JP3706603B2 - Data feature extraction device and data collation device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a data collation technique for collating data such as image data.
[0002]
[Prior art]
In recent years, image data obtained by imaging has been widely used as an information processing target due to the remarkable development of hardware technology and software technology.
For example, an image database in which a considerable amount of image data is collected and managed collectively is used for various purposes such as medical, office work, academic research, monitoring, etc. Research and development of image recognition technology to determine whether a registered object or person is active is also underway.
[0003]
By the way, it is necessary to determine whether or not a specific object is reflected in certain image data, such as when searching for image data that includes a specific object or the like from an image database in which image data obtained by imaging is accumulated. In some cases, it is possible to use a method of comparing the certain image data with data representing the specific object or the like as a line drawing or a multi-tone image.
[0004]
As a conventional image data collation technique for comparing the two data, for example, there is a method using generalized Hough transform or high-speed generalized Hough transform that speeds up the conversion.
The outline of this method will be described below.
A person who performs a search draws a specific object or the like desired to be searched on paper as a line drawing, and inputs the line drawing to a computer as binary image data through an image scanner or the like. The computer generates the template data by performing high-speed generalized Hough transform on the image data of the line drawing, and extracts the contour by the spatial differential processing of the image based on the multi-tone image data to be searched in the image database. The search target binary image data is generated, the template data is voted for the search target binary image data using the high-speed generalized Hough transform, and the number of votes for a certain point in the image is equal to or greater than a threshold value. If there is a point, it is determined that an image corresponding to a specific object or the like is captured at that point. The generalized Hough transform and the high-speed generalized Hough transform are disclosed in Non-Patent Document 1, for example.
[0005]
As another image data collation technique, there is an image data collation technique according to the prior invention by the present inventor that collates both image data by generating specific collation data by wavelet decomposition of each image data. Yes (see Patent Document 1).
Each of these various image data matching techniques has its own advantages and disadvantages.
[0006]
[Non-Patent Document 1]
Kimura Akio et al., High-speed Generalized Hough Transform-Similar Transform Invariant Arbitrary Figure Detection Method, “The Institute of Electronics, Information and Communication Engineers Journal (D-2)”, April 1998, J81-D-2, No. 4, p. . 726-734
[0007]
[Patent Document 1]
JP 2001-283221 A
[0008]
[Problems to be solved by the invention]
In the present invention, even when two objects obtained by imaging or representing the same object, person, scenery, etc. are included in the relative rotation, both data are the same object. An object of the present invention is to provide a new data matching technique that can be determined to be related to the above, and that uses a method different from the conventional technique.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, a data feature extraction device according to the present invention is a data feature extraction device that generates feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space, A three-dimensional data acquisition unit that acquires three-dimensional data, a scanning unit that sequentially focuses on each position on the xy two-dimensional plane, and every time one position is focused on by the scanning unit, around the focused position An xyz three-dimensional coordinate space of a three-dimensional position having an x coordinate and a y coordinate in the local area among the three-dimensional positions indicated by the three-dimensional data for each of the four or more predetermined local areas on the xy two-dimensional plane. In the xyz three-dimensional coordinate space, the average positions obtained for all the local regions are almost on the same plane. It is determined whether or not it exists, and the information indicating the gradient of the plane with respect to the xy two-dimensional plane and the information indicating the focused position are associated with each other only when they exist on substantially the same plane. And feature information generating means for generating the information as one element of the feature information.
[0010]
A data feature extraction method according to the present invention is a data feature extraction method for generating feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space, and acquires three-dimensional data. A three-dimensional data acquisition step, a scanning step in which attention is paid sequentially to each position on the xy two-dimensional plane, and a predetermined number of four or more located around the focused position every time one position is noted in the scanning step For each of the local regions on the xy two-dimensional plane, an average position in the xyz three-dimensional coordinate space of a three-dimensional position having x and y coordinates in the local region among the three-dimensional positions indicated by the three-dimensional data is obtained. Whether each average position obtained for all the local regions is substantially on the same plane in the xyz three-dimensional coordinate space Only when they exist on substantially the same plane, information indicating the gradient of the plane with respect to the xy two-dimensional plane and information indicating the focused position are associated with 1 of the feature information. And a feature information generation step generated as an element.
[0011]
Here, being on substantially the same plane means being present at a distance of a predetermined value or less from a certain plane.
Thereby, the characteristic information about the three-dimensional data that can be used for the comparison between the three-dimensional data can be obtained. When the three-dimensional data is collated with the obtained feature information, the three-dimensional data having the relationship rotated around the straight line perpendicular to the xy two-dimensional plane has a similar tendency to some extent appropriately between the three-dimensional data representing the same object. It can be determined that the price is high.
[0012]
A data collating apparatus according to the present invention is a data collating apparatus that collates three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space, and obtains first three-dimensional data. A three-dimensional data acquisition unit, a first scanning unit that sequentially focuses on each position on the xy two-dimensional plane, and each time one position is focused by the first scanning unit, is located around the focused position. For each of the four or more predetermined number of local regions on the xy two-dimensional plane, among the three-dimensional positions indicated by the first three-dimensional data, the xyz three-dimensional coordinates of the three-dimensional position having the x coordinate and the y coordinate in the local region An average position in the space is obtained, and whether or not each of the average positions obtained for all the local regions exists on substantially the same plane in the xyz three-dimensional coordinate space. The vector V on the xy two-dimensional plane indicating the gradient of the plane with respect to the xy two-dimensional plane is specified only when the planes are substantially on the same plane, and the vector V and information indicating the focused position are obtained. First feature information generating means to be associated with, a unit vector S starting from one position Q in the xyz three-dimensional coordinate space and going in one direction, and associated with each vector V specified by the first feature information generating means For each piece of information, a vector PQ indicating a relative positional relationship of the position Q in the xyz three-dimensional coordinate space is defined with reference to the focused position relating to the information, and a plurality of sets of the vector V and the vector PQ and a unit vector S are shown. Verification data generation means for generating verification data, second 3D data acquisition means for acquiring second 3D data, xy 2D flat A second scanning unit that sequentially focuses on each position on the top, and every time one position is focused by the second scanning unit, on a predetermined number of four or more xy two-dimensional planes located around the focused position For each of the local regions, the average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x-coordinate and the y-coordinate in the local region among the three-dimensional positions indicated by the second three-dimensional data is determined. It is determined whether or not each average position obtained for the local region is on substantially the same plane in the xyz three-dimensional coordinate space, and the xy two-dimensional of the plane is only when the average position is on the same plane. Second feature information generating means for specifying a vector V ′ on an xy two-dimensional plane indicating a gradient with respect to the plane and associating the vector V ′ with information indicating the focused position By sequentially applying each vector V in the verification data to each vector V ′ specified by the second feature information generating means, the orientation relationship between the vector V ′ and the new unit vector S ′ is a vector. Each unit vector S ′ is defined so as to be the same as the orientation relationship between the vector V applied to V ′ and the unit vector S, and according to the distribution state in the xyz three-dimensional coordinate space of all the determined unit vectors S ′. The apparatus further comprises collating means for determining a similarity tendency between the first three-dimensional data and the second three-dimensional data.
[0013]
The data collating method according to the present invention is a data collating method for collating three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space, wherein the first three-dimensional data is acquired. A three-dimensional data acquisition step, a first scanning step in which attention is paid sequentially to each position on the xy two-dimensional plane, and every time one position is noticed in the first scanning step, it is located around the focused position. For each of the four or more predetermined number of local regions on the xy two-dimensional plane, among the three-dimensional positions indicated by the first three-dimensional data, the xyz three-dimensional coordinates of the three-dimensional position having the x coordinate and the y coordinate in the local region An average position in the space is obtained, and all the average positions obtained for all the local regions are on substantially the same plane in the xyz three-dimensional coordinate space. The vector V on the xy two-dimensional plane indicating the gradient of the plane with respect to the xy two-dimensional plane is specified only when the planes are substantially on the same plane, and the vector V and the focused position A first feature information generating step for associating information indicating the unit vector S, a unit vector S starting from one position Q in the xyz three-dimensional coordinate space and going in one direction, and each of the unit feature S specified by the first feature information generating step For each piece of information associated with the vector V, a vector PQ indicating the relative positional relationship of the position Q in the xyz three-dimensional coordinate space is defined with reference to the focused position relating to the information, and a plurality of sets of vector V and vector PQ and units A collation data generation step for generating collation data indicating the vector S, and second 3D data for acquiring second 3D data An acquisition step, a second scanning step for sequentially focusing on each position on the xy two-dimensional plane, and each time one position is focused on by the second scanning step, four or more located around the focused position For each of a predetermined number of local regions on the xy two-dimensional plane, an average of three-dimensional positions having x and y coordinates in the local region among the three-dimensional positions indicated by the second three-dimensional data in the xyz three-dimensional coordinate space And determining whether or not each average position obtained for all the local regions is on substantially the same plane in the xyz three-dimensional coordinate space. As long as the vector V ′ on the xy two-dimensional plane indicating the gradient of the plane with respect to the xy two-dimensional plane is specified, the vector V ′ and the focused position are indicated. A second feature information generation step for associating the information with each other, and by sequentially applying each vector V in the matching data to each vector V ′ specified in the second feature information generation step, Each unit vector S ′ is determined such that the azimuth relationship with the unit vector S ′ is the same as the azimuth relationship between the vector V and the unit vector S applied to the vector V ′, and xyz3 of all the determined unit vectors S ′. And a collating step for determining a similarity tendency between the first three-dimensional data and the second three-dimensional data according to a distribution state in a dimensional coordinate space.
[0014]
As a result, the three-dimensional data in a rotational relationship can be appropriately collated.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an image collating apparatus according to an embodiment of the present invention will be described with reference to FIGS.
<1. Configuration>
FIG. 1 is a functional block diagram of an image collating apparatus according to an embodiment of the present invention.
[0016]
The image collation apparatus 100 is an apparatus that is used in an image retrieval system that retrieves specific image data from an image database and has a function of collating two pieces of image data. Here, in the image search system, objects such as objects, people, and landscapes are handwritten from an image database that is a set of multi-gradation two-dimensional image data (hereinafter referred to as “searched data”) formed by imaging. A function of searching for data to be searched that is obtained by imaging the same object as the two-dimensional image data (hereinafter referred to as “search reference data”) of the line drawing obtained by digitizing the line drawing drawn in (1) using a scanner or the like. It is the system which has.
[0017]
The image matching device 100 is realized as one function of a computer. As shown in FIG. 1, a search reference data acquisition unit 101, a luminance level determination unit 102, a feature data generation unit 103, a range limitation unit 104, and verification data generation. A holding unit 105, a searched data acquisition unit 106, a line drawing unit 107, a luminance level determination unit 108, a feature data generation unit 109, a collation unit 110, and an evaluation value recording unit 111 are provided. The function of each unit constituting the image collating apparatus 100 is realized by the CPU executing a control program stored in the memory of the computer.
[0018]
Hereinafter, each part which comprises this image collation apparatus 100 is demonstrated.
<1-1. Search Criteria Data Acquisition Unit 101>
The search reference data acquisition unit 101 has a function of specifying search reference data, which is one line drawing data, as a processing target and transmitting it to the luminance level determination unit 102. This search reference data is 1 if the person who performs the search draws a line drawing on paper by hand and then there is a line (point) drawn for each pixel position by a scanner or the like, and if there is no line (point). The data obtained by giving a data value of 0 to binary two-dimensional image data is, for example, data of 512 pixels in the horizontal direction and 512 pixels in the vertical direction.
[0019]
<1-2. Luminance Level Determination Unit 102>
The brightness level determination unit 102 performs a predetermined brightness determination process based on the transmitted search reference data, determines a multi-tone brightness value for each pixel of the search reference data, for example, 256 multi-tone multi-tone It has a function of generating image data.
In the luminance determination process, each pixel position on the xy two-dimensional plane related to the search reference data is sequentially set as a first scanning point, and all pixel positions in a circular area having a predetermined constant radius r1 are centered on the first scanning point. The first luminance value is approximately inversely proportional to the number of all pixel positions where the binary data value is 1, that is, the number of all pixel positions corresponding to a line or point drawn in the circular area. This is a process of determining the luminance value for the scanning point.
[0020]
That is, in the luminance determination process, the number of all pixel positions where the data value is 1 in a circular area having a constant radius r1 centered on the first scanning point is k while moving the first scanning point on the search reference data. If the luminance value is determined by assuming that the arithmetic expression for obtaining the luminance value Z at the first scanning point is, for example, the following equation 1, the number of 256 gradations from 0 to 255 is obtained. This is processing for generating gradation image data.
[0021]
[Formula 1] Z = 255 / (k + 1)
Here, for example, the radius r1 of the circular region is 10 pixels or the like, and the size of the circular region is determined so that k can vary from 0 to a value of 255 or more. Is an integer value obtained by rounding the result of 255 / (k + 1).
[0022]
In practice, when the first scanning point is at the end of the search reference data, since the position in the circular region centering on the first scanning point protrudes from the search reference data region, for example, When the radius r1 of the circular area is 10 pixels, the horizontal coordinates 10 to 501 and the vertical coordinates 10 to 501 in the search reference data in which pixels exist at each coordinate of the horizontal coordinates 0 to 511 and the vertical coordinates 0 to 511 The luminance value at each first scanning point is obtained sequentially with each coordinate position as the first scanning point, and the luminance value is not obtained for pixel positions in the range of 10 pixels from each end.
[0023]
By scanning, the coordinates of the first scanning point change in the horizontal direction in the order of (10, 10), (11, 10), (12, 10), ..., (500, 10), (501, 10). Subsequently, the next row changes in the same manner as (10, 11), (11, 11), (12, 11),..., (500, 11), (501, 11). An xy two-dimensional plane related to image data such as search reference data is a two-dimensional plane represented by an x-axis and a y-axis, and all pixel positions constituting the image data are integer values in the x-coordinate and the y-coordinate. Shall be represented. It is assumed that each pixel has a position as a point, and the distance between two pixels adjacent in the horizontal direction or the vertical direction on the xy two-dimensional plane is 1.
[0024]
FIG. 2 is a diagram illustrating the relationship between the line drawing portion and the first scanning point in the search reference data.
This figure shows a circular area 203 centered on a first scanning point 202 at a certain moment during scanning, taking as an example search reference data 200 representing a line drawing 201 in which two triangles are connected.
[0025]
In this example, since the number k of pixel positions on the line drawing in the circular area is 16, the luminance value Z of the first scanning point 202 is Z = 255 / (16 + 1) = 15 from Equation 1.
For example, when a simple straight line is represented in the search reference data, when the first scanning point is on the straight line by the above scanning, the largest number of pixel positions are included in the circular area 203. The luminance of the first scanning point is the lowest, and the luminance of the first scanning point decreases to a certain extent as the first scanning point moves away from the straight line.
[0026]
<1-3. Feature Data Generation Unit 103>
The feature data generation unit 103 has a function of performing predetermined feature extraction processing based on the multi-tone image data generated by the brightness level determination unit 102 and storing feature data that is information indicating the feature of the image.
In the feature extraction process, each position on the two-dimensional image space related to the multi-tone image data generated by the luminance level determination unit 102 is sequentially set as the second scanning point, and a predetermined constant radius centered on the second scanning point. Only when the average luminance values for the first quadrant to the fourth quadrant obtained by dividing the circular area r2 into horizontal and vertical crosses are in a planar relationship, around the second scanning point. A vector V, which is a spatial gradient of each average luminance value, and a three-dimensional coordinate P obtained by adding the luminance value for the second scanning point as another dimension data to the two-dimensional pixel position for the second scanning point. This is a process of saving a pair. Hereinafter, when distinguishing individual vectors V, V (1), V (2),..., Etc., and when distinguishing individual three-dimensional coordinates P, P (1), P (2),. .. etc., typically V (i), P (i).
[0027]
FIG. 3 is a diagram illustrating an example of the relationship between the second scanning point used in the feature extraction process and each region of the first quadrant to the fourth quadrant. In the figure, the radius r2 is 10 pixels.
Here, the feature extraction processing will be described in more detail with reference to FIG.
The multi-gradation image data generated by the brightness level determination unit 102 is data in which a luminance value is determined for each pixel position in the xy two-dimensional image space. In the feature extraction process, xy2 related to the multi-gradation image data is obtained. A portion between a negative direction of x and a negative direction of y with reference to the second scanning point in a circular region having a radius r2 centered on the second scanning point while moving the second scanning point on the dimensional image space The average value z1 of each luminance value corresponding to (region R1 in FIG. 3) and the average value of each luminance value corresponding to a portion (region R2) between the positive direction of x and the negative direction of y z2, an average value z3 of each luminance value corresponding to a portion between the negative direction of x and the positive direction of y (region R3), and a portion between the positive direction of x and the positive direction of y (region R4) ) To obtain an average value z4 of each luminance value, and using these z1 to z4, Each average luminance value of the peripheral scanning point to calculate the vector V (i) is the spatial gradient of luminance values at the periphery of determining whether or not and the second scanning point in the plane relationship.
[0028]
The second scanning point is when the multi-tone image data is composed of 502 pixels in the horizontal direction (x coordinate is 10 to 501) and 502 pixels in the vertical direction (y coordinate is 10 to 501) on the xy two-dimensional plane. , X coordinate and y coordinate move in the range of 20-491 respectively. That is, the coordinates of the second scanning point sequentially change in the horizontal direction as (20, 20), (21, 20), (22, 20), ..., (490, 20), (491, 20). Subsequently, (20, 21), (21, 21), (22, 21),..., (490, 21), (491, 21) are similarly changed for the next line.
[0029]
Whether each average luminance value of each part (region R1 to R4) around the second scanning point has a planar relationship is determined based on whether the relationship represented by the following inequality is satisfied. Done.
[Expression 2] | z1-z2-z3 + z4 | ≦ C1
In Equation 2, C1 is a constant, and a relatively small value is determined in advance so that the relationship of the inequality of Equation 2 is satisfied when z1−z2−z3 + z4 is close to 0. In the case where the maximum value of the luminance value is 255, C1 may be a single value of about 0 to 5, for example.
[0030]
When the relationship represented by the inequality of Equation 2 is satisfied, the above-described regions R1 to R4 are further rotated about the second scanning point by 30 degrees in the circumferential direction, and the region R1 ′ of the new range is obtained. Similarly, for R4 ′, the average values z1 to z4 of the luminance values in the respective regions are obtained to determine whether or not the relationship represented by the inequality of Equation 2 is satisfied, and when the relationship is still satisfied, the same is performed again. The regions R1 ′ to R4 ′ are rotated in the direction, z1 to z4 are similarly obtained for the new ranges R1 ″ to R4 ″, and it is determined whether or not the relationship represented by the inequality of Equation 2 is satisfied. Is satisfied, it is determined that the average luminance values of the regions around the second scanning point are in a planar relationship.
[0031]
Hereinafter, the vector V (i), which is the spatial gradient of the luminance value around the second scanning point, is performed only when the average luminance values of the respective regions around the second scanning point are in a planar relationship. A calculation procedure will be described. When z1−z2−z3 + z4 is not 0, z4 is corrected so that z1−z2−z3 + z4 becomes 0, and the vector V (i) is calculated by the following procedure after correction.
[0032]
First, for each of the regions R1 to R4, an average luminance value in the region is determined as a coordinate (z coordinate) of another dimension at an average position (x coordinate and y coordinate of the center of gravity) of the region on the xy two-dimensional plane. As a result, four three-dimensional coordinates are obtained.
The x coordinate of the average position of the region R1 is x1, the y coordinate is y1, the x coordinate of the average position of the region R2 is x2, the x coordinate is x2, and the x coordinate of the average position of the region R3 is x3, y coordinate. Is y3, the x coordinate of the average position of the region R4 is x4, and the y coordinate is y4, the above four three-dimensional coordinates are A1 (x1, y1, z1), A2 (x2, y2, z2), A3 (x3, y3, z3) and A4 (x4, y4, z4).
[0033]
By substituting the respective three-dimensional coordinate values of A1 to A4 into the general formula of the plane, the value of each coefficient of the general formula of the plane is calculated, and the vector V (i) is calculated by the formulas 4 and 5. Components Vx and Vy are obtained.
[Formula 3] a * x + b * y + c * z = d
In Equation 3, x, y, and z are variables, a, b, c, and d are coefficients, and “·” is a multiplication operator.
[0034]
[Formula 4] Vx = −a / c
[Formula 5] Vy = −b / c
Equation 4 represents the slope of a straight line intersecting the xz plane, that is, the displacement Δz / Δx in the z direction with respect to the displacement in the x direction, in which the plane in the three-dimensional coordinate space according to Equation 3 in which the coefficient value is determined. The plane in the three-dimensional coordinate space according to the equation 3 with a fixed coefficient value represents the inclination of the straight line intersecting the yz plane, that is, the displacement Δz / Δy in the z direction with respect to the displacement in the y direction.
[0035]
In other words, Vx represents the gradient of the luminance value in the x direction, and Vy represents the gradient of the luminance value in the y direction. Therefore, it can be said that the vector V (i) represents the spatial gradient of the luminance value.
Assuming that the value of the three-dimensional coordinate P (i) to be stored in a pair with V (i) = (Vx, Vy) thus obtained is (Px, Py, Pz), Px is converted into multi-tone image data. The x-coordinate of the second scanning point in the xy two-dimensional plane is such that Py is the y-coordinate of the second scanning point in the xy two-dimensional plane relating to the multi-tone image data, and Pz is a value obtained by the following equation (6).
[0036]
[Formula 6] Pz = (z1 + z2 + z3 + z4) / 4
The feature data generation unit 103 records and stores a plurality of sets of V (i) and P (i) obtained by such a procedure in a memory. However, as an exception, for V (i) where both Vx and Vy which are components of V (i) are approximately 0, that is, V (i) which is less than a predetermined threshold, V (i) And P (i) are not saved. For example, if the average luminance values in the regions R1 to R4 around the second scanning point are all the same value, V (i) and P (i) for the second scanning point are not stored.
[0037]
<1-4. Range limiting unit 104>
The range limiting unit 104 has a function of specifying a range in which the image to be collated exists in the search reference data. For example, a search reference having 512 pixels in the horizontal and vertical directions in response to input from a searcher Identify a rectangular area in the data. The rectangular area is defined as, for example, a range where the x coordinate is 30 to 480 and the y coordinate is 30 to 480.
[0038]
<1-5. Verification Data Generation Holding Unit 105>
The collation data generation / holding unit 105 determines a point Q and a vector S indicating a certain direction on the xyz three-dimensional space, and the vector V (i) and the three-dimensional coordinate P ( Among the pairs with i), those in which the x and y coordinates of P (i) are included in the range specified by the range limiting unit 104 are extracted, and the extracted V (i) and P (i) For each group, PQ (i) indicating the position of Q in relative coordinates from P (i) is obtained, and V (i), PQ (i), and one S are stored in a memory or the like as collation data. It has a function.
[0039]
Number of combinations of V (i) and P (i) stored by the feature data generation unit 103 that include the x and y coordinates of P (i) within the range specified by the range limiting unit 104 Is n, the collation data generation and holding unit 105 stores n vectors V (i), n three-dimensional relative coordinates PQ (i), and one vector S as collation data. It will be.
[0040]
For example, if P (1) and P (2) are within the range specified by the range limiting unit 104, the relative coordinate PQ (1) is associated with the vector V (1), and similarly the relative coordinate PQ (2) is stored in association with the vector V (2). The vectors V (i), PQ (i), and S are all data composed of an x-direction component, a y-direction component, and a z-direction component in the xyz three-dimensional coordinate space. A vector V (i) having a starting point at the point P and a vector S having a starting point at the point Q are expressed by relative positions of relative starting points and relative directions of the vectors by V (i), PQ (i) and S. , That is, a relative orientation relationship.
[0041]
<1-6. Searched data acquisition unit 106>
The to-be-searched data acquisition unit 106 sequentially extracts one multi-tone image data from an image database in which a large number of to-be-searched multi-tone image data is stored, and specifies it as a target to be collated with the search reference data. And has a function of transmitting to the line drawing unit 107.
[0042]
This searched data is multi-gradation image data generated by a digital camera or the like, and is data of, for example, 256 gradations, 512 pixels in the horizontal direction and 512 pixels in the vertical direction.
<1-7. Line Drawing Unit 107>
The line drawing unit 107 performs a spatial differentiation process for obtaining the gradient of the luminance value in the two-dimensional image plane with respect to the multi-tone image data that is one search target data transmitted from the search target data acquisition unit 106 and the result thereof. 2 is binarized, and as a result, binary image data corresponding to the line drawing is generated and transmitted to the luminance level determination unit 108. In other words, the continuous process from the spatial differentiation process to the binarization process is a process in which a portion where the spatial gradient of the luminance change of the image is steep, that is, a so-called edge portion is set to 1 and the other portions are set to 0.
[0043]
<1-8. Luminance Level Determination Unit 108>
The luminance level determination unit 108 receives the binary image data generated based on the searched data from the line drawing unit 107, performs a predetermined luminance determination process, and performs multi-gradation luminance for each pixel in the binary image data. For example, it has a function of generating multi-gradation image data of, for example, 256 gradations by determining a value. The luminance determination process for generating multi-tone image data based on the binary image data is basically the same as the luminance determination process performed by the luminance level determination unit 102 described above, that is, the image data to be processed is different. Is the same.
[0044]
<1-9. Feature Data Generation Unit 109>
The feature data generation unit 109 has a function of performing predetermined feature extraction processing based on the multi-gradation image data generated by the luminance level determination unit 108 and storing feature data indicating image features. This feature extraction process is basically the same as the feature extraction process performed by the above-described feature data generation unit 103, that is, the same except that the image data to be processed is different. If completely the same image data is input to the feature data generation units 103 and 109, the feature data generation units 103 and 109 both store feature data having the same contents.
[0045]
The feature data stored by the feature data generation unit 109 is a plurality of sets of vectors V and three-dimensional coordinates P obtained based on the multi-tone image data generated by the luminance level determination unit 108. Hereinafter, the vector V and the three-dimensional coordinate P stored by the feature data generation unit 109 are expressed in the form of a vector V ′ (j) and a three-dimensional coordinate P ′ (j).
[0046]
<1-10. Verification unit 110>
The collation unit 110 is generated by the feature data generation unit 109 with n sets of vectors V (i), relative coordinates PQ (i) and one S that are the collation data generated by the collation data generation and holding unit 105. Using the m sets of vectors V ′ (j) and the three-dimensional coordinates P ′ (j), the image shown in the search reference data in a certain range approximates the image shown in the searched data. It has a function of performing a collation process for determining whether it is a thing.
[0047]
As this collation process, the collation unit 110 first applies all V (i) in the collation data in order for each vector V ′ (j), and is indicated by V (i), PQ (i), and S. Size 1 so that the azimuth relationship between V (i) and S in the three-dimensional coordinate space and the azimuth relationship between V ′ (j) and S ′ (j, i) in the three-dimensional coordinate space are the same. Unit vector S ′ (j, i). Here, S ′ (i, j) is a unit determined by applying the vector V (i) to the vector V ′ (j) so as to maintain the same orientation relationship as the vector S with respect to the vector V (i). Meaning vector S ′, if there are n V (i) as matching data and m vectors V ′ (j) stored by the feature data generation unit 109, the unit vector S in the matching process '(I, j) is determined to be n · m.
[0048]
If the angle between the direction of vector V (i) and the direction of vector PQ (i) from point P to point Q is α, α is V (i) and PQ (i If the angle between the direction of the vector V (i) and the direction of the vector S is β, β can be obtained from V (i) and S. Further, the distance γ between the point P and the point Q can be obtained from the vector PQ (i).
[0049]
Therefore, to determine the unit vector S ′ (i, j) by applying the vector V ′ (j) to the vector V ′ (j), the starting point of the unit vector S ′ (hereinafter referred to as “point Q ′”) is obtained. A vector P′Q ′ that is at a distance γ from the position of the point P ′ (j) that is the starting point of the vector V ′ (j) and that travels from the point P ′ to the point Q ′ is the vector V ′ (j). And the unit vector S ′ is determined to be a vector directed to the angle β with respect to the vector V ′ (j).
[0050]
Next, the xyz three-dimensional coordinate space is segmented so as to be a three-dimensional array of unit cubes of a predetermined size such as one side having a length of 5, and each unit cube is positioned in the unit cube. Processing for obtaining the vector addition result of all the vectors S ′ (i, j), that is, processing for synthesizing the vectors is performed, and finally, the maximum value Smax of the combined vector obtained as the vector addition result and the vector S ′ The number Snum of the vectors S ′ (i, j) in the unit cube containing the most (i, j) is evaluated together with information for identifying the searched data acquired by the searched data acquiring unit 106. If the maximum value Smax of the vector size stored in the value recording unit 111 and obtained as a vector addition result is equal to or greater than a predetermined threshold, the image in the searched data and the search reference data The image performing a process of determining and tend similar.
[0051]
The collation unit 110 displays the determination result via, for example, a display device. The display contents of the determination result are, for example, the search target data itself, whether it is similar or dissimilar, and the value of Smax / n.
If the image data obtained as a result of drawing the search target data completely matches the search reference data, the maximum vector size Smax obtained as a result of the vector addition described above is the size of the unit cube. Theoretically, at least n is the size of 1 pixel × 1 pixel × 1 pixel. Therefore, in consideration of the influence of the pixel positions being discrete in the digital image, in practice, for example, 0.6 or 0.7 times n may be set as the predetermined threshold.
[0052]
<1-11. Evaluation Value Recording Unit 111>
The evaluation value recording unit 111, as a pair with the information for identifying the data to be searched by the collating unit 110, the maximum value Smax of the vector size obtained as the vector addition result and the vector S ′ (i, This is a storage area such as a memory in which the maximum density of j), that is, the maximum number Snum of the number in the unit cube is stored.
[0053]
The evaluation value recording unit 111 includes a line drawing unit 107, a luminance level determination unit 108, and a feature data generation unit 109 each time one piece of image data is acquired from the image database by the search data acquisition unit 106. And by the operation of the collation unit 110, information for identifying a set of data to be searched and the maximum value Smax are additionally recorded. Therefore, if the contents of the evaluation value recording unit 111 are used, the search target data identified corresponding to the largest recorded maximum value Smax is indicated in the search reference data most in the image database. It is possible to evaluate that the data to be searched has a high tendency to be similar to the image. For example, it can be evaluated that the similarity tendency is higher as the number of Snums is larger.
<2. Operation>
Hereinafter, the operation of the image collation apparatus 100 having the above-described configuration will be described.
[0054]
The image collation apparatus 100 can be broadly divided into a first process for acquiring search reference data that is binary image data of input line drawings and generating collation data, and sequentially specifying data to be searched from the image database. Then, by using the specified data to be searched and the data for collation, the second processing for collating the search reference data with the data to be searched is performed.
[0055]
FIG. 4 is a flowchart showing a first process for acquiring the search reference data and generating the matching data.
As shown in the figure, first, the search reference data acquisition unit 101 acquires search reference data and transmits it to the luminance level determination unit 102 (step S11). The brightness level determination unit 102 generates multi-gradation image data by determining the brightness value for each pixel position by moving the first scanning point on the search reference data (step S12). As a result, multi-tone image data having a high luminance value is generated for pixel positions around the low density of lines and dots indicated by the search reference data.
[0056]
Subsequently, the feature data generation unit 103 moves the second scanning point on the xy two-dimensional plane related to the multi-tone image data generated by the luminance level determination unit 102, and the periphery around the second scanning point. Only when each average luminance value of each of the four divided regions is in a planar relationship, a vector V (i) indicating a spatial gradient of the average luminance value is obtained as xy of the second scanning point. The xyz three-dimensional coordinate P (i) in which the coordinate and the average luminance value of the entire surrounding four regions are set to the z coordinate are stored in a memory or the like (step S13).
[0057]
Subsequently, the verification data generation holding unit 105 receives the specification of the range by the range limiting unit 104, determines a point Q at an arbitrary three-dimensional position, determines a vector S in an arbitrary direction, and falls within the specified range. Relative coordinates PQ indicating vectors from P (i) to point Q for all vectors V (i) including three-dimensional coordinates P (i) and all P (i) located within the specified range (I) and one vector S are stored in a memory or the like as collation data (step S14).
[0058]
FIG. 5 is a diagram illustrating V (i) and P (i) generated by the feature data generation unit 103 and the points Q and vectors S determined by the collation data generation / holding unit 105.
In the figure, for convenience of explanation, the number of V (i) and P (i) is only three, and the figure shows a mapping on the xy two-dimensional plane for each element.
[0059]
FIG. 6 is a flowchart showing a second process for identifying data to be searched and using the matching data to match the search reference data with the data to be searched.
As shown in the figure, first, the searched data acquisition unit 106 acquires searched data that is one multi-tone image data from the image database and transmits it to the line drawing unit 107 (step S21). Then, binary image data is generated by binarization that extracts a so-called edge portion of the image represented in the search target data, and is transmitted to the luminance level determination unit 108 (step S22).
[0060]
The luminance level determination unit 108 that has received the binary image data determines the luminance for each pixel position of the binary image data and generates new multi-gradation image data in the same manner as in step S12 described above (step S12). S23).
Subsequently, the feature data generation unit 109 moves the second scanning point on the xy two-dimensional plane related to the multi-gradation image data generated by the luminance level determination unit 108, and surrounds the second scanning point as a center. Each average luminance value of the four divided regions is obtained, and only when each average luminance value of each region is in a planar relationship, a vector V ′ (j) indicating a spatial gradient of the average luminance value is calculated at the second scanning point. The xyz coordinates and the average luminance value of the entire four surrounding areas are stored in a memory or the like together with the xyz three-dimensional coordinates P ′ (j) in which the z coordinates are used (step S24).
[0061]
Subsequently, the collation unit 110 uses P ′ (j) and V ′ (i) indicating the features related to the search target image stored by the feature data generation unit 109 and the collation data stored in the first process. Referring to each P ′ (j), when V (i) which is the matching data is placed on the position of P ′ (j) in the xyz three-dimensional space so as to match V ′ (j), A unit vector S ′ is arranged while maintaining the same orientation relationship from the position of V (i) to S (step S25).
[0062]
FIG. 7 shows V (i), PQ (i) and PQ (i) held by the collation data generation holding unit 105 for each set of V ′ (j) and P ′ (j) generated by the feature data generation unit 109. It is a figure which illustrates a mode that unit relationship S '(i, j) was defined by applying the relationship of S.
In the figure, for convenience of explanation, the number of V ′ (j), P ′ (j), and V (i) is only three, and the mapping on the xy two-dimensional plane is shown for each element. ing. In the figure, the notation “S (V ′ (1) ← V (2))” indicates that the unit vector S ′ is arranged as a result of fitting V (2) to V ′ (1). Yes.
[0063]
For example, when V (2) is applied to V ′ (1) placed at the position indicated by P ′ (1) extracted based on the searched data, the vector PQ (2) from P ′ (1) ), And the angle between the direction from P ′ (1) to Q ′ with respect to the vector V ′ (1) as a reference is V (2) as a reference. A point Q ′ is determined so as to be the same as the angle formed by the vector of PQ (2), and the angle formed by V ′ (1) and the unit vector S ′ is V (2) at the position of the point Q ′. A unit vector S ′ that is the same as the angle formed by the direction of the vector S with reference to is arranged.
[0064]
After arrangement of the unit vectors S ′, the collation unit 110 calculates Smax and Snum described above, determines whether or not they are similar based on Smax, displays the determination result on the display device, and further Smax and Snum are recorded in the evaluation value recording unit 111 (step S26).
The image matching apparatus 100 sequentially determines data to be searched from the image database, and repeats the procedures of steps S21 to S26 until it is determined that they are similar (step S26).
[0065]
By the first process and the second process described above, image data including those that match or are similar to the object drawn as a line drawing by the person performing the search among the image data in the image database is detected. It will be.
Note that in the feature data generation units 103 and 109, the relationship between the quadrant regions R1 to R4 around the second scanning point on the xy two-dimensional plane related to the image data, that is, the centroid of each region is positioned at the four vertices of the square. For each of the regions R1 to R4 (see FIG. 3), the coordinates of the center of gravity in the xyz three-dimensional space obtained by using the average luminance value in the region as the z coordinate of the center of gravity of the region are approximately on the same plane. Only when it exists, the change of the peripheral luminance value at the second scanning point with respect to the xy coordinate space is stored as the vector V (i) or the vector V ′ (j) and treated as indicating the characteristics of the image data. ing. In addition, only when the regions R1 to R4 are rotated by 30 degrees and 60 degrees around the second scanning point, and the coordinates in the xyz three-dimensional space of the center of gravity of each region obtained in the same manner are approximately on the same plane. The vector V (i) or vector V ′ (j) is stored.
[0066]
That is, even if the image is rotated, the feature data generation unit aims to make the mutual relationship between the feature data in the feature data group collected as a result of the scanning by the second scanning point unchanged. Only when the spatial gradient of the brightness in the peripheral region of the two scanning points is uniform, the spatial gradient of the luminance at the second scanning point obtained using the luminance at each position in the peripheral region of the second scanning point. Are stored as feature data that can be used as a basis for verification.
[0067]
Therefore, according to the image collating apparatus 100, the objects represented in the two image data to be collated are rotated even if they are in a relationship of being rotated around a straight line perpendicular to the image plane. It is possible to determine whether or not they are correct and similar, as in the case where there is no such information.
<3. Supplement>
As described above, the image collating apparatus according to one embodiment of the data collating apparatus and the data feature extracting apparatus according to the present invention has been described. However, the image collating apparatus can be partially modified, and the present invention is described above. It goes without saying that the present invention is not limited to the embodiment of the image collating apparatus. That is,
(1) The image collation apparatus can collate not only the line drawing image data and the multi-tone image data but also the two multi-tone image data and the two line drawing image data. You may do it. For example, when all the image data to be collated is line drawing image data, the line drawing by the line drawing unit 107 is omitted, and when all the image data to be collated are multi-tone image data. The line drawing similar to the line drawing unit 107 may be performed between the search reference data acquisition unit 101 and the luminance level determination unit 102.
[0068]
Also, in the case of collating two multi-tone image data, the processing by the line drawing unit and the brightness level determining unit is omitted, and the two multi-tone image data are directly processed by each feature data generating unit. You may comprise an image collation apparatus so that it may become object.
(2) The position on the xy two-dimensional plane where the first scanning point or the second scanning point is positioned in the scanning by the luminance level determination unit and the feature data generation unit of the image matching device is not necessarily the point when the pixel is represented by a point. There is no need to be on top. For example, the first scanning point or the second scanning point may change an intermediate point between the pixel point and the pixel point. Further, in order to determine the luminance at the first scanning point, binary data for pixels in the range of the circular area centered on the first scanning point is used, and the spatial gradient of the luminance at the second scanning point is obtained. For this reason, the luminance values for the pixels within the range of the circular area centered on the second scanning point are used. However, each circular area may be changed to an area of another shape, You may vary the size.
(3) In the luminance determination process by the two luminance level determination units, Equation 1 is used to determine the luminance value Z of the first scanning point from the number k of the pixel positions in the circular area centered on the first scanning point. Although used, there are more lines or points around the first scanning point, that is, the number of pixel positions where the luminance value Z becomes lower as the number of pixel positions where the value of binary data is 1 is larger or the number of pixel positions is larger. If the calculation is such that the luminance value Z increases as the number increases, the luminance value Z may be determined by a calculation other than Equation 1. That is, the luminance determination processing is performed for each position in the local area on the xy two-dimensional plane according to whether the density of pixel points having a binary data value of 1 such as a line drawing is high or low. This is a process for determining a dimension attribute value (luminance value).
(4) In the feature data generation unit, the regions R1 to R4 each have a range that includes a plurality of pixel points, but each region R1 to R4 may have a range that includes only one pixel point. Note that the feature data generation unit calculates the average position of all the points in the range of each of the regions R1 to R4 and the average value of the luminance of all the points, but each region R1 to R4 has one pixel point. In the case where only the area R1 to R4 is included, the average position of each of the areas R1 to R4 is the position of one pixel included in each area, and the average value of the luminance for each of the areas R1 to R4 is It is obtained by the luminance itself of the one pixel included in the region.
(5) In the feature data generation unit, in order to determine whether or not the luminance distribution around the second scanning point is planar, and to calculate the spatial gradient of the luminance around the second scanning point, the second scanning point A region R1 to R4 obtained by dividing a circular region into four in the periphery of the region, and in order to determine whether it is particularly planar, the regions R1 to R4 are rotated twice by 30 degrees and a determination based on Equation 2 is performed; However, whether or not the spatial distribution of luminance around the second scanning point is planar from the average position and average luminance value in each of four or more local regions around the second scanning point. If it is a method of calculating the spatial gradient of the luminance at the second scanning point as the angle between the plane and the xy two-dimensional plane if it is planar, the number of local regions is not necessarily four. The position of the local region is not limited, and the region R1 It is not limited to R4. Further, the rotation by 30 degrees by 2 degrees is only one method for ensuring that the luminance distribution around the second scanning point is sufficiently flat. For example, the second distribution on the xy two-dimensional plane is performed. Whether each pixel located at each vertex of a small regular polygon centered on the scanning point is substantially in the same plane in the xyz three-dimensional coordinate space, where the luminance is the z coordinate It may be determined whether or not the luminance distribution around the second scanning point is sufficiently planar depending on whether or not the second scanning point is present. It should be noted that only the points where the peripheral luminance distribution is sufficiently planar are treated as the information indicating the characteristics of the image, such as the spatial gradient of the luminances added to the points, so that the information is a plurality of objects to be collated. Thus, it is difficult to be affected by the deviation of the sampling direction or the processing of rotating the image in digitizing the image.
(6) The range limiting unit does not necessarily have to receive input from a person who performs a search, and may store a rectangular region in a fixed manner in advance. In addition, the range limiting unit displays an image captured by the search reference data acquisition unit via a display device or the like, and allows a user to perform a search to specify a rectangle by dragging the mouse or the like. The image range to be collated may be specified based on this.
(7) In the feature data generation unit, each component of the vector V (i) indicating the spatial gradient of luminance is calculated by Equation 4 and Equation 5, but the change in luminance and the position on the xy two-dimensional plane are calculated. Data calculated by other operations that can represent the relationship with changes may be handled in the same manner as the vector V (i). For example, when the average luminance values of the regions R1, R2, R3, and R4 are z1, z2, z3, and z4 in this order, the x component of the vector V (i) is obtained by the following equation (7): Vx The y component of (i) may be determined to be Vy obtained by the following equation (8).
[0069]
[Formula 7] Vx = z1-z2 + z3-z4
[Equation 8] Vy = z1 + z2-z3-z4
Further, a vector V (i) obtained by multiplying both its components Vx and Vy by a constant so that the magnitude of the vector V (i) is 1, is defined as a new vector V (i). (I) may be normalized.
(8) The collation data generation / holding unit generates and stores collation data that includes a plurality of sets of the vector V (i) and the vector PQ (i) and the vector S. Information indicating the azimuthal relationship between the direction and the position between the vector V (i) and the vector S is sufficient. For example, the matching data can be obtained from the direction of the vector V (i) from the vector PQ ( The angle in the direction of i), the magnitude of the vector PQ (i), and the angle from the direction of the vector V (i) to the direction of the vector S may be used.
[0070]
Note that the fitting of the vector V (i) to the vector V ′ (j) may be performed only when the magnitudes of both vectors are substantially the same. In this case, the verification data needs to include information indicating the magnitude of the vector V (i), and the luminance from the maximum to the minimum luminance between the two image data to be verified Before the collation, a method such as obtaining the vector V (i) by operations such as Equation 4 and Equation 5 so that the influence of the luminance range does not appear on the vector V (i) so as not to be affected by the difference in range. It is necessary to adjust the luminance range and use a method for obtaining the vector V (i) by the formulas 7 and 8 in the feature data generation unit.
(9) PQ (i) stored in the collation data generation / holding unit is all set so that the size of PQ (i) having the maximum size among all PQ (i) is 1. PQ (i) is defined as PQ (i), which is obtained by uniformly multiplying the size of all the components (x component, y component and z component) of PQ (i) of PQ (i). It is good also as making and storing. Of all the PQ (i), the maximum size of PQ (i) may be set to 1, and the average value of all the PQ (i) sizes may be set to 1. As a result of these normalizations, when the two image data to be collated indicate the same object and are in a relationship of enlargement / reduction, it can be determined that they are the same or similar.
(10) The collation unit of the image collation apparatus performs collation processing such as fitting, using all the data used for collation as xy two-dimensional data, that is, ignoring the z component in the xyz three-dimensional data and including the x component and the y component. Also good. In this case, a plurality of unit vectors S ′ are arranged on the xy two-dimensional plane, and both of the object to be collated are determined on the xy two-dimensional plane according to the distribution state of the starting points of the plurality of unit vectors S ′ that are directed in the same direction. The similarity tendency of images can be evaluated. Note that it can be determined that the images are more similar as the number of origins in the local range where the origins of the unit vectors S ′ directed in approximately the same direction on the xy two-dimensional plane are concentrated the most.
(11) The collation unit calculates the size of an angle formed by the direction of the unit vector S ′ in the partial region where the addition result of the unit vector S ′ is the maximum viewed from the direction of the vector S, and the calculation result It is also possible to output to the display device or the like how much the images to be collated are in an angular relationship. Further, the collation unit records the maximum value Smax of the magnitude of the combined vector, which is the addition result of the vector S ′, and the maximum number Snum of the vectors S ′ in the unit cube in the evaluation value recording unit, as well as a display device or the like. May be output.
(12) The collation unit divides the xyz three-dimensional coordinate space so as to be a three-dimensional array of unit cubes of a predetermined size, for example, each side is 5 in length, and each unit cube has its unit cube A process of obtaining the vector addition result of all the vectors S ′ (i, j) positioned at the position, that is, a process of synthesizing the vectors is performed. Finally, the maximum value Smax of the combined vector obtained as the vector addition result is And the number Snum of the vectors S ′ (i, j) in the unit cube containing the most vectors S ′ (i, j) to identify the searched data acquired by the searched data acquisition unit 106. However, the unit cube here is not necessarily limited to a cube, and a predetermined size is used instead of a cube. May be the use of a sphere having a radius of no problem even be arranged such that each sphere overlap partially when arranged as a three-dimensional array in the same manner that each sphere and cube.
(13) In addition to the form of the above-described image collation device, the data collation technique according to the present invention can be applied to collation of 3D data aggregates, and can be applied to 3D map data or 3D measured object shapes. The present invention can also be applied to a use in which similarity is determined by comparing such data with other similar data.
[0071]
If the two 3D data to be collated are in a rotational relationship with respect to a certain straight line, if the straight line cannot be predicted, a plane perpendicular to the straight line can be assumed in advance. By treating the plane perpendicular to the straight line as the xy two-dimensional image plane in the case of the image matching shown as the embodiment, the similarity can be appropriately determined. For example, 3D data A indicating the shape of a single building and 3D data B indicating the shape and positional relationship of a plurality of buildings in a section where a plurality of buildings including the building are built. If it is necessary to collate and determine whether or not the same or similar building as the 3D data A is represented in the 3D data B, the surface of the land is treated as an xy 2D plane, By performing the same processing as the processing performed by the feature data generation unit of the image matching device using the height that is the attribute value of the position as the z coordinate, the determination can be performed in the same manner as the image matching device. become.
(14) A program for causing a computer to execute each process (see FIGS. 4 and 6) in the image collating apparatus can be recorded on a recording medium or distributed and distributed via various communication paths. Examples of such a recording medium include an IC card, an optical disk, a flexible disk, and a ROM. The distributed and distributed program is used by being stored in a memory or the like that can be read by a computer, and the computer executes the program to realize the functions of the image processing apparatus described in this embodiment.
[0072]
【The invention's effect】
As is clear from the above description, the data feature extraction device according to the present invention is a data feature extraction device that generates feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space. A three-dimensional data acquisition unit that acquires three-dimensional data, a scanning unit that sequentially focuses on each position on the xy two-dimensional plane, and each time a position is focused by the scanning unit, the periphery of the focused position Xyz three-dimensional coordinates of a three-dimensional position having an x coordinate and a y coordinate in the local area among the three-dimensional positions indicated by the three-dimensional data for each of the four or more predetermined number of local areas on the xy two-dimensional plane An average position in space is obtained, and each of the average positions obtained for all the local regions is almost the same in the xyz three-dimensional coordinate space. It is determined whether or not it exists on the same plane, and the information indicating the gradient of the plane with respect to the xy two-dimensional plane is associated with the information indicating the focused position only when they exist on the same plane. And feature information generating means for generating as one element of the feature information.
[0073]
Thereby, the characteristic information about the three-dimensional data that can be used for the comparison between the three-dimensional data can be obtained. When the three-dimensional data is collated with the obtained feature information, the three-dimensional data having the relationship rotated around the straight line perpendicular to the xy two-dimensional plane has a similar tendency to some extent appropriately between the three-dimensional data representing the same object. It can be determined that the price is high.
[0074]
Here, the three-dimensional data is image data indicating a position and a luminance value on the xy two-dimensional plane for each pixel arranged on the xy two-dimensional plane, and the three-dimensional position is a luminance value for the pixel. And the x coordinate and y coordinate of the position on the xy two-dimensional plane with respect to the pixel, and the scanning means includes a pixel related to the image data on the xy two-dimensional plane. The feature information generating unit sequentially focuses on each position in the region, and the feature information generation unit performs processing for each local region on a predetermined number of four or more xy two-dimensional planes located around the position focused by the scanning unit. The average position on the xy two-dimensional plane and the average value of the luminance values for all the pixels arranged at positions on the xy two-dimensional plane having the x coordinate and the y coordinate in the xyz 3 It is determined as an average position in the original coordinate space, and it is determined whether or not each average position calculated for all the local regions is substantially on the same plane in the xyz three-dimensional coordinate space. The information indicating the gradient of the plane with respect to the xy two-dimensional plane and the information indicating the focused position may be generated as one element of the feature information.
[0075]
As a result, the feature information is generated by extracting features only in the local area where the luminance distribution is flat, so that the influence of the positional reference of sampling and the rotation processing of the image data in the digital image data generation stage is affected. It will not appear in the feature information, and if this feature information is used, it becomes possible to appropriately match two image data in a rotational relationship.
[0076]
The data feature extraction apparatus further includes line drawing data acquisition means for acquiring binary image data indicating a position on the xy two-dimensional plane and binary data for each pixel arranged on the xy two-dimensional plane; Binary image scanning means for sequentially paying attention to each position on an xy two-dimensional plane where pixels related to the value image data are arranged, and included in an area of a predetermined size centered on the position noted by the binary image scanning means Among the pixels related to the binary image data, the attribute value is determined so that the attribute value decreases as the number of pixels in which the binary data for the pixel indicates the first value increases, Attribute value determining means for associating the position with the attribute value, and each set of the position and attribute value on the xy two-dimensional plane associated with the attribute value determining means as the three-dimensional data. Transmission means for transmitting to the original data acquisition means, and the three-dimensional position is indicated by an x-coordinate and a y-coordinate of each position constituting the three-dimensional data and a z-coordinate which is an attribute value corresponding to the position. It may be a thing.
[0077]
Thus, since the line density is an attribute value for each position on the xy two-dimensional plane, when the obtained feature information is used, the binary image data representing the line drawing is compared with other image data. Even when the line is partially broken and becomes a broken line, the similarity of the image data can be determined appropriately.
The three-dimensional data is image data indicating a position on the xy two-dimensional plane and an attribute value for each pixel arranged on the xy two-dimensional plane, and the three-dimensional position is an attribute value for the pixel. It is indicated by a certain z coordinate and an x coordinate and a y coordinate of a position on the xy two-dimensional plane for the pixel, and the scanning unit is provided with a pixel related to the image data on the xy two-dimensional plane. The feature information generation unit pays attention to each position in the region sequentially, and the feature information generation unit uses the position as a reference in a region within a certain range on the xy two-dimensional plane centered on the position focused by the scanning unit. The average value z1 of each attribute value corresponding to the first local region between the negative x direction and the negative y direction, and the second local region between the positive x direction and the negative y direction Average of each attribute value z2, the average value z3 of the attribute values corresponding to the third local region between the negative x direction and the positive y direction, and the fourth local region between the positive x direction and the positive y direction The average position is obtained by obtaining the average value z4 of each attribute value corresponding to the region, and when the value of the mathematical formula | z1-z2-z3 + z4 | is equal to or less than the predetermined value d1, all the local regions are It is determined that each of the obtained average positions is on substantially the same plane in the xyz three-dimensional coordinate space, and information indicating the gradient of the plane with respect to the xy two-dimensional plane corresponds to information indicating the focused position. The attached one may be generated as one element of the feature information.
[0078]
As a result, in image data or the like in which pixels are arranged in a two-dimensional array on the xy two-dimensional plane, the gradient of the luminance value with respect to the xy two-dimensional data plane when the luminance value of the pixel is taken as the z coordinate is in the x and y directions. It is possible to easily determine whether or not the distribution of the luminance value, which is the basis for calculating the gradient of the luminance value when it is used as the feature information of the image data, is planar. . When the average luminances z1 to z4 of the regions R1 to R4 shown in the embodiment are in a planar relationship, the spatial gradient of the luminance of the scanning point is obtained on the basis of the calculation of these z1 to z4 to show the image characteristics. Using a method that is to be treated as information, for example, each of two image data obtained by imaging the same object by rotating the camera so that the relationship is rotated by 90 degrees with each other. When information indicating the characteristics is obtained, the information indicating the characteristics of each image is the same.
[0079]
The region of the certain range related to the feature information generating unit is a circular region having a constant radius centered on the position focused by the scanning unit, and the feature information generating unit is further focused by the scanning unit. The first to fourth areas determined with reference to the position are rotated by a predetermined angle with the reference position as a center, and the first to fourth areas after the rotation correspond to each other. An average value z1 ′, z2 ′, z3 ′ and z4 ′ of each attribute value is obtained, and the feature information generating means is a case where the value of the mathematical formula | z1−z2−z3 + z4 | is equal to or less than a predetermined value d1. In addition, when the value of the mathematical formula | z1′−z2′−z3 ′ + z4 ′ | is equal to or smaller than the predetermined value d1, each of the average positions obtained for all the local regions is all in the xyz three-dimensional coordinate space. Exists on almost the same plane It determines that that, those associated with information indicating the information and the focus position indicating a gradient with respect to the xy2 dimensional plane of the plane may be generated as a component of the characteristic information.
[0080]
Thereby, the characteristic information which shows the characteristic of each of the two image data which expresses the same object so that it may be in a relation rotated to an angle other than 90 degrees can be the same.
Further, the feature information generation means further associates information indicating the gradient and information indicating the focused position only when the gradient of the plane with respect to the xy two-dimensional plane is equal to or greater than a predetermined amount. It may be generated as one element of the feature information.
[0081]
As a result, since only the point where the luminance change is a certain amount is extracted as the feature of the image data, the number of data used for collation can be effectively reduced, and the calculation time required for collation can be shortened.
Further, the feature information generating means may calculate a mathematical expression representing the plane when the average positions obtained for all the local regions are substantially on the same plane, a · x + b · y + c · z = d ( a, b, c, and d are constant values), a vector V on an xy two-dimensional plane having x component and y component as -a / c and -b / c is obtained, and the magnitude of the vector V is 1 The unit vector obtained as follows may be specified as information indicating the gradient related to the plane.
[0082]
By such normalization, information indicating the characteristics of three-dimensional data such as multi-gradation image data is not affected by whether the range of attribute values corresponding to the z coordinate such as luminance is narrow or wide, and is used for collation. It can be used for general purposes.
In addition, the feature information generation unit calculates a vector V on an xy two-dimensional plane indicating the gradient related to the plane when all the average positions obtained for all the local regions are present on substantially the same plane. Specifying the vector V as information indicating the gradient related to the plane, and generating a correspondence between the vector V and the information indicating the focused position as one element of the feature information, The data feature extraction apparatus further determines a unit vector S starting from one position Q in the xyz three-dimensional coordinate space and going in one direction, and each of the unit V associated with each vector V generated by the feature information generating means. For the information, a vector PQ indicating a relative positional relationship of the position Q in the xyz three-dimensional coordinate space is defined with reference to the focused position related to the information. Collation data generated as collation data for use in collation between the three-dimensional data acquired by the three-dimensional data acquisition means and other three-dimensional data, information indicating a plurality of sets of tor PQs and unit vectors S It is good also as providing a production | generation means.
[0083]
As a result, data that can be directly used for collation is generated.
Further, the verification data generation means uniformly multiplies the size of each vector PQ so that the size of the vector PQ having the maximum size among the plurality of sets of vectors PQ is 1, Information indicating each vector PQ, each vector V, and the unit vector S obtained as a result of multiplication by a constant may be generated as the verification data.
[0084]
According to the data generated in this way, when two pieces of data to be collated indicate the same target and are in a relationship of enlargement / reduction, it can be appropriately determined.
A data collating apparatus according to the present invention is a data collating apparatus that collates three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space, and obtains first three-dimensional data. A three-dimensional data acquisition unit, a first scanning unit that sequentially focuses on each position on the xy two-dimensional plane, and each time one position is focused by the first scanning unit, is located around the focused position. For each of the four or more predetermined number of local regions on the xy two-dimensional plane, among the three-dimensional positions indicated by the first three-dimensional data, the xyz three-dimensional coordinates of the three-dimensional position having the x coordinate and the y coordinate in the local region An average position in the space is obtained, and whether or not each of the average positions obtained for all the local regions exists on substantially the same plane in the xyz three-dimensional coordinate space. The vector V on the xy two-dimensional plane indicating the gradient of the plane with respect to the xy two-dimensional plane is specified only when the planes are substantially on the same plane, and the vector V and information indicating the focused position are obtained. First feature information generating means to be associated with, a unit vector S starting from one position Q in the xyz three-dimensional coordinate space and going in one direction, and associated with each vector V specified by the first feature information generating means For each piece of information, a vector PQ indicating a relative positional relationship of the position Q in the xyz three-dimensional coordinate space is defined with reference to the focused position relating to the information, and a plurality of sets of the vector V and the vector PQ and a unit vector S are shown. Verification data generation means for generating verification data, second 3D data acquisition means for acquiring second 3D data, xy 2D flat A second scanning unit that sequentially focuses on each position on the top, and every time one position is focused by the second scanning unit, on a predetermined number of four or more xy two-dimensional planes located around the focused position For each of the local regions, the average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x-coordinate and the y-coordinate in the local region among the three-dimensional positions indicated by the second three-dimensional data is determined. It is determined whether or not each average position obtained for the local region is on substantially the same plane in the xyz three-dimensional coordinate space, and the xy two-dimensional of the plane is only when the average position is on the same plane. Second feature information generating means for specifying a vector V ′ on an xy two-dimensional plane indicating a gradient with respect to the plane and associating the vector V ′ with information indicating the focused position By sequentially applying each vector V in the verification data to each vector V ′ specified by the second feature information generating means, the orientation relationship between the vector V ′ and the new unit vector S ′ is a vector. Each unit vector S ′ is defined so as to be the same as the orientation relationship between the vector V applied to V ′ and the unit vector S, and according to the distribution state in the xyz three-dimensional coordinate space of all the determined unit vectors S ′. The apparatus further comprises collating means for determining a similarity tendency between the first three-dimensional data and the second three-dimensional data.
[0085]
As a result, the three-dimensional data in a rotational relationship can be appropriately collated.
Here, the collating means may perform the fitting of each vector V to each vector V ′ only when the difference between the magnitude of the vector V ′ and the magnitude of the vector V is less than a predetermined value. Good.
[0086]
Thereby, collation can be performed more accurately.
Further, the data collating device further includes a cubic region in which the starting point of the unit vector S ′ determined by the collating means is most included in each cubic region when the xyz three-dimensional coordinate space is divided into cubic regions of a predetermined size. In addition, it may be provided with a recording means for recording the number of unit vectors S ′ including the starting point on a recording medium.
[0087]
Thereby, the similarity between the data to be collated can be expressed numerically. Therefore, it is possible to specify data having the highest similarity.
Further, when the xyz three-dimensional coordinate space is divided into cube regions of a predetermined size, the data collating device further combines all unit vectors S ′ including the starting points in the cube regions for each cube region. It is also possible to provide recording means for obtaining the magnitude of the resultant composite vector and recording the maximum value among the magnitudes of all the obtained composite vectors on the recording medium.
[0088]
This makes it possible to obtain an index that can evaluate the degree of similarity between data with a certain degree of accuracy.
Instead of the collation data generating means, a unit vector S starting from one position Q on the xy two-dimensional plane and going in one direction is determined, and each vector V specified by the first feature information generating means For each piece of associated information, a vector PQ indicating the relative positional relationship of the position Q in the xy two-dimensional plane is defined with reference to the x coordinate and the y coordinate of the focused position related to the information, and a plurality of sets of the vector V and the vector PQ And two-dimensional collation data generating means for generating collation data indicating the unit vector S, and instead of the collating means, the collation is performed for each vector V ′ specified by the second feature information generating means. The vector V ′ and the new unit vector S ′ are applied to the vector V ′ by sequentially applying the vectors V in the business data. Each unit vector S ′ is defined so as to be the same as the azimuth relationship between V and the unit vector S, and the first three-dimensional data and the first three-dimensional data Two-dimensional matching means for determining a similarity tendency with the second three-dimensional data may be provided.
[0089]
As a result, it is possible to easily perform rough matching.
The data collating device further includes line drawing data acquisition means for acquiring binary image data indicating a position on the xy two-dimensional plane and binary data for each pixel arranged on the xy two-dimensional plane; A binary image scanning unit that sequentially focuses on each position on an xy two-dimensional plane in which pixels related to image data are arranged, and a region of a predetermined size centered on the position focused by the binary image scanning unit Among the pixels related to the binary image data, the attribute value is determined so that the attribute value becomes lower as the number of pixels in which the binary data for the pixel indicates the first value increases, and the focused position Attribute value determining means for associating the attribute value with the attribute value, and each set of the position on the xy two-dimensional plane and the attribute value associated with each other by the attribute value determining means as the three-dimensional data. Transmission means for transmitting to the dimensional data acquisition means, and the three-dimensional position is indicated by an x-coordinate and a y-coordinate of each position constituting the three-dimensional data, and a z-coordinate which is an attribute value corresponding to the position. It may be a thing.
[0090]
Thus, for example, when collating image data generated by capturing a line drawing in which the outline of a certain object is handwritten on paper with a scanner and image data obtained by capturing the same object with a digital camera When the line drawn as a solid line becomes a broken line in the process of capturing by the scanner, it is appropriate that the similarity between the two image data is higher than when the conventional image matching technology is applied. The possibility that it can be determined is increased.
[Brief description of the drawings]
FIG. 1 is a functional block diagram of an image collating apparatus according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating a relationship between a line drawing portion and first scanning points in search reference data.
FIG. 3 is a diagram illustrating an example of a relationship between a second scanning point used in feature extraction processing and each region of the first quadrant to the fourth quadrant.
FIG. 4 is a flowchart showing a first process for acquiring search reference data and generating data for collation.
5 is a diagram illustrating V (i) and P (i) generated by a feature data generation unit 103, and a point Q and a vector S determined by a collation data generation / holding unit 105. FIG.
FIG. 6 is a flowchart showing a second process for identifying data to be searched and performing matching between search reference data and data to be searched using data for matching;
FIG. 7 shows V (i), PQ (i), and V (i) and PQ (i) held by the collation data generation and holding unit 105 for each set of V ′ (j) and P ′ (j) generated by the feature data generation unit 109; It is a figure which illustrates a mode that unit relationship S '(i, j) was defined by applying the relationship of S.
[Explanation of symbols]
100 image verification device
101 Search criteria data acquisition unit
102 Luminance level determination unit
103 feature data generator
104 Range limiting part
105 Collation data generation and holding unit
106 Searched data acquisition unit
107 Line drawing part
108 Luminance level determination unit
109 Feature data generator
110 Verification unit
111 Evaluation value recording part

Claims (21)

A data feature extraction device that generates feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
Three-dimensional data acquisition means for acquiring three-dimensional data;
scanning means for sequentially paying attention to each position on the xy two-dimensional plane;
Each time one position is noticed by the scanning means, each of the local regions on a predetermined number of four or more xy two-dimensional planes located around the noticed position,
Among the three-dimensional positions indicated by the three-dimensional data, an average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x coordinate and the y coordinate in the local area is obtained, and each average obtained for all the local areas is obtained. It is determined whether or not all the positions are on substantially the same plane in the xyz three-dimensional coordinate space, and information indicating the gradient of the plane with respect to the xy two-dimensional plane and the focus only when they are on the same plane A data feature extraction apparatus comprising: feature information generation means for generating information corresponding to the indicated position as one element of the feature information. The three-dimensional data is image data indicating a position and a luminance value on the xy two-dimensional plane for each pixel arranged on the xy two-dimensional plane,
The three-dimensional position is indicated by a z-coordinate that is a luminance value for the pixel, and an x-coordinate and a y-coordinate of a position on the xy two-dimensional plane for the pixel,
The scanning unit sequentially pays attention to each position in a region where pixels related to the image data on an xy two-dimensional plane are arranged,
The feature information generating means is an xy two-dimensional image having x coordinates and y coordinates in each of the local areas on a predetermined number of four or more xy two-dimensional planes located around the position focused by the scanning means. The average position on the xy two-dimensional plane and the average value of the luminance values for all the pixels arranged at the positions on the plane are obtained as the average position in the xyz three-dimensional coordinate space, and each average obtained for all the local regions is obtained. It is determined whether or not the target positions are on substantially the same plane in the xyz three-dimensional coordinate space, and information indicating the gradient of the plane with respect to the xy two-dimensional plane only when it exists on the same plane. The data according to claim 1, wherein the information indicating the focused position is generated as one element of the feature information. Feature extraction apparatus. The data feature extraction device further includes:
line drawing data acquisition means for acquiring binary image data indicating a position on the xy two-dimensional plane and binary data for each pixel arranged on the xy two-dimensional plane;
Binary image scanning means for sequentially paying attention to each position on an xy two-dimensional plane in which pixels related to the binary image data are arranged;
Of the pixels related to the binary image data included in the region of a predetermined size centered on the position focused by the binary image scanning means, the binary data for the pixel is a pixel indicating the first value. An attribute value is determined so that the attribute value decreases as the number increases, and an attribute value determination unit that associates the focused position with the attribute value;
Transmitting means for transmitting each set of the position on the xy two-dimensional plane and the attribute value associated by the attribute value determining means to the three-dimensional data acquiring means as the three-dimensional data;
2. The three-dimensional position is indicated by an x-coordinate and a y-coordinate of each position constituting the three-dimensional data, and a z-coordinate which is an attribute value corresponding to the position. Data feature extraction device. The three-dimensional data is image data indicating a position on the xy two-dimensional plane and an attribute value for each pixel arranged on the xy two-dimensional plane,
The three-dimensional position is indicated by a z-coordinate which is an attribute value for the pixel, and an x-coordinate and a y-coordinate of a position on the xy two-dimensional plane for the pixel,
The scanning unit sequentially pays attention to each position in a region where pixels related to the image data on an xy two-dimensional plane are arranged,
The feature information generating means includes
A first local area between the negative direction of x and the negative direction of y from the reference in the region within a certain range on the xy two-dimensional plane centered on the position focused by the scanning means. Average value z1 of each attribute value corresponding to the region, average value z2 of each attribute value corresponding to the second local region between the positive direction of x and the negative direction of y, negative direction of x and y Average value z3 of each attribute value corresponding to the third local region between the positive direction and the average value of each attribute value corresponding to the fourth local region between the positive direction of x and the positive direction of y The respective average positions are obtained by obtaining z4, and when the value of the expression | z1-z2-z3 + z4 | is equal to or less than the predetermined value d1, each of the average positions obtained for all the local regions is all It is determined that they exist on substantially the same plane in the xyz three-dimensional coordinate space. The data feature extraction according to claim 1, wherein information indicating a gradient of a plane with respect to the xy two-dimensional plane and information indicating the focused position are generated as one element of the feature information. apparatus. The region of the certain range according to the feature information generating unit is a circular region having a certain radius centered on the position focused by the scanning unit;
The feature information generating means further rotates the first to fourth areas determined with the position focused by the scanning means as a reference by a predetermined angle around the reference position, and after the rotation Average values z1 ′, z2 ′, z3 ′ and z4 ′ of the attribute values corresponding to the first to fourth areas are determined,
The feature information generation means is a case where the value of the mathematical expression | z1-z2-z3 + z4 | When the value is less than or equal to the value d1, it is determined that all the average positions obtained for all the local regions are on substantially the same plane in the xyz three-dimensional coordinate space, and the gradient of the plane with respect to the xy two-dimensional plane is determined. 5. The data feature extraction apparatus according to claim 4, wherein information indicating information indicating the position of interest is generated as one element of the feature information. The feature information generating unit further associates information indicating the gradient with information indicating the focused position only when a gradient of the plane with respect to the xy two-dimensional plane is equal to or greater than a predetermined amount. 6. The data feature extraction apparatus according to claim 5, wherein the data feature extraction device is generated as one element of feature information. The feature information generation means calculates a formula representing the plane as a · x + b · y + c · z = d (a, when all the average positions obtained for all the local regions are substantially on the same plane. b, c, and d are constant values), a vector V on an xy two-dimensional plane having x and y components as -a / c and -b / c is obtained, and the magnitude of the vector V is obtained as 1. The data feature extraction apparatus according to claim 1, wherein a unit vector to be obtained is specified as information indicating the gradient related to the plane. The feature information generation unit specifies a vector V on an xy two-dimensional plane indicating the gradient related to the plane when all the average positions obtained for all the local regions are present on substantially the same plane. The vector V is identified as information indicating the gradient related to the plane, and the vector V is associated with the information indicating the focused position as one element of the feature information.
The data feature extraction device further includes:
A unit vector S starting from one position Q in the xyz three-dimensional coordinate space and starting in one direction is determined, and each piece of information associated with each vector V generated by the feature information generating unit is focused on the information. The vector PQ indicating the relative positional relationship of the position Q in the xyz three-dimensional coordinate space is defined with reference to the determined position, and information indicating the vector V, a plurality of sets of the vector PQ, and the unit vector S is acquired by the three-dimensional data acquisition unit. 2. The data feature extraction apparatus according to claim 1, further comprising collation data generation means for generating collation data for use in collation between the obtained three-dimensional data and other three-dimensional data. The verification data generation means includes:
Each vector PQ is obtained by multiplying the size of each vector PQ uniformly by a constant so that the size of the vector PQ having the maximum size among the plurality of sets of vectors PQ is 1. 9. The data feature extraction apparatus according to claim 8, wherein information indicating each vector V and the unit vector S is generated as the verification data. A data collating apparatus that collates three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
First three-dimensional data acquisition means for acquiring first three-dimensional data;
first scanning means for sequentially paying attention to each position on an xy two-dimensional plane;
Each time one position is noticed by the first scanning means, each of the four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Of the three-dimensional positions indicated by the first three-dimensional data, the average positions in the xyz three-dimensional coordinate space of the three-dimensional positions having the x coordinate and the y coordinate in the local area are obtained, and all the local areas are obtained. It is determined whether or not each average position is substantially on the same plane in the xyz three-dimensional coordinate space, and xy2 indicating the gradient of the plane with respect to the xy two-dimensional plane only when the average positions are on the same plane. First feature information generating means for identifying a vector V on a dimensional plane and associating the vector V with information indicating the focused position;
A unit vector S starting from one position Q in the xyz three-dimensional coordinate space and starting in one direction is determined, and each piece of information associated with each vector V specified by the first feature information generating unit is related to the information. Collation data for defining a vector PQ indicating the relative positional relationship of the position Q in the xyz three-dimensional coordinate space with reference to the focused position, and generating collation data indicating a plurality of sets of the vector V and the vector PQ and the unit vector S Generating means;
Second 3D data acquisition means for acquiring second 3D data;
second scanning means for sequentially focusing on each position on the xy two-dimensional plane;
Each time one position is noticed by the second scanning means, each of the four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Of the three-dimensional positions indicated by the second three-dimensional data, the average positions in the xyz three-dimensional coordinate space of the three-dimensional positions having the x coordinate and the y coordinate in the local area are obtained, and all the local areas are obtained. It is determined whether or not each average position is substantially on the same plane in the xyz three-dimensional coordinate space, and xy2 indicating the gradient of the plane with respect to the xy two-dimensional plane only when the average positions are on the same plane. A second feature information generating unit that identifies a vector V ′ on a dimension plane and associates the vector V ′ with information indicating the focused position;
By sequentially applying each vector V in the verification data to each vector V ′ specified by the second feature information generating means, the orientation relationship between the vector V ′ and the new unit vector S ′ is the vector V ′. Each unit vector S ′ is defined so as to have the same azimuth relationship between the vector V applied to the unit vector S and the unit vector S, and according to the distribution state in the xyz three-dimensional coordinate space of all the determined unit vectors S ′. A data collating apparatus comprising collating means for determining a similarity tendency between one three-dimensional data and the second three-dimensional data. The collation means performs the fitting of each vector V with respect to each vector V ′ only when the difference between the magnitude of the vector V ′ and the magnitude of the vector V is less than a predetermined value. Item 15. The data verification device according to Item 10. In the case where the xyz three-dimensional coordinate space is divided into cube regions of a predetermined size, the data collating device further includes a cubic region that includes the most starting points of the unit vector S ′ determined by the collating unit among the cubic regions. 11. The data collating apparatus according to claim 10, further comprising recording means for recording the number of unit vectors S ′ including the starting point on a recording medium. The data collating device is further obtained as a result of synthesizing all the unit vectors S ′ whose starting points are included in each cubic area when the xyz three-dimensional coordinate space is divided into cubic areas of a predetermined size. 11. The data collating apparatus according to claim 10, further comprising recording means for obtaining a magnitude of the combined vector and recording the maximum value of all the obtained magnitudes of the synthesized vectors on a recording medium. The data collating apparatus according to claim 10, wherein
Instead of the verification data generation means,
A unit vector S starting from one position Q on the xy two-dimensional plane and going in one direction is determined, and attention is paid to each information associated with each vector V specified by the first feature information generating means. A vector PQ indicating the relative positional relationship of the position Q in the xy two-dimensional plane is defined with reference to the x coordinate and the y coordinate of the determined position, and verification data indicating a plurality of sets of the vector V and the vector PQ and the unit vector S is generated. 2D collation data generation means
Instead of the verification means,
By sequentially applying each vector V in the verification data to each vector V ′ specified by the second feature information generating means, the orientation relationship between the vector V ′ and the new unit vector S ′ is the vector V ′. Each unit vector S ′ is defined so as to have the same azimuth relationship between the vector V applied to “and the unit vector S, and the first unit vector S ′ is determined according to the distribution state in the xy two-dimensional plane. A data collating apparatus comprising: a two-dimensional collation unit for determining a tendency of similarity between the three-dimensional data and the second three-dimensional data. The data verification device further includes:
line drawing data acquisition means for acquiring binary image data indicating a position on the xy two-dimensional plane and binary data for each pixel arranged on the xy two-dimensional plane;
Binary image scanning means for sequentially paying attention to each position on an xy two-dimensional plane in which pixels related to the binary image data are arranged;
Of the pixels related to the binary image data included in the region of a predetermined size centered on the position focused by the binary image scanning means, the binary data for the pixel is a pixel indicating the first value. An attribute value is determined so that the attribute value decreases as the number increases, and an attribute value determination unit that associates the focused position with the attribute value;
A transmission means for transmitting each set of the position on the xy two-dimensional plane and the attribute value associated by the attribute value determination means to the first three-dimensional data acquisition means as the three-dimensional data;
11. The three-dimensional position is indicated by an x-coordinate and a y-coordinate of each position constituting the three-dimensional data, and a z-coordinate that is an attribute value corresponding to the position. Data verification device. A data feature extraction method for generating feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
A three-dimensional data acquisition step for acquiring three-dimensional data;
a scanning step that sequentially focuses on each position on the xy two-dimensional plane;
Each time one position is noticed by the scanning step, for each of four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Among the three-dimensional positions indicated by the three-dimensional data, an average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x coordinate and the y coordinate in the local area is obtained, and each average obtained for all the local areas is obtained. It is determined whether or not all the positions are on substantially the same plane in the xyz three-dimensional coordinate space, and information indicating the gradient of the plane with respect to the xy two-dimensional plane and the focus only when they are on the same plane And a feature information generation step of generating, as one element of the feature information, an association with information indicating the position that has been performed. A data matching method for matching three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
A first three-dimensional data acquisition step of acquiring first three-dimensional data;
a first scanning step for sequentially focusing on each position on the xy two-dimensional plane;
Each time one position is noticed by the first scanning step, each of four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Of the three-dimensional positions indicated by the first three-dimensional data, the average positions in the xyz three-dimensional coordinate space of the three-dimensional positions having the x coordinate and the y coordinate in the local area are obtained, and all the local areas are obtained. It is determined whether or not each average position is substantially on the same plane in the xyz three-dimensional coordinate space, and xy2 indicating the gradient of the plane with respect to the xy two-dimensional plane only when the average positions are on the same plane. A first feature information generation step of identifying a vector V on a dimensional plane and associating the vector V with information indicating the focused position;
A unit vector S starting from one position Q in the xyz three-dimensional coordinate space and starting in one direction is determined, and each piece of information associated with each vector V specified in the first feature information generation step is related to the information. Collation data for defining a vector PQ indicating the relative positional relationship of the position Q in the xyz three-dimensional coordinate space with reference to the focused position, and generating collation data indicating a plurality of sets of the vector V and the vector PQ and the unit vector S Generation step;
A second three-dimensional data acquisition step of acquiring second three-dimensional data;
a second scanning step for sequentially paying attention to each position on the xy two-dimensional plane;
Each time one position is noticed in the second scanning step, the second three-dimensional data is obtained for each of the four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position. Among the three-dimensional positions shown, the average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x coordinate and the y coordinate in the local area is obtained, and each average position obtained for all the local areas is It is determined whether or not they exist on substantially the same plane in the xyz three-dimensional coordinate space, and the vector V ′ on the xy two-dimensional plane indicating the gradient of the plane with respect to the xy two-dimensional plane only when it exists on the same plane. And a second feature information generation step for associating the vector V ′ with information indicating the focused position;
By sequentially applying each vector V in the verification data to each vector V ′ specified in the second feature information generation step, the orientation relationship between the vector V ′ and the new unit vector S ′ is the vector V ′. Each unit vector S ′ is defined so as to have the same azimuth relationship between the vector V applied to the unit vector S and the unit vector S, and according to the distribution state in the xyz three-dimensional coordinate space of all the determined unit vectors S ′. A data collating method comprising: a collating step for determining a similarity tendency between one three-dimensional data and the second three-dimensional data. A computer program for causing a computer to perform data feature extraction processing for generating feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
The data feature extraction process includes:
A three-dimensional data acquisition step for acquiring three-dimensional data;
a scanning step that sequentially focuses on each position on the xy two-dimensional plane;
Each time one position is noticed by the scanning step, for each of four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Among the three-dimensional positions indicated by the three-dimensional data, an average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x coordinate and the y coordinate in the local area is obtained, and each average obtained for all the local areas is obtained. It is determined whether or not all the positions are on substantially the same plane in the xyz three-dimensional coordinate space, and information indicating the gradient of the plane with respect to the xy two-dimensional plane and the focus only when they are on the same plane And a feature information generation step of generating, as one element of the feature information, information associated with the information indicating the position. A computer program for causing a computer to perform data matching processing related to matching of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
The data matching process includes
A first three-dimensional data acquisition step of acquiring first three-dimensional data;
a first scanning step for sequentially focusing on each position on the xy two-dimensional plane;
Each time one position is noticed by the first scanning step, each of four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Of the three-dimensional positions indicated by the first three-dimensional data, the average positions in the xyz three-dimensional coordinate space of the three-dimensional positions having the x coordinate and the y coordinate in the local area are obtained, and all the local areas are obtained. It is determined whether or not each average position is substantially on the same plane in the xyz three-dimensional coordinate space, and xy2 indicating the gradient of the plane with respect to the xy two-dimensional plane only when the average positions are on the same plane. A first feature information generation step of identifying a vector V on a dimensional plane and associating the vector V with information indicating the focused position;
A unit vector S starting from one position Q in the xyz three-dimensional coordinate space and starting in one direction is determined, and each piece of information associated with each vector V specified in the first feature information generation step is related to the information. Collation data for defining a vector PQ indicating the relative positional relationship of the position Q in the xyz three-dimensional coordinate space with reference to the focused position, and generating collation data indicating a plurality of sets of the vector V and the vector PQ and the unit vector S Generation step;
A second three-dimensional data acquisition step of acquiring second three-dimensional data;
a second scanning step for sequentially paying attention to each position on the xy two-dimensional plane;
Each time one position is noticed by the second scanning step, each of the four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Of the three-dimensional positions indicated by the second three-dimensional data, the average positions in the xyz three-dimensional coordinate space of the three-dimensional positions having the x coordinate and the y coordinate in the local area are obtained, and all the local areas are obtained. It is determined whether or not each average position is substantially on the same plane in the xyz three-dimensional coordinate space, and xy2 indicating the gradient of the plane with respect to the xy two-dimensional plane only when it is on the same plane. A second feature information generating step of identifying a vector V ′ on the dimension plane and associating the vector V ′ with information indicating the focused position;
By sequentially applying each vector V in the collation data to each vector V ′ specified in the second feature information generation step, the orientation relationship between the vector V ′ and the new unit vector S ′ is the vector V ′. Each unit vector S ′ is defined so as to have the same azimuth relationship between the vector V and the unit vector S applied to “,” and the unit vector S ′ is determined according to the distribution state in the xyz three-dimensional coordinate space. The computer program characterized by including the collation step which determines the similarity tendency of 1 three-dimensional data and said 2nd three-dimensional data. A computer-readable recording medium storing a computer program for causing a computer to perform data feature extraction processing for generating feature information representing features of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space. ,
The data feature extraction process includes:
A three-dimensional data acquisition step for acquiring three-dimensional data;
a scanning step that sequentially focuses on each position on the xy two-dimensional plane;
Each time one position is noticed by the scanning step, for each of four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Among the three-dimensional positions indicated by the three-dimensional data, an average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x coordinate and the y coordinate in the local area is obtained, and each average obtained for all the local areas is obtained. It is determined whether or not all the positions are on substantially the same plane in the xyz three-dimensional coordinate space, and information indicating the gradient of the plane with respect to the xy two-dimensional plane and the focus only when they are on the same plane And a feature information generation step of generating, as one element of the feature information, information associated with the information indicating the position. A computer-readable recording medium recording a computer program for causing a computer to perform data collation processing related to collation of three-dimensional data indicating a plurality of three-dimensional positions in an xyz three-dimensional coordinate space,
The data matching process includes
A first three-dimensional data acquisition step of acquiring first three-dimensional data;
a first scanning step for sequentially focusing on each position on the xy two-dimensional plane;
Each time one position is noticed by the first scanning step, each of four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position,
Of the three-dimensional positions indicated by the first three-dimensional data, the average positions in the xyz three-dimensional coordinate space of the three-dimensional positions having the x coordinate and the y coordinate in the local area are obtained, and all the local areas are obtained. It is determined whether or not each average position is substantially on the same plane in the xyz three-dimensional coordinate space, and xy2 indicating the gradient of the plane with respect to the xy two-dimensional plane only when the average positions are on the same plane. A first feature information generation step of identifying a vector V on a dimensional plane and associating the vector V with information indicating the focused position;
A unit vector S starting from one position Q in the xyz three-dimensional coordinate space and starting in one direction is determined, and each piece of information associated with each vector V specified in the first feature information generation step is related to the information. Collation data for defining a vector PQ indicating the relative positional relationship of the position Q in the xyz three-dimensional coordinate space with reference to the focused position, and generating collation data indicating a plurality of sets of the vector V and the vector PQ and the unit vector S Generation step;
A second three-dimensional data acquisition step of acquiring second three-dimensional data;
a second scanning step for sequentially paying attention to each position on the xy two-dimensional plane;
Each time one position is noticed in the second scanning step, the second three-dimensional data is obtained for each of the four or more predetermined number of local regions on the xy two-dimensional plane located around the noticed position. Among the three-dimensional positions shown, the average position in the xyz three-dimensional coordinate space of the three-dimensional position having the x coordinate and the y coordinate in the local area is obtained, and each average position obtained for all the local areas is It is determined whether or not they exist on substantially the same plane in the xyz three-dimensional coordinate space, and the vector V ′ on the xy two-dimensional plane indicating the gradient of the plane with respect to the xy two-dimensional plane only when it exists on the same plane. And a second feature information generation step for associating the vector V ′ with information indicating the focused position;
By sequentially applying each vector V in the verification data to each vector V ′ specified in the second feature information generation step, the orientation relationship between the vector V ′ and the new unit vector S ′ is the vector V ′. Each unit vector S ′ is defined so as to have the same azimuth relationship between the vector V applied to the unit vector S and the unit vector S, and according to the distribution state in the xyz three-dimensional coordinate space of all the determined unit vectors S ′. And a collating step for determining a similarity tendency between the first three-dimensional data and the second three-dimensional data.

Images