are views for explaining an operation of the image matching apparatus Ic shown in FIG. 17; FIG. 19A is a view showing a specific example of an image IM11; FIG. 19B is a view showing an image in which a region equal to or greater than a first threshold is extracted from the image IM11 shown in FIG. 19A; FIG. 19C is a view showing an image in which a region equal to or greater than a second threshold larger than the first threshold is extracted from the image IM11 shown in FIG. 19A; FIG. 19D is a view showing a specific example of an image IM12; FIG. 19E is a view showing an image in which a region equal to or greater than a first threshold is extracted from the image IM12 shown in FIG. 19D; and FIG. 19F is a view showing an image in which a region equal to or greater than a second threshold, larger than the first threshold, is extracted from the image IM11 shown in FIG. 19D. [FIG. 20] FIG. 20 is a flow chart for explaining an operation of the image matching apparatus according to the present embodiment. [FIG. 21] FIG. 21 is a flow chart for explaining an operation of an image matching apparatus of a fifth embodiment according to the present invention. [FIG. 22] FIG. 22 is a flow chart for explaining an operation of an image matching apparatus according to a sixth embodiment according to the present invention. BEST MODE FOR CARRYING OUT THE INVENTION [0015] FIG. 1 is a functional block diagram illustrated in a form of hardware of an image matching apparatus of a first embodiment according to the present invention. An image matching apparatus 1 according to the present embodiment, as shown in FIG. 1, has an image input unit 11, a memory 12, a FFT processing unit 13, a coordinates transform unit 14, a Hough transform unit 15, a CPU 16, and an operation processing unit 17. For example, the image input unit 11 is connected to the memory 12. The memory 12, the FFT processing unit 13, the coordinates transform unit 14, the Hough transform unit 15, and the CPU 16 are connected by bus BS. The operation processing unit 17 is connected to the CPU 16. [0016] The image input unit 11 functions as an input unit for inputting an image from the outside. For example a registered image AIM and an image to be compared with the registered image AIM (referred to an image to be matched RIM, also) are input to the image input unit 11. The memory 12, for example, stores an image input from the image input unit 11. The memory 12 stores the registered image AIM, the image to be matched RIM, and a program PRG as shown in FIG. 1. The program PRG is executed, for example, by the CPU 16, and includes routines for realizing functions concerning a transform processing, a correlation processing, and a matching processing, according to the present invention. [0017] The FFT processing unit 13, for example under control of the CPU 16, performs a two-dimensional Fourier transform processing to an image stored in the memory 12, for example, and outputs the processed result to the coordinates transform unit 14 and the CPU 16. [0018] The coordinates transform unit 14, for example under control of the CPU 16, performs a logarithm-polar coordinates transform processing to the result of the two-dimensional Fourier transform processing performed by the FFT processing unit 13, and outputs the coordinates transformed result to the CPU 16. The operation processing unit 17 performs a predetermined processing such as a release of an electric-key in the case where the registered image AIM and the image to be matched RIM are matched, for example, based on the processed result by the CPU 16 as mentioned later. [0019] The Hough transform unit 15 performs a Hough transform processing as mentioned later under control of the CPU 16, and outputs the processed result to the CPU 16. The Hough transform unit 15 is preferably used with an exclusive circuit formed by a hardware in order to perform, for example, the Hough transform processing at a high speed. [0020] The CPU 16 performs a matching processing to the registered image AIM and the image to be matched RIM stored in the memory 12, for example, based on the program PRG according to the present embodiment of the present invention. The CPU 16 controls, for example, the image input unit 11, the memory 12, the FFT processing unit 13, the coordinates transform unit 14, the Hough transform unit 15, and the operation processing unit 17, to realize the operation according to the present embodiment. [0021] FIG. 2 is a functional block diagram illustrated in a form of a software of the image matching apparatus shown in FIG. 1. For example, the CPU 16 executes the program PRG stored in the memory 12 to realize functions of a position correction unit 161, a Hough transform unit 162, an extraction unit 163, and a judgment unit 164 as shown in FIG. 2. The position correction unit 161 corresponds to a position correction means according to the present invention, the Hough transform unit 162 corresponds to a transform means according to the present invention, the extraction unit 163 corresponds to an extraction means according to the present invention, and the judgment unit 164 corresponds to a matching means according to the present invention. [0022] The position correction unit 161, for example, corrects a positional shift to an image pattern in the registered image AIM and the image to be matched RIM stored in the memory 12 in a right and left direction and an up and down direction, an enlargement ratio, and a rotation angle shift of the two images, and outputs the corrected image to the Hough transform unit 162. In detail, for example, the positional correction unit 161 performs a position correction processing to the registered image AIM and the image to be matched RIM and outputs the result as signals S1611 and S1612 to the Hough transform unit 162. [0023] The Hough transform unit 162, for example, executes the Hough transform processing in the Hough transform unit 15 performing the exclusive Hough transform processing in hardware. In detail, for example, the Hough transform unit 162 performs the Hough transform processing to the signal SI611 which is the registered image AIM to which the position correction processing was performed, and outputs the processing result as the signal S1621. The Hough transform unit 162 performs the Hough transform processing to the signal SI612 which is the image to be matched RIM to which the position correction processing was performed, and outputs the result as a signal SI622. [0024] FIGs. 3A and 3B are views for explaining an operation of the Hough transform unit shown in FIG. 2. The Hough transform unit 162 performs an image processing, for example, to the first image and the second image respectively, by which the points in each image are transformed to curved patterns and by which linear components in each image are transformed to a plurality of overlapped curved patterns, based on a distance pO from a reference position O to a shortest point PO in a straight line LO passing through a point in the image and an angle 6 between a straight line nO passing though the reference position 0 and the shortest point PO and a reference axis including the reference position O, to thereby generate a first transformed image and a second transformed image. [0025] For simplifying explanations, for example, as shown in FIG. 3A, it is assumed that, in an x-y plane, there is the straight line LO, and a position P (xl, yl), a position P2 (x2, y2), and a position P3 (x3, y3) on the straight line LO. It is assumed that a straight line through the origin O and perpendicular to the straight line LO is the line nO, so the straight line nO and the x-axis as the reference axis are crossed at an angle 60 and that the origin O and the straight line LO are separated with a distance |pO|, for example. Here, |pO| indicates the absolute value of pO. The straight line LO can be expressed by parameters (pO, 00). The Hough transform processing performed to coordinates (x, y) in the x-y plane, is defined by formula (1), for example. [0026] [Formula 1] [0027] For example, when the Hough transform processing expressed by the formula (1) is performed to the positions PI, P2, and P3, the positions PI, P2, and P3 are transformed to curved lines in a p-0 space as shown in FIG. 3B. In detail, by the Hough transform processing, the point Pl(xl, yl) is transformed to a curved line PLl (xl-cos0 + yl-sin9) , the point P2 (x2, y2) is transformed to a curved line PL2 (x2-cos0 + y2-sin6) , and the point P3(x3, y3) is transformed to a curved line PL3 (x3-cos0 + y3-sin0) . Patterns of the curved lines (curved patterns) PLl, PL2, and PL3 are crossed at a crossed point CP(pO, 00) in the p-0 space. The crossed point P(pO, 00) in the p-0 space corresponds to the linear component LO in the x-y space. Conversely, as shown in FIG. 3A, the linear component LO on the x-y plane corresponds to the crossed point CP of the curved patterns PLl, PL2, and PL3 on the p-0 space. [0028] As described above, the Hough transform processing is performed to a two-binary image, it may judge which linear components dominates in the x-y plane before the transform processing, based on the degree of the overlap of the curved patterns in the above resultant p-0 space. [0029] FIGs. 4A to 4F are views for explaining an operation of the Hough transform unit shown in FIG. 2. The Hough transform unit 162 performs the Hough transform processing to the registered image AIM to which the position correction processing was performed, shown in FIG. 4A for example, to generate the image SI621 shown in FIG. 4C, and performs the Hough transform processing to the image to be matched RIM to which the position correction processing was performed, shown in FIG. 4B, to generate the image SI622. In the respective pixels included in the images S1621 and S1622, a value corresponding to the degree of the overlap of the curved patterns is set. In the present embodiment, in the images to be displayed based on a predetermined gradation, a pixel(s) in which the degree of the overlap of the curved patterns is large, is displayed in white. As mentioned later, the matching unit 1642 performs a matching processing based on the degree of the overlap of the curved patterns, so it performs the matching processing based on the linear component in the original x-y space. [0030] The extraction unit 163, from the respective first transformed image and the second transformed image, extracts regions each of which indicates the degree of the overlap of the curved patterns in the transformed image equal to or greater than a threshold set in advance. In detail, the extraction unit 163, for example, extracts the region of which the degree of the overlap of the curved patterns in the transformed image is equal to or greater than the threshold set in advance, from the signal S1621 as the first transformed image shown in FIG. 4C, generates an image S1631 shown in FIG. 4E, and outputs the same to the matching unit 1642. Also, the extraction unit 163, for example, extracts the region of which the degree of the overlap of the curved patterns in the transformed image is equal to or greater than the threshold set in advance, from the signal' 81622 as the second transformed image shown in FIG. 4D, generates an image SI632 shown in FIG. 4F, and outputs the same to the matching unit 1642. By performing the extraction processing, for example, noise components different from the linear components in the registered image AIM and the image to be matched RIM, in the x-y space, for example, point components, are deleted. [0031] The judgment unit 164 performs the matching of the first image and the second image based on the degree of the overlap of the patterns in the first transformed image and the second transformed image and a matching or mismatching of the patterns in the two images. In detail, the judgment unit 164 performs the matching, for example, to the signal S1631 and the signal S1632, and outputs the matched result as a signal S164. [0032] The judgment unit 164, for example, as shown in FIG. 2, has a similarity generation unit 1641 and a matching unit 1642. The similarity generation unit 1641, for example, performs a comparison processing to a plurality of different positional relationships in the first transformed image and the second transformed image, and generates a similarity as the correlation value based on the compared result. In detail, the similarity generation unit 1641, for example, performs the comparison processing to a plurality of different positional relationships in the two images of the signal S1631 and the signal S1632, and generates the similarity as the correlation value based on the compared result. [0033] For example, the similarity generation unit 1641, assuming the two images as f1(m, n) and f2(m, n), calculates a similarity Sim by applying the following formula (2) and outputs the result as S1641. [0034] (Formula Removed) [0035] FIGs. 5A to 5C are views for explaining an operation of the similarity generation unit shown in FIG. 2. The similarity generation unit 1641, for example, in the case where the similarity of two images including linear components (also, referred to linear shapes) shown in FIGs. 5A and 5B is generated, generates the similarity corresponding to a number of the crossed points CP of the two images as shown in FIG. 5C. Here, for simplifying explanations, the linear components are indicated by black pixels of a bit value xl' and others are indicated by white pixels of a bit value X0'. [0036] The matching unit 1642 performs the matching of the image to be matched RIM and the registered image AIM based on the signal S1641 indicating the similarity generated by the similarity generation unit 1641. For example, the matching unit 1642 judges that the registered image AIM and the image to be matched RIM are matched in the case where the similarity is larger than a predetermined value. [0037] FIG. 6 is a flow chart for explaining an operation of the image matching apparatus according to the present embodiment shown in FIG. 1. While referring to FIG. 6, the operation of the image matching apparatus, mainly an operation of the CPU, will be described. [0038] For example, the registered image AIM and the image to be matched RIM are input from the image input unit 11, and the memory 12 stores the respective input data. At step ST101, for example as shown in FIG. 2, the position correction unit 161 performs the position correction processing based on the image patterns in the registered image AIM and the image to be matched RIM stored in the memory 12, in detail, corrects a positional shift in a right and left direction and an up and down direction, a magnification ratio, and the rotation angle shift in the two images, and outputs the corrected images as the signals S1611 and S1612 to the Hough transform unit 162. [0039] At step ST102, the Hough transform unit 162, for example, performs the image processing to the signal S1611 being the registered image AIM to which the position correction processing was performed shown in FIG. 4A, the image processing transforming the positions in each image to the curved pattern PL and the linear components L in each image to a plurality of the overlapped curved patterns PL as shown in FIG. 3A, based on the distance pO from the reference position O to the shortest point PO in the straight line LO passing through the point in the image and the angle 0 between the straight line nO passing through the reference position O and the shortest point PO and the x-axis as the reference axis including the reference position O, to generate the signal SI62 as the transformed image in the p-0 space as shown in FIG. 4C. The Hough transform unit 162 performs the Hough transform processing, in the same way, to the signal SI612 being the image to be matched RIM to which the position correction processing was performed as shown in FIG. 4B, to generate the signal S1622 as the transformed image in the p-9 space as shown in FIG. 4D, for example. [0040] At step ST103, the extraction unit 163 extracts the regions each of which indicates the degree of the overlap of the curved patterns in the transformed image equal to or greater than the threshold set in advance, from the transformed image S1621 and the transformed image S1622 respectively. In detail, as described above, the value corresponding to the degree of the overlap of the curved patterns is set in the respective pixels in the images S1621 and S1622, and in the image to be displayed based on a predetermined gradation, portions of which the degree of the overlap of the curved patterns are high, are displayed in white. The extraction unit 163, for example, extracts the region of which the degree of the overlap of the curved patterns in the transformed image SI621 shown in FIG. 4C is equal to or greater than the threshold set in advance, generates the image SI631 shown in FIG. 4E for example, and outputs the same to the matching unit 1642. The extraction unit 163, for example, extracts the region of which the degree of the overlap of the curved patterns in the transformed image Si622 shown in FIG. 4D is equal to or greater than the threshold set in advance, generates the image SI632 shown in FIG. 4F for example, and outputs the same to the matching unit 1642. [0041] The matching unit 1642 performs the matching of the image to be matched RIM and the registered image AIM based on the degree of the overlap of the patterns in the transformed images S1631 and S1632 and the matching or mismatching of the patterns in the transformed images S1631 and S1632. [0042] In detail, at step S1104, the similarity generation unit 1641 performs the comparison processing to a plurality of the different positional relationships in the transformed images S1631 and S1632, generates the similarity Sim as the correlation value based on the compared result, for example, as expressed by the formula (2) , and outputs the same as the signal S1641. [0043] At step ST105, the matching unit 1642 performs the matching of the image to be matched RIM and the registered image AIM based on the similarity Sim functioning as the correlation value generated by the similarity generation unit 1641. In detail, the matching, unit 1642 judges that the registered image AIM and the image to be matched RIM are matched in the case where the similarity Sim is greater than the threshold set in advance, otherwise judges that they are mismatched in the case where the similarity Sim is equal to or less than the threshold. For example, the operation processing unit 17 performs a predetermined processing such as release of an electric-key in the case where the image matching apparatus according to the present embodiment is applied to a vein pattern matching apparatus in a security field. [0044] As described above, the matching apparatus is provided with: the Hough transform unit 15 performing the Hough transform processing to the registered image AIM and the image to be matched RIM respectively, in detail, performing the image processing to the registered image AIM and the image to be matched RIM, by which the points in each image are transformed to the curved pattern PL and the linear components in each image are transformed to a plurality of the overlapped curved-patterns PL based on the distance p from the reference position O to the shortest point PO in the straight line L passing through the point in the image and the angle 0 between the straight line nO passing through the reference position O and the shortest point PO and the x-axis as the reference axis including the reference position O, and generating the transformed image SI621 and the transformed image S1622; and the judgment unit 164 performing the matching of the image to be matched RIM and the registered image AIM based on the degree of the overlap of the patterns in the transformed image SI621 and the transformed image SI622 generated by the Hough transform unit 15 and the matching or mismatching of the patterns in the transformed image S1621 and the transformed image S1622. So, the matching images including the featured linear components can be performed at a high accuracy. [0045] Namely, the Hough transform unit 15 generates the transformed images under consideration of the linear components (also, referred to a linear shapes) in the registered image AIM and the image to be matched RIM by performing the Hough transform processing, and the judgment unit 164 performs the matching of the transformed images based on the degree of the overlap of the curved patterns. So, the matching images including the featured linear components can be performed at a high accuracy. [0046] Further, if the extraction unit 163 is provided at a subsequent stage of the Hough transform unit 162, the extraction unit 163 extracts the region of which the degree of the overlap of the curved patterns PL in either the transformed image S1621 or the transformed image SI622 is equal to or greater than the threshold set in advance, namely, deletes point components functioning as noise components on the registered image AIM and the image to be matched RIM and extracts only the linear components, and generates the images S1631 and S1632, and the judgment unit 164 performs the matching of the image to be matched RIM and the registered image AIM based on the matching or mismatching of the patterns in the extracted region. Consequently, the matching processing further improved and at a higher accuracy can be performed. [0047] And, the judgment unit 164 is provided with: the similarity generation unit 1641 performing the comparison processing to a plurality of the different positional relationships in the transformed images, and generating the similarity as the correlation value based on the compared result; and the matching unit 1642 performing the matching of the image to be matched RIM and the registered image AIM based on the generated similarity Sim. Then, the judgment unit 164 generates the similarity as the correlation value by applying a simple calculation and performs the matching processing based on the similarity. Consequently, the matching processing can be performed at a high speed. [0048] In the present embodiment, though the position correction processing is performed at step ST101, but, it is not limited thereto. For example, in the case where the position correction processing is unnecessary, the Hough transform unit 15 may perform the Hough transform processing to the registered image AIM and the image to be matched RIM respectively. Therefore, a processing load is reduced and the matching processing can be performed at a high speed. [0049] In the present embodiment, though, at step ST103, the extraction unit 163 extracts the region of which the degree of the overlap of the curved patterns PL is equal to or greater than the threshold set in advance, but, it is not limited thereto. For example, the extraction unit 163 dose not perform the extraction processing, and the judgment unit 164 may perform the matching of the image to be matched RIM and the registered image AIM based on the degree of the overlap of the patterns in the image SI621 and the image SI622 and the matching or mismatching of the patterns therein. Therefore, the processing load is reduced and the matching processing can be performed at a high speed. [0050] FIG. 7 is a block diagram showing a position correction unit of the image matching apparatus of a second embodiment according to the present invention. The image matching apparatus la according to the present embodiment, in a functional block diagram illustrated as a hardware, has approximately the same configuration as the image matching apparatus 1 according to the first embodiment, and, for example, has the image input unit 11, the memory 12, the FFT processing unit 13, the coordinates transform unit 14, the Hough transform unit 15, the CPU 16 and the operation processing unit 17 as shown in FIG. 1. In the functional block illustrated as a software, the image matching apparatus la has approximately the same configuration as the image matching apparatus 1 according to the first embodiment, and the CPU 16 executes the program PRG to realize the function of the position correction unit 161, the Hough transform unit 162, the extraction unit 163, and the judgment unit 164. Components the same as those of the above embodiment are assigned the same notations (references) and explanations thereof are omitted. And only different points will be described. [0051] The different points are the followings: the position correction unit 161, as a position correction processing, generating the correlation value based on phase components which are results of the rotation angle correction processing of the registered image AIM and the image to be matched RIM or the enlargement ratio correction processing thereof, and the Fourier transform processing thereof, and performing the position correction processing to the registered image AIM and the image to be matched RIM based on the generated correlation value. In detail, the position correction unit 161 according to the present embodiment, as shown in FIG. 7, realizes functions of the magnification information and rotation information unit 21, the correction unit 22, the phase-only correlation unit 23, the shift information generation unit 24, and the correction unit 25 by the CPU 16 executing the program PRG and controlling the FFT processing unit 13 and the coordinates transform unit 14, for example. [0052] The magnification information ana rotation information unit 21 and the correction unit 22 perform a rotation angle correction processing or an enlargement ratio correction processing thereof to the registered image AIM and the image to be matched RIM. The magnification information and rotation information unit 21 generates a magnification information and/or a rotation information based on the registered image AIM and the image to be matched RIM, and outputs the same as a signal S21 to the correction unit 22. The magnification information includes an information indicating an enlargement and/or reduction ratio of the registered image AIM and the image to be matched RIM. The rotation information includes an information indicating a rotation angle of the registered image AIM and the image to be matched RIM. [0053] In detail, for example, the magnification information and rotation information unit 21 has a Fourier and Mellin transform unit 211, a phase-only correlation unit 212, and a magnification information and rotation information generation unit 213. The Fourier and Mellin transform unit 211 performs a Fourier and Mellin transform as mentioned later to the respective image information and outputs signals SA211 and SR211 indicating the respective transformed results to the phase-only correlation unit 212. [0054] In detail, the Fourier and Mellin transform unit 211 has Fourier transform units 21111 and 21112, logarithmic transform units 21121 and 21122, and logarithm-polar coordinates transform units 21131 and 21132. [0055] The Fourier transform unit 21111 performs Fourier transform as expressed by the following formula (3) to generate a Fourier image data Fl(u, v) when the registered image AIM is assumed as f1(m, n) in the case where the registered image AIM is N x N pixel image, for example, and outputs the same to the logarithmic transform unit 21121. The Fourier transform unit 21112 performs Fourier transform as expressed by the following formula (4) to generate a Fourier image data F2(u, v) when the image to be matched RIM is assumed as f2(m, n) in the case where the image to be matched RIM is N x N pixel image, for example, and outputs the same to the logarithmic transform unit 21122. [0056] The Fourier image data Fl(u, v) is formed by an amplitude spectrum A(u, v) and a phase spectrum ®(u, v) as expressed by the following formula (3). The Fourier image data F2(u, v) is formed by an amplitude spectrum B(u, v) and a phase spectrum1u, v) as expressed by the following formula (4). (Formula Removed) [0059] The logarithmic transform units 21121 and 21122 perform logarithmic processings to amplitude components of the Fourier image data Fl(u, v) and F2(u, v) generated by the Fourier transform units 21111 and 21112. The logarithmic processing to the amplitude components is carried out to emphasize a high frequency component including a detail featured information of the image data. [0060] In detail, the logarithmic transform unit 21121 performs the logarithmic processing to the amplitude components A(u, v) as expressed by the following formula (5) to generate A'(u, v), and outputs the same to logarithm-polar coordinates transform unit 21131. The logarithmic transform unit 21122 performs the logarithmic processing to the amplitude components B(u, v) as expressed by the following formula (6) to generate B'(u, v), and outputs the same to logarithm-polar coordinate transform unit 21132. [0061] [Formula 5] (Formula Removed) [0063] The logarithm-polar coordinates transform units 21131 and 21132 transform the signals output from the logarithmic transform units 21121 and 21122 to signals of a logarithm-polar coordinates system (for example, (log(r), 0). Generally, for example, when a position (x, y) is defined such as the following formulas (7) and (8) , (j, is equal to log(r) when R is equal to e^, so there is unanimously (log(r), 0) corresponding to any point (x, y). The logarithm-polar coordinates transform units 21131 and 21132 perform a coordinate transform processing based 0065] on the above nature. [0064] [Formula 7] [0066] In detail, the logarithm-polar coordinates transform units 21131 and 21132 define a set (r±, 0j) expressed by the following formula (9) and a function f(ri, 63) expressed by the following formula (10). [0067] (Formula Removed) [0068] (Formula Removed) [0069] The logarithm-polar coordinates transform units 21131 and 21132 perform the logarithm-polar coordinates transform to the image data A' (u, v) and B' (u, v) as expressed by the following formulas (11) and (12) by applying the set (r±, 0j) and the function f (ri, 6j) defined by the formulas (9) and (10) to generate 6j) and pB(ri, 0-j) , and output the same as the signal SA211 and the signal SR211 to the phase-only correlation unit 212. [0070] [formula 11] [0072] FIG. 8 is a view for explaining an operation of the Fourier and Mellin transform unit 211 shown in FIG. 7. The image f 1 (m, n) and the image f2 (m, n) , for example, include rectangle regions Wl and W2 having different predetermined angles with respect to x-axis and y-axis . In the Fourier and Mellin transform unit 211, for example as shown in FIG. 8, the Fourier transform processing is performed to the image f 1 (m, n) by the Fourier transform units 21111 to thereby generate a Fourier image data Fl(u, v) , and the image data pA(r, 0) is generated by the logarithmic transform unit 21121 and the logarithm-polar coordinates transform unit 21131. [0073] In the same way, the Fourier transform processing is performed to the image f2(m, n) by the Fourier transform units 21112 to thereby generate a Fourier image data F2(u, v), and the image data pB(r, 6) is generated by the logarithmic transform unit 21122 and the logarithm-polar coordinates transform unit 21132. [0074] As described above, the images f 1 (m, n) and f2(m, n) are transformed from Cartesian coordinates to the logarithm-polar coordinates system (also, referred to a Fourier and Mellin space) by the Fourier transform and the logarithm-polar coordinates transform. In the Fourier and Mellin space, there is a nature that the component is moved along a log-r axis based on a scaling of an image and moved along a 6-axis based on a rotation angle of the image. By applying the above nature, the scaling (magnification information) and a rotation angle of the images fl(m, n) and f2(m, n) may be obtained based on an amount of the shift along the log-r axis in the Fourier and Mellin space and an amount of the shift along the 6-axis. [0075] The phase-only correlation unit 212, for example, applies a phase only correlation method used in a phase-only filter (symmetric phase-only matched filter: SPOMF) to the signal SA211 and the signal SR211 respectively indicating a pattern data output from the Fourier and Mellin transfer 211 and obtains the amount of the parallel movement thereof. The phase-only correlation unit 212, as shown in FIG. 7, has Fourier transform units 2120 and 2121, a combination (synthesizing) unit 2122, a phase extraction unit 2123, and an inverse-Fourier transform unit 2124. [0076] The Fourier transform units 2120 and 2121 perform Fourier transform to the signals SA211 (pA(m, n)) and SR211 (pB(m, n)) output from the logarithm-polar coordinates transform units 21131 and 21132 by applying the following formulas (13) and (14). Here, X(u, v) and Y(u, v) indicate Fourier coefficients. The Fourier coefficient X(u, v) is formed by an amplitude spectrum C(u, v) and a phase spectrum 9(u, v) as expressed by the following formula (13). The Fourier coefficient Y(u, v) is formed by an amplitude spectrum D(u, v) and a phase spectrum (j)(u, v) as expressed by the following formula (14) . [0077] [0078] (Formula Removed) [0079] The combination unit 2122 combines X(u, v) and Y(u, v) generated by the Fourier transform units 2120 and 2121, and obtains the correlation. For example, the combination unit 2122 generates X(u, v)-Y*(u, v) and outputs the same to the phase extraction unit 2123. Here, 5f*(u, v) is defined as a complex conjugate of Y(u, v) . [0080] The phase extraction unit 2123 deletes an amplitude component based on a combined signal output from the combination unit 2122 and extracts a phase information. For example, the phase extraction unit 2123 extracts the phase component which is Z(u, v) = e3(8(u/ v)~ »