JP3852285B2 - 3D shape measuring apparatus and 3D shape measuring method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is a three-dimensional method based on a triangulation method in which a pattern projection image obtained by irradiating a measurement target with pattern light is imaged from different directions by a plurality of imaging means, and distance information is obtained based on a change in the pattern The present invention relates to a shape measuring apparatus and a three-dimensional shape measuring method.
[0002]
[Prior art]
As a method for acquiring a three-dimensional shape, there are an active method (Active vision) and a passive method (Passive vision). Active methods include (1) a laser method that emits laser light, ultrasonic waves, etc., measures the amount of reflected light and the arrival time from an object, and extracts depth information, and (2) a special pattern light source such as slit light. A pattern projection method for estimating the target shape from image information such as geometric deformation of the target surface pattern using (3), a method for obtaining three-dimensional information by forming contour lines with moire fringes by optical processing, etc. There is. On the other hand, the passive method uses knowledge about the appearance of an object, light source, illumination, shadow information, etc., and uses monocular stereoscopic vision to estimate 3D information from a single image, depth information of each pixel by triangulation principle And binocular stereopsis for estimating
[0003]
In general, the active method has higher measurement accuracy, but the measurable range is often small due to limitations of the light projecting means. On the other hand, the passive method is general-purpose and has few restrictions on objects. The present invention relates to a pattern projection method which is a three-dimensional measuring apparatus using this active method.
[0004]
In the pattern projection method, reference pattern light is projected onto a target object, and photographing is performed from a direction different from the direction in which the reference pattern light is projected. The photographed pattern is deformed depending on the shape of the object. By associating the observed deformation pattern with the projected pattern, three-dimensional measurement of the object can be performed. The problem with the pattern projection method is how to reduce the miscorrespondence and simplify the correspondence between the deformation pattern and the projected pattern. Accordingly, various pattern projection methods (spatial pattern coding, moire, color coding) have been proposed.
[0005]
As a typical example of spatial coding, there is a configuration disclosed in, for example, Japanese Patent Laid-Open No. 5-3327375. In this example, a laser light source and a lens system that shapes laser light into a slit shape, a scanning device that scans and irradiates the shaped laser light onto an object, a camera that detects reflected light from the object, and these It consists of a device to control.
[0006]
A striped pattern is formed on the object by the laser beam scanned from the scanning device between a portion irradiated with the laser beam and a portion not irradiated with the laser beam. By irradiating the laser beam with a plurality of different patterns, the object is divided into N identifiable portions. The shape of the object can be calculated by determining in which part the pixels on the image captured by the camera from different positions are included.
[0007]
In order to increase the resolution, it is necessary to perform scanning with a plurality of lasers and to perform imaging with a plurality of cameras. For example, in order to divide the screen into 256 areas, it is necessary to shoot 8 times. For this reason, it is difficult to capture a fast-moving object, and it is necessary to securely fix the imaging system during scanning. Therefore, even if the apparatus itself is simple, it is difficult to easily perform imaging.
[0008]
As a means for reducing the number of times the pattern is projected, there is color coding disclosed in Japanese Patent Laid-Open No. 3-192474. In color coding, when q and k are set to a predetermined natural number of 2 or more, using colors of q colors or more, two adjacent slit lights are not the same color, but by adjacent k slit lights. A pattern encoded so that the color sequence appears only once is projected, the color of the slit is detected from the observed image, and the slit number is obtained from the color sequence of the corresponding slit. From the slit number, the irradiation direction of the slit can be calculated, and the distance can be calculated in the same manner as in the spatial coding example.
[0009]
However, in color coding, there is a problem in that the amount of calculation for code restoration is large because codes are restored from the arrangement of code strings. Furthermore, when dividing into 256 areas using three colors of R, G, and B, it is necessary to know the arrangement of the eight slit lights around the slit where the code is to be known, and the slits are continuously long. It is only suitable for measuring objects that can be shaped.
[0010]
There is a spatial pattern coding method disclosed in Japanese Patent No. 2565885 as means for easily restoring a slit and projecting a pattern coded once. In this patent, there are three or more gradation regions by three or more shades, three or more colors, or a combination of shades and colors, and at least three types of floors at the intersection of the boundary of the gradation regions. According to the type and order of gradations that are in contact with the intersection of the projected images produced by projecting the pattern onto the object to be measured. A main code is assigned, and a combination code obtained by combining the main code or the main code of the intersection with the main code of the surrounding intersection is assigned as a feature code for identifying the intersection.
[0011]
However, in the above-described method, depending on the object to be photographed, the coding may be lost and correct code association may not be possible. For example, when the pattern sequence projected by the projection pattern light source is “12345678”, the pattern sequence captured by the camera may be recognized as “1267” depending on the structure to be imaged, or the pattern sequence “758” In some cases, an inverted pattern sequence is obtained. In addition, it is difficult to associate the projected pattern with the photographed pattern sequence due to the shape and reflectance of the object.
[0012]
In color coding, when slits are grouped at the time of decoding, this problem is avoided by using a method that does not perform decoding for patterns that may be missing or inverted. In the spatial pattern coding method, the possibility of the aforementioned error is reduced by using a two-dimensional pattern, but in principle, the same error occurs depending on the object. Therefore, in the above-described method, although excellent accuracy can be obtained in shooting in a special situation in a laboratory or a situation in which a target object is limited, deterioration in accuracy cannot be denied in a general imaging situation in which a target is not limited. Further, in the method of performing light projection using a light source, when a target having a wide range is targeted, a three-dimensional shape cannot be obtained for a portion where the light projection does not reach. In addition, since a shadow area generated by the object blocking the projected pattern cannot be measured, there is an area where the distance cannot be obtained even though it is actually visible.
[0013]
Therefore, in Japanese Patent Application No. 10-191063, (Japanese Patent Laid-Open No. 2000-9442), and Japanese Patent Application No. 10-247796 (Japanese Patent Laid-Open No. 2000-65542) related to the same applicant as this application, the projected pattern is fed back and a new one is added. We proposed a 3D image capturing device that does not depend on the object by generating code.
[0014]
[Problems to be solved by the invention]
The three-dimensional image photographing apparatus described in Japanese Patent Application No. 10-191063 (Japanese Patent Application Laid-Open No. 2000-9442) and Japanese Patent Application No. 10-247796 (Japanese Patent Application Laid-Open No. 2000-65542) uses a light projection pattern having a plurality of intensities and a plurality of wavelengths. This is realized by projecting what is encoded by At this time, the projection pattern changes due to the influence of the luminance information of the subject, the material, and the like, and when calculating the three-dimensional shape, an error occurs and an appropriate three-dimensional shape cannot be measured. Therefore, the three-dimensional imaging apparatus is arranged on the same optical axis as the light projecting element, monitors the amount of change in the light projection pattern according to the subject information, performs re-encoding, and measures the three-dimensional shape. .
[0015]
However, the above configuration is based on the premise that a new code obtained by one camera is generated in another camera as the same new code. This requires that all cameras in the device have the same output characteristics and output the same signal for a certain lightness on the target object. It is difficult in terms of cost to prepare a plurality of cameras that strictly satisfy this condition. Even if it is prepared, it is unavoidable that variations occur due to changes in timekeeping or fluctuations in the operating environment during use.
[0016]
Furthermore, the new code generated by the texture on the target object is not always unique. This is because the same new code can be generated by different textures at different points on the target object. That is, when the reflectance at the point p (x, y) is r (x, y) and the luminance of the projected code is i (x, y), two different points P1 (x1, y It is assumed that the relationship between the reflectance of y1) and the point P2 (x2, y2) and the projected luminance is the relationship of the following formula (1).
[Expression 1]
r (x1, y1) / r (x2, y2) ≒ i (x2, y2) / i (x1, y1) ... (1)
[0017]
If two different points P1 (x1, y1) and point P2 (x2, y2) on the target object are in the relationship of the above equation (1), the relationship of the following equation (2) is satisfied.
[Expression 2]
r (x1, y1) × i (x1, y1) ≒ r (x2, y2) × i (x2, y2) ... (2)
[0018]
As a result, the brightness of the image photographed by the camera is the same at two different points P1 (x1, y1) and P2 (x2, y2) on the target object. Therefore, when the association is attempted based on the brightness of the captured image of each camera, it is difficult to associate the point captured by one camera with the point captured by the other camera. In particular, in a configuration in which association processing is performed using captured images using a plurality of cameras, output values that differ from camera to camera may vary from camera to camera depending on the characteristics of each camera. If the images are different from each other, the image projected with the projection pattern as described above may be taken with a plurality of cameras, and the points taken with one camera may be associated with the points taken with the other camera. Matching may be difficult or inaccurate.
[0019]
The present invention has been made for the purpose of solving the above-described problems, and re-encoding can be performed based on accurate association even when using a group of cameras whose output characteristics vary. An object of the present invention is to provide a highly reliable three-dimensional shape measurement method and apparatus that can be made less susceptible to erroneous correspondence by reducing the dependence on the texture of the target object.
[0020]
[Means for Solving the Problems]
The present invention solves the above-mentioned object, and the first aspect thereof is
A light projecting means for projecting a coded slit pattern given by using a light source, and a first camera having substantially the same viewpoint and the same optical axis as the pattern projected by the light projecting means; In the three-dimensional shape measuring apparatus composed of a second camera arranged on an optical axis different from that of the first camera,
Output value storage means for storing output value information between edges of a pattern projection image captured by the first and second cameras, wherein a pattern is projected on a reference object and a pattern is measured on a measurement object. Output value storage means for storing output values of the first camera and the second camera when projected,
Reflection coefficient calculation means for calculating a reflection coefficient for the measurement object based on output values of the first camera when a pattern is projected onto a reference object and when a pattern is projected onto a measurement object ,
Output value prediction for calculating the output prediction value of the second camera by calculating the reflection coefficient and the second camera output value when a pattern is projected onto the reference object stored in the storage means Means,
An output value that is the closest to the predicted output value of the second camera calculated by the output value predicting means from the output value information between edges of the pattern projection image that is projected onto the measurement object and captured by the second camera And an associating means for selecting an edge having a selected point as a corresponding point for the output value of the first camera;
The three-dimensional shape measuring apparatus is characterized by having
[0021]
Furthermore, in one embodiment of the three-dimensional shape measuring apparatus of the present invention, the apparatus has edge search means for detecting a similar pattern in the output value of the second camera when a pattern is projected onto a measurement object, and the output When a similar pattern is detected by the edge search means, the value predicting means outputs the output value of the second camera and the reflection coefficient when a pattern is projected onto the reference object stored in the output value storage means An output predicted value of each edge in the second camera is calculated based on the reflection coefficient calculated by the calculation means, and the association means calculates the corresponding edge of the second camera based on the comparison with the similar pattern. It has the structure which performs the process which selects this.
[0022]
Furthermore, in one embodiment of the three-dimensional shape measurement apparatus of the present invention, the light emitted from the light projecting means and the light incident on the first camera are separated by a beam splitter, and the light projection The means and the first camera are arranged so as to be optically coaxial with each other.
[0023]
Furthermore, in one embodiment of the three-dimensional shape measuring apparatus of the present invention, the light projecting means has a light source that generates light in an invisible region, and the first camera has a filter that transmits light in the invisible region and an invisible region light. It has a filter that blocks light.
[0024]
Furthermore, in one embodiment of the three-dimensional shape measurement apparatus of the present invention, the second camera is configured by a plurality of cameras that capture the measurement object at different angles, and each of the plurality of second cameras is captured. The distance information obtained based on the projected pattern is synthesized to obtain a three-dimensional image.
[0025]
Furthermore, the second aspect of the present invention provides
A light projecting step of projecting a given slit pattern using a light source;
A first camera having substantially the same viewpoint and optical axis as the pattern projected by the projecting step, and a second camera disposed on an optical axis different from the first camera, An output value storage step of storing output value information between edges in the captured pattern image when the pattern is projected onto the reference object and when the pattern is projected onto the measurement object;
A reflection coefficient calculation step of calculating a reflection coefficient for the measurement object based on output values of the first camera when a pattern is projected onto a reference object and when a pattern is projected onto a measurement object; ,
Output value prediction for calculating the output prediction value of the second camera by calculating the reflection coefficient and the second camera output value when a pattern is projected onto the reference object stored in the storage means Steps,
An output value that is the closest to the predicted output value of the second camera calculated by the output value predicting means from the output value information between edges of the pattern projection image that is projected onto the measurement object and captured by the second camera And an associating step of extracting an edge having a selected point as a corresponding point for the output value of the first camera;
There is a three-dimensional shape measurement method characterized by comprising:
[0026]
Furthermore, in one embodiment of the three-dimensional shape measurement method of the present invention, the method includes an edge search step of detecting a similar pattern in the output value of the second camera when a pattern is projected onto a measurement object, and the output In the value prediction step, when a similar pattern is detected in the edge search step, the output value of the second camera and the reflection coefficient when a pattern is projected onto the reference object stored in the output value storage means An output predicted value of each edge in the second camera is calculated based on the reflection coefficient calculated by the computing means, and the association step calculates the corresponding edge of the second camera based on the comparison with the similar pattern. A process of selecting is executed.
[0027]
Furthermore, in one embodiment of the three-dimensional shape measurement method of the present invention, the second camera is configured by a plurality of cameras that capture images of the measurement object at different angles, and each of the plurality of second cameras is captured. And a step of obtaining a three-dimensional image by synthesizing distance information obtained based on the projected pattern.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the three-dimensional shape measurement apparatus and the three-dimensional shape measurement method of the present invention will be described in detail with reference to the drawings.
[0029]
First, the principle of distance data acquisition using recoding processing will be described. FIG. 1 is a block diagram showing a configuration example of a three-dimensional shape measuring apparatus that executes distance data acquisition using recoding processing. FIG. 2 shows the positional relationship between the light source and the image sensor.
[0030]
In the configuration shown in FIG. 2, the three-dimensional shape measuring apparatus includes three cameras 101 to 103 and a projector 104. The illustrated distances I1, I2, and I3 are made equal so that the distance relations of the cameras are aligned. The cameras 3 and 103 and the projector 104 are arranged so that their optical axes coincide using a half mirror 105 as a beam splitter. The cameras 1 and 101 and the cameras 2 and 102 are arranged on both sides of the cameras 3 and 103 and the projector 104 so that their optical axes are different. The distance between the central optical axis and the optical axes on both sides is the baseline length L.
[0031]
The projector 104 includes a light source 106, a mask pattern 107, an intensity pattern 108, and a prism 109. Here, as the light source 106, a light source in an invisible region using infrared light or ultraviolet light can be used. In this case, each camera is configured as shown in FIG. That is, the incident light 310 is divided into two directions by the prism 301, one of which passes through the invisible region (infrared or ultraviolet) transmission filter 302 and enters the imaging device (for example, a CCD camera) 303, and the other enters the invisible region. The light enters the imaging device 305 through the (infrared and ultraviolet) cutoff filter 304.
[0032]
In addition, the light source 106 illustrated in FIG. 2 is not limited to the visible region or the invisible region, and a light source having a wavelength band that can be imaged may be used. In this case, the cameras 3 and 103 use progressive scan type CCD cameras, and the configuration of the cameras 1 and 101 and the cameras 2 and 102 is not particularly limited. However, considering the correspondence with the cameras 3 and 103, a CCD camera having the same configuration is desirable. A pattern is projected from the light source 106, and the three cameras 1 to 3 (101 to 103) take images simultaneously. Each camera obtains light that has passed through the filters 304 and 305 (see FIG. 3) by the imaging devices 303 and 305, thereby performing batch acquisition of images.
[0033]
The configuration of the three-dimensional shape measuring apparatus will be described with reference to FIG. As shown in the figure, the cameras 1 and 101 store the luminance information obtained by shooting in the luminance value memory 121 and store the shooting pattern in the pattern image memory 122. Similarly, the cameras 2 and 102 store the luminance information in the luminance value memory 123 and store the shooting pattern in the pattern image memory 124. The cameras 3 and 103 store the luminance information in the luminance value memory 125 and store the shooting pattern in the pattern image memory 126. The projector 104 divides each slit into cells on a square lattice and stores them in the frame memory 127 in order to refer to a coded pattern created in advance later.
[0034]
A three-dimensional image is obtained as follows using the stored photographing pattern and luminance information. The following operations are common to both the combination of the cameras 1, 101 and the cameras 3, 103, and the combination of the cameras 2, 102 and the cameras 3, 103. Here, the combination of the cameras 1, 101 and the cameras 3, 103 is taken as an example. explain.
[0035]
In FIG. 1, the area dividing unit 128 performs area division of the shooting pattern shot by the cameras 3 and 103. Then, a region where the intensity difference between adjacent slit patterns is equal to or smaller than the threshold is extracted as a region 1 where light from the projector does not reach, and a region where the intensity difference between the slit patterns is equal to or larger than the threshold is extracted as a region 2. To do. The re-encoding unit 129 re-encodes the extracted area 2 using the shooting pattern stored in the pattern image memory 126 and the projection pattern stored in the frame memory 127.
[0036]
FIG. 4 is a flowchart for performing re-encoding. First, each slit pattern is divided vertically for each slit width (step S11) to generate a square cell. The average value of the intensity is taken for each generated cell, and the average value is set as the intensity of each cell (step S12). Intensity between cells corresponding to the projection pattern and shooting pattern is compared in order from the center of the image, and the pattern changes due to factors such as the reflectance of the object and the distance to the object, so the intensity between the cells exceeds the threshold. It is determined whether they are different (step S13). If the difference is not more than the threshold value, the recoding is finished for all the photographed cells (step S17).
[0037]
If the difference is greater than the threshold, it is determined whether the cell has a new intensity (step S14). When the cell has a new strength, a new code is generated and assigned (step S15). If it is not a cell having a new intensity, it is coded using a sequence of slit patterns that can be distinguished from other appearing parts (step S16). This completes the recoding (step S17).
[0038]
FIG. 5 shows an example of the coding of the slit pattern. FIG. 5A shows a projection pattern coded by the arrangement of the slits, and the intensity is 3 (strong), 2 (medium), and 1 (weak), respectively. ) Is assigned. In FIG. 5B, since the strength changes in the third cell from the left and a new code appears, a code of 0 is newly assigned. In FIG. 8C, since the existing code appears in the third cell from the left and the second cell from the top, the vertical code is [232] and the horizontal code is set as a new code from the cell code. Recode as [131]. This re-encoding is equivalent to projecting a complex pattern such as a two-dimensional pattern at a site where the shape of the object is rich in change, and projecting a simple pattern at a site where there is little change. This process is repeated and recoding is performed by assigning unique codes to all cells.
[0039]
FIG. 6 shows an example in which the encoded pattern is projected onto the plate 606 disposed in front of the wall 605 using the cameras 601 to 603 and the projector 604. The pattern encoded here is the slit pattern shown in FIG. At this time, in the images obtained by the camera 601 and the camera 602, areas 801 and 901 which are shadows of the plate 606 are generated as shown in FIGS. In this example, a slit pattern as shown in FIG. 10 is obtained as a newly coded pattern on the surface of the plate 606.
[0040]
Next, referring back to FIG. The code decoding unit 130 on the camera 1, 101 side extracts the projection pattern from the pattern image memory 122 and divides it into cells in the same manner as described above. Then, the code of each cell is detected using the code re-encoded by the re-encoding unit 129, and the slit angle θ from the light source is calculated based on the detected code. FIG. 11 is a diagram showing a distance calculation method in spatial coding. The slit angle θ of the cell to which each pixel belongs, the x coordinate on the image taken by the camera 1, the focal length F which is a camera parameter, and the base length L. From the above, the distance Z is calculated by the following equation (3).
[0041]
[Equation 3]
Z = (F × L) / (x + F × tan θ) (3)
[0042]
The calculation of the distance Z is similarly performed in the code decoding unit 131 on the camera 2 and 102 side. For the above-described region 1, the distance is calculated as follows. In the area 1, since the pattern detection using the projected pattern cannot be performed, the corresponding point search unit 132 uses the luminance information read from the luminance value memories 121, 123, and 125 of the cameras 1 to 3 to perform the parallax. Is detected, and the distance is calculated based on this. Since the distance is calculated for the areas other than the area 1 by the above-described operation, the minimum distance of the area 1 can be obtained, and the pixels that can be associated are also limited. Using these restrictions, the pixels are correlated to detect the parallax d, and the distance Z is calculated by the following equation (4) using the pixel size λ which is a camera parameter.
[0043]
[Expression 4]
Z = (L × F) / (λ × d) (4)
[0044]
With the distance information obtained by the combination of the cameras 3 and 103 and the cameras 1 and 101 by the above-described method, the distance information of the area 801 that is a shadow of the plate shown in FIG. 8 cannot be detected. On the other hand, the distance information of the area 901 that is the shadow of the plate shown in FIG. 9 cannot be detected from the distance information obtained by the combination of the cameras 3 and 103 and the cameras 2 and 102. However, it is possible to calculate the distance information of the region 801 that is the shadow of the plate shown in FIG. Therefore, in the distance information integration unit 133 in FIG. 1, the distance information calculated by the pair of the cameras 3, 103 and the cameras 1, 101 and the distance information calculated by the cameras 3, 103 and the cameras 2, 102 are used. Distance information for all pixels of the image (FIG. 12) is acquired. The distance information obtained by the above operation is stored in a three-dimensional image memory in association with the luminance image of the camera 3, for example, thereby performing three-dimensional shape measurement.
[0045]
FIG. 13 shows a three-dimensional shape measurement processing system configuration including a projector 1302 that irradiates pattern light onto the measurement target 1301 and two imaging units 1303 and 1304 that execute imaging of the measurement target onto which the pattern light is projected. The imaging unit 1303 is configured to take an image through a half mirror 1305 as a beam splitter in order to take an image in the same direction as the projector 1302. Distance data to be measured by the above-described method based on the imaging patterns of the imaging units 1303 and 1304. That is, this is a configuration for obtaining a three-dimensional shape.
[0046]
The pattern projected onto the measurement target by the light projector 1302 is, for example, the pattern shown in FIG. 14A. When the measurement target has a shape as shown in FIG. 14B, the pattern image irradiated on the measurement target. The patterns photographed by the two imaging means are as shown in FIG. FIG. 15A is a pattern image projected onto a measurement object by the pattern projection projector, and FIG. 15B is a pattern image photographed by the cameras 1 and 1501 arranged at the same optical axis position as the pattern projection projector. FIG. 15C shows a pattern image captured by the camera 2 1502 that captures an object from a different direction from the camera 1.
[0047]
In the configuration of FIG. 15, when re-encoding based on the pattern images photographed by the cameras 1 and 1501 and the cameras 2 and 1502, an association process is performed in consideration of the characteristics and output of the camera, and an accurate distance is obtained. The processing of the 3D shape measuring apparatus and 3D shape measuring method of the present invention for acquiring data will be described.
[0048]
This will be described in detail with reference to the flowchart of FIG. First, in order to measure the output characteristics of the cameras 1 and 1501, the cameras 2 and 1502, and the respective cameras in the configuration of FIG. 15, a reference object that is not a measurement target is placed, and a pattern is projected onto the reference object (see FIG. 16). S101), and the projected projection pattern is imaged by each camera (S111, S121). This state is called a standard state. It is desirable to use a flat plate having a uniform reflection coefficient as the reference object.
[0049]
Edges are extracted (S112, S122) from an image obtained by photographing the projection pattern projected on the reference object with each camera, and an output value corresponding to each code is calculated. The output value can be calculated by calculating an average value of luminance values (S113, S123) for a plurality of pixels existing between edges. An output value corresponding to each code is calculated for each camera (camera 1, 1501, and camera 2, 1502 in FIG. 15), and the correspondence between each code and the output value is stored in a memory (S114, S124).
[0050]
Specifically, the n-th edge position in the image obtained by photographing the projection pattern projected on the reference object with the camera is En, and the left and right codes are C and C respectively._left(n), C_right(n). C_left(n) and C_rightThe output value of the camera 1,1501 for the code of (n) is O_left-1(n), O_right-1(n) and the output value of the camera 2 1502 is O_left-2(n), O_right-2(n) is stored in association with each code.
[0051]
In the memory, the output value O of the camera 1,1501_left-1(n), O_right-1(n) and the output value O of the camera 2 1502_left-2(n), O_right-2The value of (n) is stored in association with each code. Specifically, when the coded projection pattern has 3 codes assigned with 3 (strong), 2 (medium), and 1 (weak) as intensities, the left and right codes 3, 2 at each edge position , Or 1 corresponding to the output value of each camera is stored in the memory.
[0052]
With the above processing, the correspondence between each code and the output value of each camera in the standard state is measured. Note that the output value measurement processing of each camera in the standard state using the reference object can be performed in advance before measurement, and does not need to be repeated for each measurement.
[0053]
Next, a pattern is projected onto an actual measurement object (S102), and imaging (S115, S125) is performed with each camera that has performed the output value measurement process described above. Next, edges are extracted from the images captured by the cameras 1 and 1501 by edge detection processing (S116). Since the cameras 1 and 1501 have the same optical axis and the same viewpoint as the pattern projection apparatus, the edge position can be detected at substantially the same position as in the standard state described above. That is, for the captured image of the measurement object, the edge position in the standard state: En, the left and right code: C_left(n), C_right(n) can be made to correspond.
[0054]
However, the output value of the camera is changed by the texture such as the shape and pattern of the target object. Due to this change, the actual output values of the left and right cameras 1, 1501 at the edge position: En are, for example, O ′_left-1(n), O '_right-1(n). O '_left-1(n), O '_right-1Regarding the calculation of (n) (S117), the camera output value O in the standard state_left-1(n), O_right-1The same procedure as when (n) was calculated may be taken. At this time, the ratio O ′ / O between the camera output value in the standard state and the camera output value of the photographing data of the measurement object can be regarded as a kind of reflectance in each stripe on the target object. The ratio of the nth edge position to the camera output values at the left and right positions of En is expressed by the following equation (5).
[0055]
[Equation 5]
R_left(n) = O '_left-1(n) / O_left-1(n) ... (5) -L
R_right(n) = O '_right-1(n) / O_right-1(n) ... (5) -R
In the above formula, R_left(n), R_right(n) can be regarded as a reflection coefficient indicating the reflectance in each stripe on the target object.
[0056]
Using the above equation, the reflection coefficient Rn indicating the reflectance at each edge position of the captured images of the cameras 1 and 1501 is calculated (S118). Also, in the captured images of the cameras 2 and 1502, the same processing as in the case of the cameras 1 and 1501 is executed to detect the edge positions, and the edge positions of the cameras 2 and 1502 are output values at the left and right positions of En. Edge brightness as: O '_left-2(n), O '_right-2(n) is calculated (S127).
[0057]
Next, the inter-edge luminance information in the measured object pattern projection image of the camera 1 calculated in step S117, the inter-edge luminance information in the measured object pattern projection image of the camera 2 calculated in step S127, and further calculated in step S118. Using the reflection coefficient R, the association process based on the epipolar line (S103) is executed.
[0058]
Details of the association processing (S103) are shown in FIG. Output value O of camera 2,1502 in standard state₂Based on the predicted value O ′ of the output value in the measurement state of the camera 2_2-preThe calculation process (S201) is executed. Reflection coefficient calculated in S118 in the camera 1,1501: R_left(n), R_rightBy using (n) and the output value in the standard state of the cameras 2 and 1502, the value that the camera 2 and 1502 will output in the measurement state, that is, the camera predicted output value can be estimated. The predicted output value of the camera 2, 1502 at the left and right positions of En is the nth edge position is O ′._left-2-pre(n), O '_right-2-pre(n).
[0059]
When the outputs of the cameras 2 and 1502 have different inclinations in the standard state and the measurement state but have the linearity of the output, the predicted output value of the cameras 2 and 1502 is expressed by the following equation (6). Can be estimated.
[0060]
[Formula 6]
O '_left-2-pre(n) = R_left(n) × O_left-2(n) ... (6) -L
O '_right-2-pre(n) = R_right(n) × O_right-2(n) ... (6) -R
[0061]
According to the above equation, the predicted output value O ′ of the camera 2, 1502 at the left and right positions of En is determined as the nth edge position._left-2-pre(n), O '_right-2-pre(n) is calculated.
[0062]
Next, the predicted output value O ′ of the camera 2, 1502 obtained by the above-described processing in step S202._left-2-pre(n), O '_right-2-preThe actual output value O ′ in the measurement state of the camera 2 1502 having the value closest to (n)_left-2(n), O '_right-2(n) is searched, and based on the search edge based on the predicted output value, an association process is performed on the epipolar line corresponding to the measurement image of the camera 1. The correspondence is O 'on the epipolar line._left-2-pre(n), O '_right-2-preO 'approximately equal to (n)_left-1(n), O '_right-1This can be done by searching for an edge having (n). Thus, the cameras 1 and 1501 and the cameras 2 and 1502 do not have the same output characteristics, that is, O ′._{left (right) -1}(n) and O '_{left (right) -2}Even when (n) does not match, the camera 2 1502 can reliably detect the corresponding point. After the corresponding points are determined, the distance can be measured by a general triangulation principle.
[0063]
FIG. 18 shows an apparatus configuration for executing the processing flows of FIGS. The configuration of FIG. 18 is a block diagram showing a configuration in which the corresponding point search unit in the configuration shown in FIG. 1 and two luminance value memories corresponding to two cameras are extracted. The corresponding point search unit 1710 includes a reflection coefficient calculation unit 1711, an output value prediction unit 1712, and an association unit 1713, and luminance information between edges acquired based on the captured image of the camera 1 is stored in the output value storage unit 1720. The luminance information between the edges stored and acquired based on the captured image of the camera 2 is stored in the output value storage unit 1730.
[0064]
The output value storage unit 1720 and the output value storage unit 1730 store the output value of the captured image of each camera in the standard state and the measurement state.
[0065]
First, when the pattern is projected onto the reference object, the output values of the first camera and the second camera calculated by the calculation means are stored in the output value storage means 1720 and the output value storage means 1730, and then the measurement object The output value of the image captured by each camera by projecting the pattern is stored in the output value storage means 1720 and the output value storage means 1730.
[0066]
The reflection coefficient calculation means 1711 of the corresponding point search unit 1710 calculates the reflection coefficient by calculating each output value stored in the output value storage means 1720 by applying the above equation (5). Next, the output value predicting unit 1712 applies a calculation based on the calculated reflection coefficient and the output value stored in the output value storage unit 1730, that is, the output value of the second camera by applying the above-described equation (6). The output value predicting process for predicting is executed.
[0067]
Next, the associating unit 1713 calculates the predicted output value O ′ of the camera 2 calculated by the output value predicting unit 1712._left-2-pre(n), O '_right-2-pre(n) and the actual output value O ′ in the measurement state of the camera 1,1501 stored in the output value storage means 1720._left-1(n), O '_right-1The association process with (n) is performed. For the association, an association process on the epipolar line can be applied.
[0068]
Next, a configuration example in which the reflection coefficient Rn is used as one of the parameters at the time of encoding will be described. This will be described with reference to the flowcharts of FIGS. FIG. 19 is a processing flow based on the premise that there is a luminance value pattern approximated on the epipolar line, and FIG. 20 is a processing assuming both cases where there is a luminance value pattern approximated on the epipolar line and not. It is a flow.
[0069]
This will be described from the flow of FIG. The procedure similar to the processing flow in FIG. 16 is executed until the reflection coefficient Rn is calculated, and the camera 1 and the camera 2 acquire output values of the captured images of the cameras in the standard state and the measurement state, respectively. It is assumed that the reflection coefficient R is obtained based on the standard state and measurement state images.
[0070]
First, an edge group α having an approximate luminance value pattern is searched for on the epipolar line of the camera 2 for associating the images of the cameras 1 and 1501 with the images of the cameras 2 and 1502 (S301).
[0071]
For example, in the case of having output values approximated at two different edge positions a and b on the epipolar line of the cameras 2 and 1502, the left and right output values of the edge position a: O ′_left-2(n, a), O '_right-2(n, a) and left and right output values of edge position b: O ′_left-2(n, b), O '_right-2This is a case where (n, b) is an approximate relationship as shown in the following formula (7).
[0072]
[Expression 7]
O '_left-2(n, a) ≒ O '_left-2(n, b) (7) -L
O '_right-2(n, a) ≒ O '_right-2(n, b) (7) -R
[0073]
An edge group having such an approximate luminance pattern is searched on the epipolar line of the camera 2 1502. Note that, here, a description will be given as a process of extracting and comparing only two adjacent luminance values on the left and right (L, R) of the edge, but by selecting three or more adjacent luminance values, a pattern having an approximate relationship is selected. You may search.
[0074]
In step S301, the luminance value pattern [O ′ approximated on the epipolar line of the cameras 2 and 1502 is displayed._left-2(n, a), O '_right-2(n, a)], [O '_left-2(n, b), O '_right-2When a plurality of edge groups α (a, b...) of (n, b)]... are detected, in step S302, the luminance pattern [O ′ of the edge group α (a, b._left-2(n, a), O '_right-2(n, a)] or [O '_left-2(n, b), O '_right-2(n, b)] is executed to detect the edge group β (n1, n2,...) in the camera 1,1501 having a luminance pattern similar to [n, b)].
[0075]
The edge positions of the cameras 1 and 1501 corresponding to the plurality of edge groups α (a, b...) Approximated on the epipolar line of the cameras 2 and 1502 are the reflection coefficients R calculated in step S118 in FIG. Can be used to calculate. For example, the reflection coefficient Rn at the edge position En can be determined only from the reference state image of the cameras 1 and 1501, and the following equation (8) is calculated using the reflectance R (n) at each edge position.
[0076]
[Equation 8]
O_pre-left= R_left(n) × O_left-2(n) ... (8) -L
O_pre-right= R_right(n) × O_right-2(n) (8) -R
[0077]
The edge positions of the cameras 1, 1501 corresponding to a plurality of approximate edge groups α (a, b,...) On the epipolar line of the cameras 2, 1502 are n1, n2,. Is calculated.
[0078]
[Equation 9]
O_pre-left-n1= R_left(n1) × O_left-2(n1) (9) -Ln1
O_pre-right-n1= R_right(n1) × O_right-2(n1) (9) -Rn1
O_pre-left-n2= R_left(n2) × O_left-2(n2) (9) -Ln2
O_pre-right-n2= R_right(n2) × O_right-2(n2) (9) -Rn2
[0079]
Calculated by the above formula (O_pre-left-n1, O_pre-right-n1), (O_pre-left-n2, O_pre-right-n2) Is a set of outputs indicating candidates corresponding to the left and right output values of either of the edge positions a or b of the cameras 2 and 1502.
[0080]
Next, each of the edge groups β (n1, n2,...) In the camera 1 and each of the approximate patterns of the edge groups α (a, b,...) On the epipolar line on the camera 2 are compared. The combination with a small difference is set as the corresponding edge (S303).
[0081]
Specifically, one edge a: (O ′) of a plurality of edge groups α (a, b...) Approximated on the epipolar line of the camera 2 1502_left-2(n, a), O '_right-2(n, a)) is (O_pre-left-n1, O_pre-right-n1), (O_pre-left-n2, O_pre-right-n2), The corresponding point of the edge a: on the epipolar line of the camera 2 1502 can be determined. Similarly for edge b (O '_left-2(n, b), O '_right-2(n, b)) is (O_pre-left-n1, O_pre-right-n1), (O_pre-left-n2, O_pre-right-n2), The corresponding point of edge b: can be determined. As a comparison method, for example, the following equation (10) is calculated.
[0082]
[Expression 10]
Σ1 = (O_pre-left-n1 -O '_left-2(n, a))²+ (O_pre-right-n1 -O '_right-2(n, a))²
+ (O_pre-left-n2 -O '_left-2(n, b))²+ (O_pre-right-n2 -O '_right-2(n, b))²
Σ2 = (O_pre-left-n2 -O '_left-2(n, a))²+ (O_pre-right-n2 -O '_right-2(n, a))²
+ (O_pre-left-n1 -O '_left-2(n, b))²+ (O_pre-right-n1 -O '_right-2(n, b))²(10)
[0083]
By calculating the above equation (10) and adopting the smaller correspondence, the output of the camera 2, 1502 (O ′_left-2(n, a), O '_right-2(n, a)) as an output corresponding to the output of the camera 1,1501 (O_pre-left-n1, O_pre-right-n1), (O_pre-left-n2, O_pre-right-n2) Is selected, and the edge position having the selected output is set as the corresponding point in the captured image of the camera 1 at the edge position a.
[0084]
In this embodiment, a reflectance is added as an encoding parameter, and with this configuration, even when a luminance value pattern that approximates the captured image of the camera 2 1502 is output, it can be reliably associated. . After the corresponding points are determined, the distance can be measured by a general triangulation principle.
[0085]
Next, the flow of FIG. 20 will be described. FIG. 20 is a processing flow assuming both cases where there is a luminance value pattern approximated on the epipolar line and where there is no luminance value pattern. The procedure similar to the processing flow in FIG. 16 is executed until the reflection coefficient Rn is calculated, and the camera 1 and the camera 2 acquire output values of the captured images of the cameras in the standard state and the measurement state, respectively. It is assumed that the reflection coefficient R is obtained based on the standard state and measurement state images.
[0086]
First, an edge group α having an approximate luminance value pattern is searched for on the epipolar line of the camera 2 for associating the images of the cameras 1 and 1501 with the images of the cameras 2 and 1502 (S401).
[0087]
If there is no edge group α having a luminance value pattern to be approximated (No in S402), the process proceeds to step S406, and the process of associating the measurement image of the camera 1 and the measurement image of the camera 2 on the epipolar line is performed. Executed. Since the epipolar line has no approximate pattern, the matching process can be executed without any problem.
[0088]
On the other hand, when there is an edge group α having an approximate luminance value pattern on the epipolar line of the camera 2 (Yes in S402), the processes of steps S403 to S405 are executed. The processes in steps S403 to S405 are the same as those in steps S301 to S303 in FIG.
[0089]
In the process of FIG. 20, when there is no luminance value pattern approximated on the epipolar line, normal association processing is executed, and prediction using the reflectance is performed only when there is a luminance value pattern approximated on the epipolar line. Value calculation is executed to accurately associate a plurality of approximate patterns.
[0090]
FIG. 21 shows an apparatus configuration for executing the processing flows of FIGS. 19 and 20. The configuration of FIG. 21 is a block diagram showing a configuration in which the corresponding point search unit in the configuration shown in FIG. 1 and two luminance value memories corresponding to two cameras are extracted. The corresponding point search unit 1710 includes a reflection coefficient calculation unit 1711, an output value prediction unit 1712, an association unit 1713, and an edge search unit 1714. The luminance information between edges acquired based on the captured image of the camera 1 is output. Luminance information between edges which is stored in the value storage unit 1720 and acquired based on the captured image of the camera 2 is stored in the output value storage unit 1730.
[0091]
The output value storage unit 1720 and the output value storage unit 1730 store the output value of the captured image of each camera in the standard state and the measurement state.
[0092]
First, when the pattern is projected onto the reference object, the output values of the first camera and the second camera calculated by the calculation means are stored in the output value storage means 1720 and the output value storage means 1730, and then the measurement object The output value of the image captured by each camera by projecting the pattern is stored in the output value storage means 1720 and the output value storage means 1730.
[0093]
Next, the edge search unit 1714 projects a pattern onto the measurement object stored in the output value storage unit 1730 and approximates a plurality of edge groups α (a, a on the epipolar line from the image captured by the camera 2. b ..) is detected. That is, an edge that satisfies the relationship of the above-described equation (7) is detected. When the edge group α (a, b,...) Having a similar pattern is detected, the luminance pattern [O ′ of the edge group α (a, b,._left-2(n, a), O '_right-2(n, a)] or [O '_left-2(n, b), O '_right-2(n, b)] is executed to detect the edge group β (n1, n2,...) in the measurement state image of the camera 1 having a luminance pattern similar to [n, b)].
[0094]
In the detection processing of the edge group β (n1, n2,...) Of the camera 1, first, the reflection coefficient calculation means 1711 calculates each output value stored in the output value storage means 1720 by applying the above-described equation (5). The reflection coefficient is calculated. Next, the output value predicting unit 1712 calculates output values for the edge positions n1 and n2 of the edge group β (n1, n2,...) Based on the above-described equations (8) and (9).
[0095]
Next, in the association unit 1713, each of the approximate patterns of each of the edge groups β (n1, n2,...) In the camera 1 and the edge groups α (a, b,...) On the epipolar line on the camera 2 is displayed. A comparison is executed, and a process using a combination having a smaller difference as a corresponding edge is executed.
[0096]
As described above, in the apparatus and method of the present invention, the inter-edge luminance of images captured by a plurality of cameras by projecting a pattern onto a reference object is obtained as an output value, and a code indicating an irradiation pattern and each camera The correspondence with the output value is stored in the memory, and then pattern irradiation is executed on the measurement object, edge extraction and edge-to-edge luminance calculation are executed, and the measurement object is measured based on the luminance information acquired by the camera 1. The reflection coefficient is obtained, the predicted output of the camera 2 is calculated based on the obtained reflection coefficient, and the edge having the closest value from the measured object pattern irradiation image of the camera 2 is calculated based on the calculated predicted output of the camera 2 Since the distance information is acquired by extracting and associating with the measurement value of the camera 1, it does not depend on the texture of the measurement object or the output characteristics of the camera, Correspondence processing between images taken by the camera can be performed accurately, it is possible to obtain an accurate distance data.
[0097]
In the above-described embodiment, only one second camera is shown. However, a plurality of second cameras (relative to the first camera) are arranged on an optical axis different from that of the first camera. The measurement object may be photographed by the second camera at various different angles, and the distance information obtained based on the projection pattern photographed by each of the plurality of second cameras may be combined to obtain a three-dimensional image. Good.
[0098]
The present invention has been described in detail above with reference to specific embodiments. However, it is obvious that those skilled in the art can make modifications and substitutions of the embodiments without departing from the gist of the present invention. In other words, the present invention has been disclosed in the form of exemplification, and should not be interpreted in a limited manner. In order to determine the gist of the present invention, the claims section described at the beginning should be considered.
[0099]
【The invention's effect】
As described above, the three-dimensional shape measurement apparatus and the three-dimensional shape measurement method of the present invention have variations in output characteristics, or the environment and time-dependent changes even though the output characteristics were originally gathered. Re-encoding can be performed even when a group of cameras with variations is used, and by reducing the dependence on the texture of the target object, erroneous correspondence is less likely to occur, and highly reliable three-dimensional shape measurement It becomes possible to do.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration example of a three-dimensional shape measuring apparatus according to the present invention.
FIG. 2 is a block diagram illustrating a camera configuration example of the three-dimensional shape measurement apparatus according to the present invention.
FIG. 3 is a diagram illustrating an imaging configuration of a three-dimensional shape measuring apparatus according to the present invention.
FIG. 4 is a diagram showing a processing flow of the three-dimensional shape measuring apparatus of the present invention.
FIG. 5 is a diagram illustrating an example of encoding a projection pattern of the three-dimensional shape measuring apparatus according to the present invention.
FIG. 6 is a diagram showing an example of a photographing configuration of the three-dimensional shape measuring apparatus according to the present invention.
FIG. 7 is a diagram showing an example of a projection pattern of the three-dimensional shape measuring apparatus according to the present invention.
FIG. 8 is a diagram showing an example of a slit pattern photographed by the camera 1 of the three-dimensional shape measuring apparatus of the present invention.
FIG. 9 is a diagram showing an example of a slit pattern photographed by the camera 2 of the three-dimensional shape measuring apparatus of the present invention.
FIG. 10 is a diagram showing an example of a newly coded slit pattern in the three-dimensional shape measuring apparatus of the present invention.
FIG. 11 is a diagram showing a distance calculation method by a spatial encoding method of the three-dimensional shape measurement apparatus of the present invention.
FIG. 12 is a diagram showing an example of a slit pattern photographed by the camera 3 of the three-dimensional shape measuring apparatus of the present invention.
FIG. 13 is a diagram showing a pattern image photographing configuration using two imaging means in the three-dimensional shape measuring apparatus of the present invention.
FIG. 14 is a diagram for explaining a specific example of a pattern and a measurement object in the three-dimensional shape measurement apparatus of the present invention.
FIG. 15 is a diagram illustrating a specific example of a light projection pattern image and an imaging pattern image in the three-dimensional shape measurement apparatus of the present invention.
FIG. 16 is a process flow (part 1) for executing the association process in consideration of the output characteristics of the camera in the three-dimensional shape measurement apparatus of the present invention.
FIG. 17 is a process flow (part 2) for executing the association process in consideration of the output characteristics of the camera in the three-dimensional shape measurement apparatus of the present invention.
FIG. 18 is a block diagram showing an apparatus configuration (Example 1) for executing association processing in consideration of output characteristics of a camera in the three-dimensional shape measurement apparatus of the present invention.
FIG. 19 is a processing flow (part 1) for executing processing using a reflection coefficient as a parameter in encoding in consideration of the output characteristics of the camera in the three-dimensional shape measurement apparatus of the present invention.
FIG. 20 is a processing flow (part 2) for executing processing using a reflection coefficient as a parameter in encoding in consideration of the output characteristics of the camera in the three-dimensional shape measurement apparatus of the present invention.
FIG. 21 is a block diagram showing an apparatus configuration (example 2) for executing association processing in consideration of output characteristics of a camera in the three-dimensional shape measurement apparatus of the present invention.
[Explanation of symbols]
101 Camera 1
102 Camera 2
103 Camera 3
104 Floodlight
105 half mirror
106 Light source
107 Mask pattern
108 Strength pattern
109 prism
121, 123, 125 Luminance value memory
122, 124, 126 Pattern image memory
127 frame memory
128 area division
129 Recoding section
130,131 code decoding unit
133 Integration unit of distance information
134 3D memory
301 prism
302,304 Transmission filter
303,305 Imaging device
601 602 603 Camera
604 Floodlight
605 wall
606 board
801, 901 shadow area
1301 Measurement target
1302 Projector
1303, 1304 Imaging means
1501 Camera 1
1502 Camera 2
1710 Corresponding point search unit
1711 Reflection coefficient calculation means
1712 Output value prediction means
1713 Association means
1714 Edge search means
1720 Output value storage means
1730 Output value storage means

Claims (8)

A light projecting means for projecting a coded slit pattern given by using a light source, and a first camera having substantially the same viewpoint and the same optical axis as the pattern projected by the light projecting means; In the three-dimensional shape measuring apparatus composed of a second camera arranged on an optical axis different from that of the first camera,
Output value storage means for storing output value information between edges of a pattern projection image captured by the first and second cameras, wherein a pattern is projected on a reference object and a pattern is measured on a measurement object. Output value storage means for storing output values of the first camera and the second camera when projected,
Reflection coefficient calculation means for calculating a reflection coefficient for the measurement object based on output values of the first camera when a pattern is projected onto a reference object and when a pattern is projected onto a measurement object ,
Output value prediction for calculating the output prediction value of the second camera by calculating the reflection coefficient and the second camera output value when a pattern is projected onto the reference object stored in the storage means Means,
An output value that is the closest to the predicted output value of the second camera calculated by the output value predicting means from the output value information between edges of the pattern projection image that is projected onto the measurement object and captured by the second camera And an associating means for selecting an edge having a selected point as a corresponding point for the output value of the first camera;
A three-dimensional shape measuring apparatus comprising: The three-dimensional shape measurement apparatus further includes:
Edge search means for detecting a similar pattern in the output value of the second camera when a pattern is projected onto a measurement object;
The output value predicting means includes
When a similar pattern is detected by the edge search means, the output value of the second camera when the pattern is projected onto the reference object stored in the output value storage means and the reflection coefficient calculation means are calculated. Calculating an output predicted value of each edge in the second camera based on the reflection coefficient,
The association means includes
The three-dimensional shape measurement apparatus according to claim 1, further comprising: a process of selecting a corresponding edge of the second camera based on a comparison with the similar pattern. The light emitted from the light projecting means and the light incident on the first camera are separated by a beam splitter, and the light projecting means and the first camera are optically coaxial with each other. The three-dimensional shape measuring apparatus according to claim 1, wherein the three-dimensional shape measuring apparatus is arranged to be The light projecting unit includes a light source that generates light in an invisible region, and the first camera includes a filter that transmits light in the invisible region and a filter that blocks light in the invisible region. The three-dimensional shape measuring apparatus described. The second camera is composed of a plurality of cameras that capture the measurement object at different angles, and combines the distance information obtained based on the projection pattern captured by each of the plurality of second cameras to obtain a three-dimensional image. The three-dimensional shape measuring apparatus according to claim 1, wherein the three-dimensional shape measuring apparatus is configured to obtain an image. A light projecting step of projecting a given slit pattern using a light source;
A first camera having substantially the same viewpoint and optical axis as the pattern projected by the projecting step, and a second camera disposed on an optical axis different from the first camera, An output value storage step of storing output value information between edges in the captured pattern image when the pattern is projected onto the reference object and when the pattern is projected onto the measurement object;
A reflection coefficient calculation step of calculating a reflection coefficient for the measurement object based on output values of the first camera when a pattern is projected onto a reference object and when a pattern is projected onto a measurement object; ,
Output value prediction for calculating the output prediction value of the second camera by calculating the reflection coefficient and the second camera output value when a pattern is projected onto the reference object stored in the storage means Steps,
An output value that is the closest to the predicted output value of the second camera calculated by the output value predicting means from the output value information between edges of the pattern projection image that is projected onto the measurement object and captured by the second camera And an associating step of extracting an edge having a selected point as a corresponding point for the output value of the first camera;
A three-dimensional shape measuring method characterized by comprising: The three-dimensional shape measurement method further includes:
An edge search step of detecting a similar pattern in the output value of the second camera when a pattern is projected onto a measurement object;
The output value prediction step includes:
When a similar pattern is detected in the edge search step, the output value of the second camera when the pattern is projected onto the reference object stored in the output value storage means and the reflection coefficient calculation means are calculated. Calculating an output predicted value of each edge in the second camera based on the reflection coefficient,
The three-dimensional shape measurement method according to claim 6, wherein the associating step performs a process of selecting a corresponding edge of the second camera based on a comparison with the similar pattern. In the three-dimensional shape measurement method,
The second camera is composed of a plurality of cameras that capture the measurement object at different angles, and combines the distance information obtained based on the projection pattern captured by each of the plurality of second cameras to obtain a three-dimensional image. The three-dimensional shape measuring method according to claim 6, further comprising a step of acquiring an image.

Images