JP2014059164A - Shape measurement device and shape measurement method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a shape measurement device and a shape measurement method that enable a measurement of a measurement target object in contactless and with higher accuracy.SOLUTION: A shape measurement device 1 according to the present invention comprises: a light source array 11; a grating plate 12 that forms a first-dimensional grating; a camera 13 that photographs a measurement target object 21 having a shadow of the grating due to the first-dimensional grating projected by a lighting of the light source in the light source array 11; and a control unit 50 that calculates a phase of the shadow of the grating projected onto the measurement target object 21 by the light source or a phase of a moire fringe consisting of the first-dimensional grating and the shadow of the grating, respectively, for each pixel of the camera 13 and calculates a three-dimensional space coordinate relating to the measurement target object 21 on the basis of the calculated phases. The grating plate 12 is arranged within a field of view of the camera 13. Regarding a photographing image by the camera 13, a predetermined number of pixels are allocated in a predefined range for calculating the phase of the shadow of the grating or the phase of the moire fringe.

Description

  The present invention relates to a shape measuring apparatus and a shape measuring method that can perform the shape and displacement of an object to be measured in a non-contact and highly accurate manner.

  As a method for measuring the three-dimensional shape of the measurement target object, an optical method capable of high-speed response without contact is advantageous. There are many non-contact optical shape measurement methods, but the laser scanning method, light cutting method, and grid projection method are often used. The laser scanning method is used for measuring large objects such as remains and buildings. It is known that the light cutting method and the grid projection method are used for measuring relatively small objects such as inspection of industrial parts.

  Among the grid projection methods, there is a method using a phase shift method as a method for non-contact and three-dimensional measurement of the shape of a measurement target object such as an object or a human body. In the phase shift method, a lattice image and an interference fringe image are sequentially photographed by one camera while changing the phase, and a phase distribution of the lattice is obtained based on a plurality of lattice images and interference fringe images whose phases are changed. Is.

  As a method using the phase shift method, for example, using a camera and a projector, in order to perform highly accurate shape measurement without being affected by lens aberration of these cameras or projectors, a reference plate on which a grating is drawn is used. Rather than calculating the lens center coordinates of the camera or projector from the image, the optical path through which the line of sight for each pixel of the camera passes from the two-dimensional grid fixed to each reference plane in the plurality of reference planes, and the light projected from the projector There are known a shape measuring method and a shape measuring apparatus that calculate all of the optical paths and calculate spatial coordinates as intersections of these optical paths (see, for example, Patent Document 1).

  There is another technique for measuring the shape of an object based on a plurality of reference planes using a camera and a projector. For example, a plurality of reference planes having X and Y axis planes orthogonal to each other are set at predetermined intervals in the normal direction of the reference plane (that is, the Z axis direction perpendicular to the X and Y axis planes) and measured. A line of sight of a camera that shoots a point S on the object in order to determine the coordinates of the point S on the surface of the object by arranging the power object between the reference surfaces located at both ends of the plurality of reference surfaces And a point where each of the light rays from the projector passing through the point S on the object intersects the plurality of reference planes, and a line formed by the intersecting points on the line of sight of the camera and the intersecting point on the light rays from the projector A technique is known in which an intersection point with a straight line is obtained, two reference planes closest to the point are selected from the Z coordinate of the intersection point, and the shape of the object is measured using the two selected reference planes. (For example, refer to Patent Document 2).

  On the other hand, as yet another technique for measuring the shape of an object based on a plurality of reference planes using a camera and a projector, when a reference plane having X and Y axis planes is translated by a minute amount in the normal direction By using a predetermined two-dimensional pattern or a spatial division pattern for the plurality of reference planes, even if the luminance distribution of the spatial division pattern projected onto the object is not cosine-like, the spatial division pattern projected onto the object 2. Description of the Related Art A shape measuring apparatus that can accurately measure a shape even when the pitch is unequal is known (see, for example, Patent Document 3).

  In addition, phase shift is performed by switching light sources by using a projection device (projector) having three or more light sources arranged in a row for projecting a lattice pattern and a grating plate that passes the projection light to form a lattice pattern. (Hereinafter, such a method is referred to as a “light source switching phase shift method”), and using a table or an approximate expression that correlates the phase of each lattice pattern with the three-dimensional spatial coordinates on a one-to-one basis. A shape measuring device that determines the shape of a measurement target is disclosed (for example, see Patent Document 4).

  By applying the shape measuring device using the phase shift method as described above, it is possible to measure the time change of the three-dimensional shape of the measurement target object. For example, a shape measuring apparatus based on the phase shift method can be used for measurement analysis of human body motion, machine vibration, vehicle collision deformation, and airbag high-speed deformation.

  On the other hand, among the lattice projection methods, there is a method using a shadow moire method as a method for inspecting a warp, a bend, a twist, a scratch, or the like for a semiconductor device, a printed circuit board, or a conveyer used for parts conveyance or the like ( For example, see Patent Document 5).

Japanese Patent No. 2913021 Japanese Patent No. 3446020 Japanese Patent No. 4873485 JP 2011-242178 A JP 2009-246036 A

  In the shape measurement of the grid projection method disclosed in Patent Documents 1 to 3, some moving device for moving the grid to be projected is required. On the other hand, in the shape measurement of the lattice projection method (moire topography) based on the phase shift method disclosed in Patent Document 4, the phase shift can be performed by switching the lighting of each of a large number of light sources. Is possible. FIG. 14 shows a schematic diagram of a lattice projection method (moire topography) apparatus in the prior art. In the lattice projection apparatus shown in FIG. 14, the distance d between the projector and the lattice is the distance between the lattice and the measurement object (z−d) so that the lattice is not reflected during imaging by the camera. It is necessary to arrange so as to be much shorter than). For this reason, the pitch of the grid shadow becomes large at the position of the measurement target object, making it difficult to improve measurement accuracy.

  On the other hand, the shadow moiré method disclosed in Patent Document 5 is effective for applications that require highly accurate measurement. FIG. 15 shows a schematic diagram of a shape measuring apparatus using the shadow moiré method in the prior art. As shown in FIG. 15, in the case of shape measurement by the shadow moire method, it is advantageous when the distance d between the light source and the grating is much longer than the distance z between the light source and the measurement target object. . In FIG. 15, the pitch of the grating on the object to be measured is the same as that shown in FIG. 14, but in the case of the grating projection method shown in FIG. 14, a fine grating must be prepared, and the grating pitch is the light source. If the size is smaller than, the projected grating becomes blurred and the contrast becomes poor, and it becomes difficult to project a fine grating. In this regard, in the case of the shadow moire method shown in FIG. 15, even if the pitch of the grating is coarse, a fine grating can be easily projected onto the object, and the resolution of shape measurement can be improved. However, since the camera captures the measurement target object through the grid, the grid itself blocks the field of view of the camera, and a portion of the measurement target object that cannot be captured by the camera occurs. If the pitch of the grid is much smaller than the pixel pitch of the camera, it can be said that there is almost no effect, but in the inspection of small defects, it is a big practical problem that there is a part that can not be photographed by the camera due to the grid It becomes. In order to solve this, the technique assumed from the prior art involves photographing with a camera while moving the grating at a high frequency. However, this technique requires a mechanism for operating the grating with high accuracy and high frequency, resulting in an increase in apparatus cost.

  That is, in the conventional light source switching phase shift method, there is a one-dimensional grating between the light source and the measurement target object, and the shadow of the grating projected on the measurement target object is photographed with a camera, and the phase of the grating is determined from this photographed image. The shape was measured by analysis. At this time, when the image was captured by the camera, only a one-dimensional lattice shadow on the object was photographed without capturing the lattice, and the shape was obtained by analyzing the phase of the lattice shadow. Therefore, in order not to capture the grating of the grating plate on the camera, the arrangement is such that the distance between the grating and the measurement target object is much larger than the distance between the light source and the grating. However, the larger the distance between the grid and the measurement target object, the larger the grid shadow will be, so the pitch of the grid shadow projected onto the measurement target object is large, and the accuracy is reduced accordingly. was there. Here, in order to make the shadow of the lattice as fine as possible, it is conceivable to use a so-called shadow moire method in which a lattice plate is disposed near the measurement target object. However, in this case, since the grating of the grating plate is usually reflected in the camera in addition to the shadow of the grating, there is no effective data necessary for phase analysis in the portion where the grating is reflected, and further contrivance is necessary. Become.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a shape measuring device and a shape measuring method that can measure a measurement target object in a non-contact manner with higher accuracy. It is in.

In the present invention, a new shape that enables precise shape measurement by a new measurement method that combines the advantages of the shadow moire method and the phase shift method (hereinafter referred to as “phase shift shadow measurement method”). Configure the measuring device. Regarding the phase shift shadow measurement method, a technique for obtaining three-dimensional spatial coordinates of a measurement target object based on the phase of the grid shadow projected onto the measurement target object by a light source, and a one-dimensional grid (hereinafter simply referred to as “lattice”) on a grid plate And a technique for obtaining three-dimensional spatial coordinates related to the measurement target object based on the phase of the moire fringes formed by the shadow of the lattice. In the specification of the present application, the term “moire fringes” refers to a striped pattern composed of a one-dimensional lattice on a lattice plate and a shadow of the lattice.

  That is, the shape measuring apparatus of the present invention is a shape measuring apparatus that measures the shape of a measurement target object in a non-contact manner, from a light source array composed of at least one point light source or a linear light source, and straight lines arranged at equal intervals. A grating plate that forms a one-dimensional grating by a light transmission region, a camera that images a measurement target object on which a shadow of the grating by the one-dimensional grating is projected by turning on the light source in the light source array, and the measurement target by the light source Calculate the phase of the shadow of the grid projected on the object or the phase of the moire fringe composed of the shadow of the one-dimensional grid and the grid for each pixel of the camera, and relates to the measurement target object based on the calculated phase A control unit for obtaining three-dimensional spatial coordinates, the lattice plate is disposed in the field of view of the camera, When calculating the phase of the shadow of the child, a predetermined number of pixels is assigned to a predetermined range (preferably, a range of 1 lattice pitch or 1/2 lattice pitch) in an area other than the lattice portion of the lattice plate, and the primary When calculating the phase of the moiré fringe composed of the original grid and the shadow of the grid, a predetermined pixel (preferably a range of one grid pitch) including a set of the one-dimensional grid and the shadow of the grid The control unit calculates a phase of a shadow of the lattice or a phase of a moire fringe composed of the shadow of the one-dimensional lattice and the lattice based on the pixel value of the predetermined number of pixels. It is characterized by that.

  In the shape measuring apparatus of the present invention, the light source array includes at least three point light sources or linear light sources arranged at equal intervals corresponding to a plurality of types of phase shifts, and the control unit includes the at least 3 Among the two light sources, a light source that performs phase shift of the shadow of the projected grid is determined, and each of the determined light sources is sequentially turned on, and the shadow of the grid or the moire fringe projected on the measurement target object respectively. Means for controlling the camera to take a picture, and calculating a phase of a shadow of the grating or a phase of the moire fringe projected on the measurement target object by each of the determined light sources for each pixel of the camera; It is characterized by.

  In the shape measuring apparatus of the present invention, the light source array includes one point light source or a linear light source, and the shape measuring apparatus is predetermined on a plane parallel to the lattice plate with respect to the measurement target object. The control unit controls the camera to capture shadows of the lattice or the moire fringes projected on the measurement target object in synchronization with the predetermined speed, The apparatus further comprises means for calculating the phase of the shadow of the grid projected on the measurement target object or the phase of the moire fringe for each pixel of the camera. The “arrangement relationship that synchronizes at a predetermined speed” does not necessarily have to be a constant speed, and refers to an arrangement relationship that is synchronized even when the speed is not stable.

  In the shape measuring apparatus of the present invention, the light source array includes at least three point light sources or linear light sources arranged at equal intervals corresponding to a plurality of types of phase shifts. The control unit is arranged to move relative to the target object at a predetermined speed on a plane parallel to the lattice plate, and the control unit is synchronized with the predetermined speed and projects the projection among the at least three light sources. A light source for phase shift of the shadow of the grating is determined, and the camera is controlled so that each of the determined light sources is sequentially turned on and the shadow of the grating or the moire fringe projected on the measurement target object is photographed. The phase of the shadow of the grating or the phase of the moire fringe projected on the measurement target object by each of the determined light sources is calculated for each pixel of the camera. Characterized in that it has a means that.

  Further, in the shape measuring apparatus of the present invention, the camera comprises a plurality of cameras for photographing a measurement target object onto which a shadow of the lattice by the one-dimensional lattice is projected by turning on a light source in the light source array with different lines of sight, The control unit controls the plurality of cameras so as to photograph the shadows of the lattice or the moire fringes projected on the measurement target object, respectively, and based on the images photographed by the plurality of cameras. The phase of the grid shadow or the phase of the moire fringe formed of the grid shadow projected onto the measurement target object is calculated for each pixel of the camera.

  In the shape measuring apparatus of the present invention, the control unit may perform the measurement based on a table or an approximate expression in which the phase of the one-dimensional lattice and the three-dimensional spatial coordinates prepared in advance for each pixel are associated one-to-one. It has the means which calculates | requires the three-dimensional space coordinate regarding a target object, It is characterized by the above-mentioned.

  In the shape measuring apparatus of the present invention, the number N of pixels that are parallel to the surface of the lattice plate and continuous in the direction perpendicular to the straight line of the light transmission region of the one-dimensional lattice with respect to the pixel row at the time of photographing by the camera N is an integer greater than or equal to 1) corresponding to one pitch of a one-dimensional lattice in the lattice plate, and has a means for calculating the luminance value of the moire fringes from the average value of the luminance of the continuous N pixels. And When N = 1, moire fringes can be taken directly.

  Further, in the shape measuring apparatus of the present invention, the camera comprises a plurality of cameras for photographing a measurement target object onto which a shadow of the lattice by the one-dimensional lattice is projected by turning on a light source in the light source array with different lines of sight, The number N of pixels that are parallel to the plane of the grid plate and perpendicular to the straight line of the light transmission region of the one-dimensional grid (N is an integer equal to or greater than 1) with respect to the pixel rows in the measurement range of each of the plurality of cameras Corresponding to ½ pitch of the one-dimensional lattice in the lattice plate, and at the first pixel position in the number N of consecutive pixels, the shadow of the lattice or the It has a means to calculate the brightness | luminance value of a moire fringe, It is characterized by the above-mentioned.

  Further, in the shape measuring apparatus of the present invention, the line of sight of the camera has a plurality of pixels in which the half period of the phase (Φ) of the shadow of the grid projected on the measurement target object in an area other than the grid portion of the grid plate is The control unit has means for calculating the phase of the shadow of the grid from each luminance value obtained by phase shifting by the number of pixels (N). It is characterized by having. Note that when N = 4, the phase Φ of the shadow of the grating is expressed by the following equation (22).

  Further, in the shape measuring apparatus of the present invention, a part of the one-dimensional lattice is transparent, and the line of sight of the camera is arranged so as to photograph a shadow of the lattice from the transparent portion. To do. It is particularly effective when the object to be measured moves because only a narrow range needs to be transparent.

  Furthermore, the shape measuring method of the present invention includes a light source array composed of at least one point light source or a linear light source, a lattice plate that forms a one-dimensional lattice with light transmission regions composed of straight lines arranged at equal intervals, and the light source array. A shape measurement method for measuring the shape of a measurement target object in a non-contact manner by a shape measurement device including a camera that captures a measurement target object onto which a shadow of the lattice formed by the one-dimensional lattice is projected by turning on a light source In the control procedure of the control unit, when calculating the phase of the shadow of the lattice with respect to the image captured by the camera, a predetermined range (preferably, one lattice pitch) is used in an area other than the lattice portion of the lattice plate. Alternatively, a predetermined number of pixels is assigned to a range of 1/2 lattice pitch), and the phase of the moire fringes formed by the one-dimensional lattice and the shadow of the lattice is determined. A step of setting a measurement range by assigning a predetermined number of pixels to a predetermined range (preferably, a range of one grid pitch) including a set of the one-dimensional grid and a shadow of the grid, Controlling the photographing of the camera so as to photograph in the measurement range an object to be measured on which the shadow of the lattice by the one-dimensional lattice is projected by turning on the light source in the light source array; and the measurement range from the photographed image The luminance of the pixel is calculated from the pixel value, and the phase of the shadow of the lattice projected onto the measurement target object by the light source from the luminance or the phase of the moire fringe composed of the shadow of the one-dimensional lattice and the lattice is calculated. And calculating each three-dimensional space coordinate related to the measurement target object based on the calculated phase. And features.

  According to the present invention, the three-dimensional shape of the measurement target object can be measured with higher accuracy.

It is a figure which shows the structure of the shape measuring apparatus of one Embodiment by this invention. It is a block diagram of the control unit in the shape measuring device of one embodiment by the present invention. It is a figure which shows an example of the operation | movement flow in the shape measuring apparatus of one Embodiment by this invention. (A), (b) relates to a shape measuring apparatus according to an embodiment of the present invention, when measuring the shape of an object to be measured by a light source array composed of a plurality of light sources, a lattice plate, and one camera. It is a figure which shows the structural example and its operation principle. (A), (b) is a figure which shows an example of the relationship between the measurement pixel row | line | column in the measurement range of the camera in the light source switching phase shift shadow grating method, a grating | lattice, and a projection grating | lattice regarding the shape measuring apparatus of one Embodiment by this invention. It is. (A), (b) is another example of the relationship between the measurement pixel column in the measurement range of the camera in the light source switching phase shift shadow grid method, and the grid and the projection grid in the shape measurement apparatus according to the embodiment of the present invention. FIG. The figure which shows the example of a structure at the time of measuring the shape of a measuring object object with the light source array which consists of several light sources, a grating | lattice plate, and two cameras regarding the shape measuring apparatus of one Embodiment by this invention, and its operation principle. It is. (A), (b), (c), and (d) relate to the shape measurement apparatus 1 according to an embodiment of the present invention, and the relationship between the measurement pixel array, the grid, and the projection grid in the measurement range of two cameras. It is a figure which shows an example. (A), (b) is a figure which shows an example of the relationship between the measurement pixel row | line | column in the measurement range of the camera in the light source switching phase shift shadow moire method, a grating | lattice, and a projection grating | lattice regarding the shape measuring apparatus of one Embodiment by this invention. It is. (A), (b) relates to a shape measuring apparatus according to an embodiment of the present invention by a light source array configured as one line light source (for example, a linear LED light source), a grating plate, and one camera. It is a figure which shows the structural example at the time of measuring the shape of a measurement object, and its operation principle. (A), (b) is projected by one line light source (for example, a linear LED light source) in the case where the shape measuring apparatus according to the embodiment of the present invention is configured to have a relative movement with respect to an object to be measured. 5A is a top view illustrating an example of a relationship between a measurement pixel row in a measurement range of a camera that captures a shadow of a grid, a grid, and a projection grid, and FIG. ) Is an example of photographing a half pitch of a lattice. (A), (b), (c), (d) is an example of the relationship between the measurement pixel row, the grid, and the projection grid in the measurement range of two cameras in the shape measuring apparatus according to the embodiment of the present invention. FIG. (A), (b) is the relationship between the measurement pixel sequence in the measurement range of the camera, the grid, and the projection grid in the configuration in which there is a relative movement with respect to the measurement target object in the shape measurement apparatus according to an embodiment of the present invention. It is a figure which shows an example. It is the schematic of the grating | lattice projection method (moire topography) apparatus in a prior art. It is the schematic of the shape measuring device by the shadow moire method in a prior art. It is the schematic of the conventional grating | lattice projection method. (A), (b), (c) is a figure which shows an example of the brightness | luminance and phase distribution of the grating | lattice (interference fringe) which concern on the shape measuring apparatus of one Embodiment by this invention. It is a figure which shows the relationship between the phase shift amount and luminance which concern on the shape measuring apparatus of one Embodiment by this invention. It is a figure which shows the principle which performs shape measurement by the calibration process by the total space table formation technique using many reference planes concerning the shape measurement apparatus of one Embodiment by this invention. It is a figure which shows the relationship between the phase of a certain 1 pixel of a camera, and a three-dimensional space coordinate in the total space table formation technique which concerns on the shape measuring apparatus of one Embodiment by this invention. It is a figure which shows the relationship between the phase of 1 pixel with a camera, and the height (measurement position) of a measurement target object in the total space table formation method using many reference planes concerning the shape measuring apparatus of one Embodiment by this invention. . It is a figure which shows the structure of the shape measuring device by the light source switching phase shift method concerning the shape measuring device of one Embodiment by this invention, and its operation principle. (A), (b) is a figure which shows the example of arrangement | positioning of the light source array and grating | lattice plate which concern on the shape measuring device of one Embodiment by this invention. It is a figure which shows the principle of operation regarding the shape measuring apparatus of one Embodiment by this invention.

  Hereinafter, a shape measuring apparatus according to an embodiment of the present invention will be described with reference to the drawings.

〔Device configuration〕
FIG. 1 is a diagram showing a configuration of a shape measuring apparatus 1 according to an embodiment of the present invention. Shape measuring apparatus 1 of this embodiment is an apparatus for measuring a three-dimensional shape of the measurement object 21, five light sources L _-2 arranged in equal intervals and a _{line, L} -1, _{L 0,} L ₁ and It includes a light source array 11 composed of L _2, the grid plate 12 having a grating surface comprising a one-dimensional grating, a camera 13, and a control unit 50.

X-axis, Y-axis and made of Z-axis coordinate points x, y, in order to define the three-dimensional space by z, 5 single light source _L -2 constituting the light source array _{11, L -1, L 0,} L 1 and The position of one of the light sources of L ₂ (the position of L _{0 in} the example shown in FIG. 1) is the origin O (ie, x = 0, y = 0, z = 0), and each of the lattice plates 12 The arrangement direction of the light transmission region 12b is the X axis, the straight line of the light transmission region 12b is the Y axis, and the normal direction of the lattice plane of the lattice plate 12 is the Z axis. The measurement target object 21 is arranged in the Z-axis direction. Note that the position of the origin O can also be defined as the center position between the light sources at both ends in the light source array 11, but here the center position of the five light sources is defined as the origin O.

Light source array 11 includes five light sources _{L -2,} L _-1, has a configuration that is linearly arranged to L 0, _{L 1} and _{L 2} at regular intervals. In FIG. 1, an example of five light sources arranged at equal intervals in the X-axis direction as the light source array 11 will be representatively described. However, as described later, one light source can be used depending on the application. When the light source switching phase shift method is used, at least three light sources may be used. Moreover, the light emitting diode (LED) is used as a light source, and each LED can be comprised as a line light source or a point light source. That is, since the output of the LED chip is small, a plurality of LED chips are arranged in the Y-axis direction to be a line light source that is parallel to the Y-axis, and each line light source is arranged at equal intervals in the X-axis direction. _{-2, L -1,} it is possible to configure the L 0, _{L 1} and _{L 2.} The interval between the five light sources is l. Hereinafter, the five light sources _{L -2, L -1, L 0} , L consists of the surface (X-axis and Y-axis of the z = 0 containing ₁ and _{L 2} X, surface to be the Y axis parallel z = 0 the plane ) Is referred to as a “light source surface”. That is, the distance in the Z-axis direction from the light source array 11 to the surface of the measurement target object 21 is set to z (in other words, the position in the Z-axis direction from the light source surface is a coordinate point z representing “height”). it can.

Grid plate 12, five light sources _{L -2,} L _-1, (in the example shown in FIG. 1, X-axis direction) L 0, _{L 1} and _{L 2} having an array direction with respect to the vertical direction (i.e., Y-axis A light-transmitting region 12b composed of a straight line (direction) is arranged on the light shielding region 12a at equal intervals. For example, the grid plate 12 can be configured by printing a black one-dimensional grid pattern on the surface of a transparent glass or plastic material, and a slit (light blocking area 12a) with a slit ( A light transmission region 12b) may be provided. The grating plate 12 is configured such that a light irradiated from the light source array 11 passes through the grating plate 12 so that a one-dimensional grating is projected onto the measurement target object 21. The lattice interval of each light transmission region 12b constituting the one-dimensional lattice is p. In the embodiment illustrated in FIG. 1, the light source surface and the lattice surface are parallel, and the distance between the light source surface and the lattice surface is d. Also, the reference point E (phase φ = 0) is the shortest distance from the Z-axis among the central positions of the light transmission regions 12b constituting the one-dimensional grating. Note that the luminance distribution of the projected one-dimensional grating needs to be a cosine wave. For this reason, even if the luminance distribution (transmittance distribution) on the grid plate 12 is a rectangular wave, the primary projected onto the measurement target object 21 by the light emitted from the light source of the light source array 11 provided at the position of the distance d. The original lattice has a substantially cosine wave shape on the image captured by the camera 13. It does not have to be strictly a cosine wave. In particular, the error can be canceled by using the all-space tabulation technique described later. The light source array 11 and the grating plate 12 constitute a projector 15 that projects a one-dimensional grating.

  The configuration of the shape measuring apparatus 1 according to an embodiment of the present invention is configured to combine the advantages of the shadow moiré method and the advantages of the light source switching phase shift method in order to enable precise shape measurement. The grid plate 12 is arranged at a position (distance D) close to the measurement target object 21, and the grid plate 12 and the measurement target object 21 are arranged in the field of view of the camera 13. In the shape measuring apparatus 1 according to the present invention, based on the grid plate 12 arranged in the field of view of the camera 13, the phase of the grid shadow (projected grid) projected on the measurement target object 21 or the grid and grid shadows is used. By calculating the phase of the moiré fringes, the height z of the measurement target object 21 is obtained from this phase. The distance D between the grid plate 12 and the measurement target object 21 is determined by the resolution to be measured for the measurement target object 21, the light source pitch of the light source array 11, the grid pitch of the grid plate 12, and the number of pixels (resolution) of the camera 13. , Can be adjusted according to the application. Note that whether to measure the shape of the measurement target object 21 from the phase of the moire fringe or the phase of the shadow of the grating depends on the application. For example, when the lattice pitch in the captured image of the camera 13 is smaller than the pixel pitch of the camera 13 in the captured image, the lattice cannot be captured, but moiré fringes can be captured. On the other hand, when the pitch of the grid in the captured image is much larger than the pixel pitch of the camera 13 in the captured image, the phase of the grid may be directly analyzed.

In the following description, these light sources _{L -2, L -1, L 0} , the brightness distribution of _{L 1} and _{L 2} in the observation area z = constant x, with respect to the y-axis direction, equal and uniform Assume. If it is not uniform, the distribution may be considered as a coefficient.

(Controller unit)
As described above, in the shape measuring apparatus 1 according to the present invention, the phase or primary phase of the grid shadow (projection grid) projected on the measurement target object 21 based on the grid plate 12 arranged in the field of view of the camera 13. The height z of the measurement target object 21 is obtained by analyzing the phase of the moire fringes composed of the original lattice and the shadow of the lattice. For this reason, the control unit 50 sets the pixel position of the measurement range of the camera 13 and controls the switching lighting of the light source when performing phase shift, and the relationship between the height z of the measurement target object 21 and the luminance, or The relationship between the height z of the measurement target object 21 and the phase of the projection grating due to the phase shift is obtained, and control is performed so as to obtain the height z of the measurement target object 21 from this phase.

  FIG. 2 is a block diagram of the control unit 50 in the shape measuring apparatus 1 according to an embodiment of the present invention. As shown in FIG. 2, the control unit 50 sets the pixel position of the measurement range of the camera 13, and controls the switching of the light source when performing the phase shift, and on the measurement target object 21 due to the lighting of the light source. A camera control signal for photographing the projected one-dimensional lattice is supplied to the camera 13, and an image signal obtained by photographing the one-dimensional lattice is acquired from the camera 13. Further, the control unit 50 directly obtains the height coordinate z of the measurement target object 21 based on the luminance value for each pixel obtained from the photographed image signal, or by each light source by the light source switching phase shift method. The phase of the one-dimensional lattice projected on the measurement target object 21 is calculated, and the height coordinate z related to the measurement target object 21 is obtained from this phase based on a table prepared in advance or an all space table forming technique using an approximate expression. More specifically, the control unit 50 includes a control unit 51 and a memory 52. The control unit 51 includes a measurement range setting unit 511, a light source determination unit 512, a camera photographing processing unit 513, a pixel-by-pixel luminance calculation unit 514, a phase calculation unit 515, and a three-dimensional spatial coordinate calculation unit 516.

  In the control of the shape measuring apparatus 1 of the present embodiment, a technique for obtaining three-dimensional spatial coordinates related to the measurement target object based on the phase of the grid shadow projected onto the measurement target object 21 by the light source, and the one-dimensional grid and grid shadow There is a technique for obtaining a three-dimensional spatial coordinate related to the measurement target object based on the phase of the moiré fringes formed of Therefore, the measurement range setting unit 511 calculates a predetermined range (preferably one grid pitch or 1/2) in an area other than the grid portion of the grid plate 12 when calculating the phase of the grid shadow with respect to the image captured by the camera 13. When a predetermined number of pixels is assigned to (grid pitch range) and the phase of a moire fringe composed of a one-dimensional lattice and a lattice shadow is calculated, a predetermined range including a set of one-dimensional lattice and lattice shadow (preferably By assigning a predetermined number of pixels to (one grid pitch range), the pixel position of the measurement range of the camera 13 is set, and the light source determination unit 512 is instructed regarding the measurement range. Here, the relationship between the measurable distance range, the number of light sources corresponding to the number N of phase shifts, the light source pitch l, and the grating pitch p of the grating plate 12 is determined in advance as the configuration of the shape measuring apparatus 1. I can leave.

  The light source determination unit 512 controls lighting of the light source, determines a combination of light sources (three or more light sources) that performs lighting switching for phase shift according to the measurement range, and sequentially determines the determined light sources. A light source switching control signal for lighting is supplied to the light source array 11.

  The camera photographing processing unit 513 supplies a camera control signal for photographing a one-dimensional lattice projected on the measurement target object 21 by turning on the light source to the camera 13 and acquires an image signal of an image photographed from the camera 13.

  The pixel-by-pixel luminance calculation unit 514 calculates a luminance value at each pixel position from the acquired image signal.

  In the case of the light source switching phase shift method, the phase calculation unit 515 calculates the phase of the one-dimensional grating by each light source based on the luminance value for each pixel obtained from the image signal of the captured image. Hereinafter, the processing for obtaining the phase φ is referred to as “phase analysis processing”. Further, x and y coordinates are obtained for each pixel from the phase θ by, for example, phase analysis using a Fourier transform lattice method.

  The three-dimensional spatial coordinate calculation unit 516 obtains the height coordinate z from the phase obtained for each pixel in the measurement target object 21 based on a table prepared in advance or an all space table forming technique using an approximate expression, and Three-dimensional spatial coordinate data is stored in the memory 52.

  For the phase analysis processing related to the projection grating, a table or approximate expression that associates the phase φ of the one-dimensional grating with the spatial coordinates (x, y, z) is created in advance for each pixel by applying a total space table forming technique. In the distance measurement, the spatial coordinates (that is, the coordinates of the point S on the measurement target object 21) can be obtained from the phase of the one-dimensional lattice obtained from each pixel with reference to this table or the approximate expression. More specifically, as a function of the three-dimensional space coordinate calculation unit 516 in the control unit 50, a three-dimensional space composed of the x and y coordinates of the X and Y axes determined by the pixel position of the camera 13 and the z coordinate of the Z axis. For coordinates, a table or an entire space table by an approximate expression is created for each z coordinate at a predetermined interval measured using a predetermined reference flat plate (hereinafter also referred to as “reference plane”) as disclosed in Patent Document 1. The three-dimensional spatial coordinates related to the measurement target object 21 are derived from the phase φ of the one-dimensional lattice calculated for the measurement target object 21 with reference to these tables or approximate expressions held in the memory 52. What is necessary is just to comprise so.

  That is, the control unit 50 calculates the phase θ of the one-dimensional grating by each light source from the luminance at each pixel position obtained from the captured image of the camera 13, and based on a table prepared in advance or an all-space table forming technique using an approximate expression. A process for obtaining the height coordinate z related to the measurement target object 21 can be performed. The x and y coordinates can be obtained from phase analysis by the Fourier transform lattice method, for example. Therefore, the z coordinate is obtained for each pixel of the camera 13, and the x and y coordinates are also obtained.

  Hereinafter, the measurement principle based on the total space table forming technique will be described in detail.

In the following description, it is assumed that the respective light emission luminance distributions (light emission directivity characteristics) of the five light sources L ₋₂ , L ₋₁ , L ₀ , L ₁ and L ₂ are uniform and equal within the observation range. If it is not uniform, the luminance value obtained from the captured image may be corrected using the distribution as a coefficient. An example in which the light source surface and the lattice surface are arranged in parallel will be described. Here, when the light source surface and the lattice surface are not parallel, the distance d between the light source surface and the lattice surface is a distance in the Z-axis direction from the origin O of the light source array 11 to the lattice surface of the lattice plate 12. The non-uniformity of the light emission luminance distribution (light emission directivity) of each light source may be corrected similarly.

[Calibration processing method of shape measuring apparatus 1 and control method by control unit]
Next, a calibration processing method and a control method of the shape measuring apparatus 1 of the present embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of an operation flow in the shape measuring apparatus 1 according to an embodiment of the present invention.

(Calibration process)
As described above, in the control of the shape measuring apparatus 1 of the present embodiment, a technique for obtaining three-dimensional spatial coordinates related to the measurement target object based on the phase of the grid shadow projected onto the measurement target object 21 by the light source, and the one-dimensional There is a technique for obtaining a three-dimensional spatial coordinate related to the measurement target object based on the phase of a moire fringe composed of a lattice and a shadow of the lattice. First, regarding the image captured by the camera 13, when calculating the phase of the shadow of the grid, it is preliminarily set within a predetermined range (preferably, a range of 1 grid pitch or 1/2 grid pitch) in an area other than the grid portion of the grid plate 12. A predetermined number of pixels is allocated, and when calculating the phase of a moire fringe composed of a one-dimensional grid and a grid shadow, a predetermined range including a set of one-dimensional grid and grid shadow (preferably with a grid pitch of one grid pitch). The pixel position (x, y) in the measurement range of the camera 13 is determined by assigning a predetermined number of pixels to (range) (step S1).

  The shape measuring apparatus 1 is configured by determining the grating pitch p, the light source pitch l, and (N, M) in the measurement range (step S2). Note that (N, M) is determined by switching the light source according to the measurement distance so that the total phase shift amount when the phase shift is performed N times is −2πM, as shown in equation (21) described later. This means that the light source for performing is determined in advance.

  Next, calibration processing is executed based on the values of the grating pitch p, the light source pitch l, and (N, M) (step S3). Specifically, the two-dimensional plane coordinate x is located at one height coordinate z among a plurality of height coordinates z determined based on the combination of the grating pitch p, the light source pitch l, and (N, M). , Y is arranged. The control unit 50 turns on one of the plurality of light sources determined based on the combination of the grating pitch p, the light source pitch l, and (N, M), and displays an image of a one-dimensional grating on the arranged reference plane. Project. The camera 13 captures an image of a one-dimensional lattice projected on the reference plane. Subsequently, the imaging of the one-dimensional grating is repeated for all of the plurality of height coordinates z determined based on the combination of the control unit 50, the grating pitch p, the light source pitch l, and (N, M). Among the phases calculated from the image of the one-dimensional lattice, a phase value having a 1: 1 correspondence with the height coordinate z of the reference plane is calculated for each pixel of the camera 13.

  Next, the control unit 50 creates an entire space table for the height z and the phase φ having a 1: 1 correspondence relationship, and stores and holds them in the memory 52 (step S4). Specifically, the control unit 50 calculates the two-dimensional plane coordinates x, y on the reference plane for each pixel of the camera 13, and the height coordinate z corresponding to the calculated phase value for each pixel of the camera 13 and A table or approximate expression of three-dimensional space coordinates composed of two-dimensional plane coordinates x and y is created and configured as an all space table, and stored in the memory 52. Thereby, the calibration process is completed.

(Control method by control unit)
In actual measurement of the measurement target object 21, the measurement range setting unit 511 of the control unit 50 sets a measurement range related to the measurement target object 21, and instructs the light source determination unit 512 about the measurement range (step S5). Even during actual measurement, when calculating the phase of the grid shadow with respect to the image captured by the camera 13, a predetermined range (preferably 1 grid pitch or 1/2 grid pitch) in an area other than the grid portion of the grid plate 12 is used. A predetermined number of pixels is assigned to a predetermined range (preferably, including a set of one-dimensional grid and grid shadow) when calculating the phase of the moire fringe composed of the one-dimensional grid and grid shadow. A predetermined number of pixels is assigned to a range of one grid pitch.

  Subsequently, the light source determination unit 512 of the control unit 50 determines a combination of light sources (three or more light sources) that performs lighting switching for phase shift according to the measurement range, and sequentially lights each determined light source. Is supplied to the light source array 11 to turn on one of the five light sources (step S6).

  Subsequently, the camera imaging processing unit 513 of the control unit 50 captures the measurement target object 21 onto which the one-dimensional lattice is projected via the lattice plate 12 with the camera 13 (step S7).

  Subsequently, the camera imaging processing unit 513 of the control unit 50 determines whether or not all of the light sources that perform the phase shift determined in Step S2 have been photographed (Step S8). If all the light sources that perform phase shifting have been photographed (step S8: Y), the process proceeds to step S9. On the other hand, if not all of the light sources that perform phase shifting have been imaged (step S8: N), steps S6 and S7 are repeated until all of the imaging is completed. Accordingly, it is possible to obtain an image signal obtained by photographing the measurement target object 21 while sequentially turning on the light source that performs the phase shift and shifting the phase of the one-dimensional grating projected onto the measurement target object 21. This image signal can be obtained as a luminance value for each pixel.

  That is, the pixel-by-pixel luminance calculation unit 514 of the control unit 50 determines each luminance value by the phase shift for each pixel of the camera 13 (step S9).

  Subsequently, the phase calculation unit 515 of the control unit 50 calculates the phase φ of the one-dimensional grating projected onto the measurement target object 21 from the calculated luminance value (step S10).

  Subsequently, the phase calculation unit 515 of the control unit 50 refers to the entire space table based on the obtained phase φ, and determines the three-dimensional space coordinates of the measurement target object 21 (step S11). In this manner, the shape of the measurement target object 21 can be obtained by performing the phase analysis process on the captured image.

Referring to FIG. 1, the camera 13 _displays a measurement target object 21 onto which a one-dimensional lattice is projected by lighting one of five light sources L ₋₂ , L ₋₁ , L ₀ , L ₁ and L _2. Take a picture. As the camera 13, a CCD sensor, a CMOS sensor, or the like can be used. For the x and y coordinates, for example, the phase in the X-axis direction and the Y-axis direction are obtained by the Fourier transform grid method, and the phase connection is further performed to obtain the x-coordinate and y-coordinate at each point ( For example, see Japanese Patent No. 3281918).

  In addition, a member for enlarging or reducing an image such as a lens may be disposed between the light source and the grating plate 12 or between the grating plate 12 and the measurement target object 21.

  In the shape measuring apparatus 1 of the above-described embodiment, the light source array 11 is composed of at least three point light sources or linear light sources arranged at equal intervals corresponding to a plurality of types of phase shifts. Of the at least three light sources, a light source that performs phase shift of the shadow of the grid to be projected is determined, and each of the determined light sources is sequentially turned on, and the shadow or moire fringe of the grid projected on the measurement target object 21 respectively. The camera 13 is controlled to take a picture, and the phase of the shadow of the grating or the phase of the moire fringes projected on the measurement target object 21 by each of the determined light sources is calculated for each pixel of the camera 13. That is, the shape measuring apparatus 1 according to the above-described embodiment is an aspect using the light source switching phase shift method.

  On the other hand, as will be described later in detail, as one aspect of the shape measuring apparatus 1 of the present embodiment, the light source array 11 is composed of one point light source or a linear light source. When the relative position of the object 21 moves relative to the object plate 21 at a predetermined speed (hereinafter, also simply referred to as “relative movement”), the control unit 50 synchronizes with the predetermined speed. Then, the camera 13 is controlled so as to capture a grid shadow or moire fringe projected on the measurement target object 21, and the phase of the grid shadow or moire fringe projected on the measurement target object 21 is determined by the pixel of the camera 13. It can comprise so that it may each calculate for every. That is, in this case, the phase shift method using relative movement is used.

  Furthermore, as will be described in detail later, as one aspect of the shape measuring apparatus 1 of the present embodiment, the light source array 11 includes point light sources or linear shapes arranged at at least three equal intervals corresponding to the number of types of phase shifts. When the shape measuring apparatus 1 is arranged to move relative to the measurement target object 21 at a predetermined speed on a plane parallel to the grid plate 12, the control unit 50 includes at least three. Among the light sources, a light source that performs phase shift of the shadow of the grid to be projected is determined, and each of the determined light sources is sequentially turned on so as to capture the shadow or moire fringe projected on the measurement target object 21. The camera 13 is controlled, and the phase of the grid shadow or the moire fringe projected on the measurement target object 21 by each of the determined light sources is calculated for each pixel of the camera 13. It can be configured to. That is, in this case, the phase shift method using both light source switching and relative movement is used.

  Further, as will be described in detail later, as an aspect of the shape measuring apparatus 1 of the present embodiment, the camera 13 is a measurement target on which a grid shadow is projected by the grid plate 12 by turning on the light sources in the light source array 11 with different lines of sight. It can be composed of a plurality of cameras that photograph the object 21. In this case, the control unit 50 controls the plurality of cameras so that the shadows or moire fringes of the lattice projected onto the measurement target object 21 are photographed at different locations, and based on the images photographed by the plurality of cameras. The phase of the shadow of the grating projected on the measurement target object 21 or the phase of the moire fringes can be calculated for each pixel of the camera 13.

  In particular, when the shape measuring device 1 is in an arrangement relationship in which the shape measuring device 1 moves relative to the measurement target object 21 at a predetermined speed, if the luminance value of the moire fringe is detected, the luminance change is used to detect the measurement target object 21. The shape can be measured.

  Hereinafter, the operation principle of the shape measuring apparatus 1 related to the above-described several aspects will be described in more detail.

  FIG. 4 shows a configuration example when measuring the shape of the measurement target object 21 with the light source array 11 composed of a plurality of light sources, the lattice plate 12, and one camera 13 in the shape measuring apparatus 1 of the present embodiment. The operation principle is shown. FIG. 4A is a top view, and FIG. 4B is a side view. The “phase shift shadow measurement method” according to the present invention includes a technique of performing phase shift by switching light sources using the shadow of the grating (hereinafter referred to as “light source switching phase shift shadow measurement method”), and a shadow of the grating. There is a technique (hereinafter referred to as “relative movement phase shift shadow measurement method”) in which phase shift is performed by relative movement using the above. First, in the “light source switching phase shift shadow measurement method”, the phase of a grid shadow (projection grid) projected onto the measurement target object 21 in an area other than the grid portion of the grid plate 12 is obtained, and shape measurement is performed from this phase. The technique to perform is called the “light source switching phase shift shadow grating method”, the phase of moire fringes formed by the grating and the shadow of the grating (projection grating) is obtained, and the technique of measuring the shape from this phase is called “light source switching phase shift shadow moire method”. The method will be described in order.

(Light source switching phase shift shadow grating method)
As shown in FIG. 4, normally, when a lattice is captured in the field of view of the camera 13, image data of the portion where the lattice is captured cannot be used for analysis, and information on the portion is lost. Therefore, when the light source switching phase shift shadow grating method is used, the shape measuring apparatus 1 is projected onto the measurement target object 21 by the camera 13 in an area other than the grating portion in the grating plate 12 arranged in the field of view of the camera 13. The grid shadow (projection grid) is observed, the brightness change of the grid shadow is caused by switching the light source, and the phase of the grid shadow is calculated from this brightness change. For example, FIG. 5 shows an example of the relationship between the measurement pixel array in the measurement range of the camera 13 in the light source switching phase shift shadow grating method, the grating, and the projection grating in the shape measuring apparatus 1 of the present embodiment. FIG. 5A is a top view, and FIG. 5B is a side view. FIG. 6 shows another example of the relationship between the measurement pixel array in the measurement range of the camera 13 in the light source switching phase shift shadow grating method, the grating, and the projection grating in the shape measuring apparatus 1 of the present embodiment. 6A is a top view, and FIG. 6B is a side view. 5 and 6, both of the images taken by the camera 13 calculate the phase of the grid shadow when calculating the phase of the grid shadow in a predetermined range (preferably one grid pitch or 1 / A predetermined number of pixels is assigned to a range of 2 lattice pitches. As described above, since the number of pixels (that is, an integer of 1 or more) set in advance in a range corresponding to one period at the grating pitch, the phase analysis of the shadow of the grating can be performed with high accuracy.

  Next, consider the case where two cameras 13-1 and 13-2 are used as shown in FIG. The first camera 13-1 is installed so that a portion that cannot be seen by the grid can be seen by the second camera 13-2. By extracting only the data in which the objects of both images of the two cameras taken in this way are visible, or by averaging the corresponding images, all of the shadows of the projected grid can be seen. It will be. As a result, it is possible to analyze the luminance change or the phase at all locations related to the shape measurement region of the measurement target object 21. FIG. 8 shows an example of the relationship between the measurement pixel array, the grid, and the projection grid in the measurement range of the two cameras 13-1 and 13-2 in the shape measuring apparatus 1 of the present embodiment. In this example, FIG. 8A is a top view when the first camera is imaged, and FIG. 8B is illustrated as a place where the shadow of a certain grid advances when the first camera is imaged (light source switching). 2 are side views of “1” and “2”). FIG. 8C is a top view of the second camera (“3” and “4” illustrated as places where the shadow of a certain grid advances when the light source is switched), and FIG. It is a side view at the time of 2nd camera imaging.

  That is, in FIG. 8, by switching the light source according to the light source switching phase shift method using the two cameras 13-1 and 13-2, N phase-shifted images can be taken. Measurement is possible by the light source switching phase shift method. When two cameras 13-1 and 13-2 are used, the cameras are synchronized to perform simultaneous shooting so that N pixels continuous in the x-axis direction of each camera become one pitch of the grid. Adjust the magnification. Thus, by performing lighting switching of N consecutive light source positions, the shadow of the grating advances by one pitch, and the phase of the grating corresponding to each pixel of each camera is phase-shifted by one period. Therefore, it is possible to take N frames of images phase-shifted by each camera and obtain phase-shifted grid shadow luminance data. From this luminance data, the phase φ of the one-dimensional grating is calculated, and the three-dimensional spatial coordinates of the measurement target object 21 can be determined by referring to the entire space table based on the obtained phase φ. By using two cameras, there is no portion that cannot be measured because only one camera cannot see the lattice, and high resolution analysis is possible.

(Light source switching phase shift shadow moire method)
On the other hand, when the light source switching phase shift shadow moiré method is used, the shape measuring apparatus 1 moiré fringes formed from the grid shadow (projection grid) projected onto the measurement target object 21 and the grid within the field of view of the camera 13. Is observed, and the luminance change of the shadow of the grating is caused by switching the light source, and the phase of the moire fringe is calculated from this luminance change. When calculating the phase of the grid shadow with respect to the image taken by the camera 13, a predetermined range including a set of one-dimensional grid and grid shadow in an area other than the grid portion of the grid plate 12 (preferably having a grid pitch of 1 grid pitch). A predetermined number of pixels is assigned to (range). Thus, since the range of the phase analysis processing is set to a predetermined number of pixels (that is, an integer of 1 or more), the phase analysis of the grid shadow can be performed with high accuracy.

  In the “light source switching phase shift shadow moire method” as well, as shown in FIG. 8, a plurality of cameras (for example, two cameras 13-1 and 13-2) are used, so that one camera is used. With this method, there is no part that cannot be measured because the grid is not visible, and high resolution analysis is possible.

  Whether to use the light source switching phase shift shadow grating method or the light source switching phase shift shadow moire method may be determined according to the application. However, even when measuring the shape of the same object to be measured 21, when the light source switching phase shift shadow grating method is used, the phase of the shadow of the grating changes by 2π in the range of one pitch of the grating, but the light source switching phase shift shadow is generated. When the moire method is used, the phase of the moire fringes = the phase of the grating in the grating plate 12−the phase of the shadow of the grating is measured, and therefore the length of one period of the moire fringes generated when one light source is turned on is longer than the grating pitch. For this reason, it is necessary to determine the light source pitch so that a phase change of 2π occurs due to the light source switching.

  That is, in the light source switching phase shift shadow grating method, the phase of the grating shadow itself is analyzed. On the other hand, in the light source switching phase shift shadow moire method, in order to observe the moire fringes that appear when the lattice and the shadow of the lattice are observed simultaneously, one pitch of the lattice is adjusted to N pixels continuous in the x-axis direction of the camera 13, Since the average brightness of the N pixels in the portion where both the lattice and the shadow of the lattice are reflected becomes the brightness of the moire fringes, the phase of the moire fringes is analyzed. In both the light source switching phase shift shadow grating method and the light source switching phase shift shadow moire method, the phase shift can be performed by switching the lighting of the light sources of the light source array 11.

  FIG. 9 shows an example of the relationship between the measurement pixel array in the measurement range of the camera 13 in the light source switching phase shift shadow moiré method, the grating, and the projection grating in the shape measuring apparatus 1 of this embodiment. In this example, FIG. 9A is a top view and FIG. 9B is a side view. In FIG. 9, the magnification is set so that N pixels in the x-axis direction of one camera 13 become one pitch of the lattice (N = 4 in the figure), and the average of consecutive N pixels in one imaging. By taking the luminance, the luminance change of the first pixel (the pixel “1” in the figure) can be obtained. Regarding the installation of one camera 13, the magnification and the arrangement are adjusted so that moire fringes are obtained in units of N pixels. In FIG. 9, the pixel columns for which the average luminance is taken are shown by shifting the rows for convenience of explanation, but the average is obtained by shifting one pixel at a time in the same row. When there is no relative movement with respect to the measurement target object, even if the average luminance of N pixels is calculated while shifting one pixel at a time in the X axis direction, one phase-shifted moire fringe cannot be obtained, but relative movement with respect to the measurement target object 21 When there is, one period of moire fringes can be obtained.

(Relative moving phase shift shadow measurement method)
As described above, the “phase shift shadow measurement method” according to the present invention includes a grid shadow separately from the “light source switching phase shift shadow measurement method” in which phase shift is performed by switching light sources using the shadow of the grating. There is a “relative movement phase shift shadow measurement method” in which phase shift is performed by relative movement. In this “relative movement phase shift shadow measurement method”, by using relative movement instead of light source switching, measurement can be performed in the same manner as shown in FIGS. 5 to 8, and “relative movement phase shift shadow grating method”. I will call it. Further, by using relative movement instead of light source switching, measurement can be performed in the same manner as in the case shown in FIG. 9, which is referred to as “relative movement phase shift shadow moire method”.

  In the light source switching phase shift shadow grating method and the light source switching phase shift shadow moire method described above, the phase shift is realized by switching the lighting of the light sources of the light source array 11. On the other hand, when the configuration is based on the “relative movement phase shift shadow grating method” in which the phase shift of the shadow of the grating is performed based on the relative movement, only one light source can be obtained without switching the light source. In the “relative movement phase shift shadow grid method”, when the N pixels of the camera 13 are fixed, even when the measurement target object 21 is a flat plate, when the shadow grid method is used, the measurement target object 21 is accompanied by relative movement. Since the phase of the upper point is shifted, the phase analysis of the shadow of the grating can be performed.

  On the other hand, in the “relative movement phase shift shadow moiré method” in which the phase shift of the moire fringes composed of the grating and the shadow of the grating is performed based on the relative movement, the shadow of the grating and the grating is assumed when only one light source is used without switching the light source. Since the average of N pixels that are continuous in the x-axis direction of the portion where both of them are captured is taken, the point on the measurement target object 21 cannot be tracked, the phase of the moire fringe is not shifted, and the phase analysis cannot be performed. There is a case. Thus, in the “relative movement phase shift shadow moiré method”, even when taking the average of N pixels, three or more light sources of the light source array 11 are prepared and the phases of the moiré fringes are shifted by switching these light sources. To be able to do Thereby, the phase of moire fringes can be analyzed.

  As described above, in general, when a lattice is captured in the field of view of the camera 13, image data of a portion where the lattice is captured cannot be used for analysis, and information on the portion is lost. Therefore, when the configuration has “relative movement with respect to the measurement target object”, it is possible to acquire the moire fringes and the phase of the grid shadow related to the grid shadow (projection grid) necessary for the shape measurement of the measurement target object 21. . By acquiring moire fringes, the change is more sensitive than the measurement value from a simple luminance change, and therefore the measurement accuracy is improved.

  For example, the shape measuring apparatus 1 is moved with respect to the measurement target object 21 in the x, y plane, or the measurement target object 21 is moved with respect to the shape measuring apparatus 1 in the x, y plane, or the shape measuring apparatus 1 is moved. The shape measuring apparatus 1 moves relative to the measurement target object 21 by moving both the measurement target object 21 and the measurement target object 21 in different directions or different speeds on the x and y planes, thereby realizing a phase shift related to the shadow of the lattice. I can do it. As described above, the moiré fringes can be photographed by taking the average of N pixels by making the N pixels of the camera 13 correspond to one pitch of the lattice.

  In both of the “relative movement phase shift shadow grating method” and the “relative movement phase shift shadow moire method”, as shown in FIG. 7, two cameras 13-1 and 13-2 may be used. it can.

  Furthermore, in order to phase shift the moiré fringes relating to the shadow of the grating simultaneously with the relative movement, the light source array 11 may be composed of a plurality of light sources and each light source may be switched. In this case, both a phase shift due to switching of the light source and a phase shift due to movement of the measurement target object 21 occur simultaneously. Therefore, if the N pixels in the x-axis direction of the camera 13 are made to correspond to ½ pitch of the lattice, each pixel chasing the measurement target object 21 corresponds to N phase shifts in a half cycle. Note that the light source switching timing and lighting switching order are adjusted so that both the phase shift due to the light source switching and the phase shift due to the measurement target object 21 do not cancel each other. For example, the light source switching order may be reversed with respect to the movement direction of the measurement object 21.

  Next, as a further application example of the shape measuring apparatus 1 according to the present invention, a case where the measurement target object 21 moves at a constant speed with respect to the relative movement phase shift shadow measurement method will be described.

[When the object to be measured moves at a constant speed]
The measurement target object 21 can be placed on a predetermined moving table and moved at a constant speed, or the measurement target object 21 itself can be used as a measurement target conveyor. In this case, the phase shift of the shadow of the grating can be calculated with only one line light source (for example, a linear LED light source) without switching light sources. FIG. 10 shows an object to be measured by a light source array 11 configured as one line light source (for example, a linear LED light source), a lattice plate 12, and one camera 13 in the shape measuring apparatus 1 of the present embodiment. The example of a structure at the time of measuring the shape of 21 and its operation principle are shown. In this example, FIG. 10 (a) is a top view and FIG. 10 (b) is a side view. In FIG. 10, regarding the light source array 11, even if the grating is illuminated by one line light source (linear LED light source), the phase shift is automatically performed by the movement of the measurement target object 21. Therefore, one light source and one camera are sufficient.

  FIG. 11A shows a shadow of a grid projected by one line light source (for example, a linear LED light source) when the shape measuring apparatus 1 of the present embodiment is configured to have a relative movement with respect to the measurement target object. It is a top view which shows an example of the relationship between the measurement pixel row | line | column in the measurement range of the camera 13, and the grating | lattice, and a projection grating | lattice which images. FIG. 11B shows a grid projected by one line light source (for example, a linear LED light source) when the shape measuring apparatus 1 according to this embodiment is configured to have a relative movement with respect to the measurement target object. It is a top view which shows another example of the relationship between the measurement pixel row | line | column in the measurement range of the camera 13 which images the shadow of this, and a grating | lattice and a projection grating | lattice. That is, FIG. 11 shows a measurement technique based on the “relatively moving phase shift shadow grating method”, FIG. 11A shows a phase shift method in which one pitch of the grating is divided into N, and FIG. The phase shift method which divides | segments a half pitch into N is shown.

(Relative moving phase shift shadow lattice method)
In FIG. 11, the magnification and position of the camera 13 are adjusted so that the length of one pitch of the grid becomes N pixels continuous in the x-axis direction of the camera 13. When the measurement target object 21 is photographed every 1 / N pitch by the camera 13, N images with the phase shift of the grid shadow can be obtained, so that the measurement target object 21 can be measured over the whole. If a line light source (for example, a linear LED light source) is always turned on at this time, an integration effect occurs, and even if a binary waveform is projected, it becomes a trapezoidal wave, and a synergistic effect with blur makes it almost a sine wave, improving accuracy. To do. Note that the relationship between the number of pixels N and the moving speed in this relative movement phase shift shadow measurement method is that the total phase shift amount when the phase shift is performed N times in the light source switching phase shift method represented by Equation (21) is − This corresponds to the fact that the light source switching is set to be 2πM.

  Next, consider the case of using two cameras 13-1 and 13-2 from the configuration shown in FIG. The part of the measurement target object 21 that was not visible on the grid by the first camera 13-1 is installed so that it can be seen by the second camera 13-2. By extracting only the data that can see the objects of both images of the two cameras taken in this way, or by averaging the corresponding images, all of the projected grids will be visible Become. As a result, it is possible to analyze the luminance change or the phase of the shadow of the grid at all locations related to the shape measurement region of the measurement target object 21. FIG. 12 shows an example of the relationship between the measurement pixel array, the grid, and the projection grid in the measurement range of the two cameras 13-1 and 13-2 in the shape measuring apparatus 1 of the present embodiment. In this example, FIG. 12A is a top view at the time of imaging with the first camera, and FIG. 12B is illustrated at the time of imaging with the first camera (where the same point on the object has advanced). 1 "and" 2 "). FIG. 12C is a top view at the time of imaging by the second camera (“3” and “4” illustrated as places where the same point on the object has advanced), and FIG. It is a side view at the time of camera imaging. That is, the numbers “1”, “2”, “3”, and “4” indicate locations where the same point on the object has advanced, and each uses luminance data that is delayed by one frame in the order of the numbers. It can be phase shift data.

  That is, in FIG. 12, using the two cameras 13-1 and 13-2 in this way, it is possible to shoot N phase-shifted images using the movement of the measurement target object 21, Measurement is possible by the phase shift method. When two cameras 13-1 and 13-2 are used, the cameras are synchronized to perform simultaneous shooting so that N pixels continuous in the x-axis direction of each camera become one pitch of the grid. Adjust the magnification. As a result, N pixels in the traveling direction are imaged N times while the measurement target object 21 advances a half pitch of the one-dimensional lattice. In N times of shooting, the shadow of the grating advances by half a pitch, and the phase of the grating is shifted by one period when the cameras 13-1 and 13-2 are used. Therefore, it is possible to take N frames of images phase-shifted by the cameras 13-1 and 13-2, and to obtain luminance data of the phase-shifted grid shadow for one period. From this luminance data, the phase φ of the one-dimensional grating is calculated, and the three-dimensional spatial coordinates of the measurement target object 21 can be determined by referring to the entire space table based on the obtained phase φ. Therefore, by using the two cameras 13-1 and 13-2, the detection speed of the phase shift is doubled compared to the case of using the single camera 13.

(Relative movement phase shift shadow moiré method)
FIG. 13 shows an example of the relationship between the measurement pixel array in the measurement range of the camera 13, the grid, and the projection grid when the shape measurement apparatus 1 of the present embodiment is configured to have relative movement with respect to the measurement target object. . In this example, FIG. 13 (a) is a top view and FIG. 13 (b) is a side view. In FIG. 13, the magnification is set so that N pixels in the x-axis direction of one camera 13 become one pitch of the lattice (N = 4 in the figure), and the average of consecutive N pixels in one imaging. Moire fringes can be photographed by taking brightness. In FIG. 13, for the convenience of explanation, the pixel column for taking the average luminance is shown by shifting the rows, but the average is obtained by shifting one pixel at a time in the same row. Further, in this example, since the measurement target object 21 moves at a constant speed, a phase shift relating to the moire fringe occurs, so that one cycle of moire fringes can be obtained. However, in the “relative movement phase shift shadow moiré method” in which the phase shift of the moire fringes composed of the grating and the shadow of the grating is performed based on the relative movement, the shadow of the grating and the grating is assumed when only one light source is used without switching the light source. Since the average of N pixels that are continuous in the x-axis direction of the portion where both of them are captured is taken, the point on the measurement target object 21 cannot be tracked, the phase of the moire fringe is not shifted, and the phase analysis cannot be performed. There is a case. Therefore, simultaneously with relative movement, in order to phase shift the moire fringes relating to the shadow of the grating, the light source array 11 may be composed of a plurality of light sources, and each light source may be switched. In this case, both a phase shift due to switching of the light source and a phase shift due to movement of the measurement target object 21 occur simultaneously. Further, if N pixels in the x-axis direction of the camera 13 are made to correspond to ½ pitch of the lattice, each pixel chasing the measurement target object 21 corresponds to N phase shifts. As described above, the light source switching timing and the lighting switching order are adjusted so that both the phase shift due to the light source switching and the phase shift due to the measurement target object 21 do not cancel each other. For example, the light source switching order may be reversed with respect to the movement direction of the measurement object 21.

The shape measuring apparatus 1 of this embodiment has the following advantages.
(1) Compared with the conventional grid projection method, a finer grid pitch can be used on the measurement target object 21, and the analysis accuracy can be improved.
(2) By sequentially lighting a plurality of line light sources, it is possible to perform a phase shift of a grating shadow or a moire fringe. In particular, two cameras 13-1 and 13-2 are installed, and the phase of the measurement range of each camera is shifted by π, so that the two cameras 13-1 and 13-2 are related to the measurement target object 21. It is possible to eliminate parts that cannot be shot.
(3) When the measurement target object 21 is moving at a constant speed, a phase shift by a plurality of line light sources is not necessarily required, and high-speed and high-precision analysis can be performed with only one line light source.

[Operation Principle of Light Source Switching Phase Shift Shadow Measurement Method]
Next, in order to deepen the understanding of the above-mentioned “light source switching phase shift shadow measurement method”, the lattice projection method, phase shift method, full space table conversion method, and light source switching phase shift method, which are the basis thereof, will be described in order. I will keep it.

(Lattice projection)
A schematic diagram of the grid projection method is shown in FIG. When a grating is projected onto the measurement target object 21 placed on the reference flat plate 14 by the projector 10 and photographed by the camera 13 from the side of the projector 10, the projected grating is distorted according to the shape of the object. In the lattice projection method, the shape is obtained by analyzing the distortion.

  The luminance values I (x, y) of the grating and the interference fringes are generally distributed in a wave shape on the space (x, y) as shown in FIG. This is expressed as follows.

  Here, the point (x, y) is one point in the captured image, a (x, y) and b (x, y) represent the luminance amplitude and the background luminance, respectively, and φ (x, y) is Represents the phase of the grating.

  In the case of a lattice image, the phase can be expressed as a whole real number. FIG. 17B shows the phase distribution as a repetition from −π to π, and FIG. 17C shows this phase continuously. Also included are phases exceeding -π to π obtained as a result of review. The phase in FIG. 17B is called a lapped phase and is expressed by θ (−π ≦ θ <π). The phase in FIG. 17C is called an unwrapped phase or a continuous phase, and a phase after phase connection, and is expressed by φ. There is a one-to-one relationship between the phase φ and the three-dimensional coordinates (x, y, z), and the three-dimensional coordinates can be obtained by obtaining this phase. By obtaining the three-dimensional coordinates of each point reflected in the camera 13, the three-dimensional shape of the object can be known.

(Phase shift method)
A phase shift method capable of accurately obtaining a phase from luminance information at one point will be described. Equation (1) has three unknowns, a (x, y), b (x, y), and φ (x, y). Therefore, basically, if there are three or more independent experimental data, the unknown is unknown. Can be determined. The phase shift method is a technique for obtaining a phase distribution from a plurality of images obtained by photographing a plurality of lattice images while changing the phase of the lattice by one period. Since the luminance changes for one period in all the pixels, the phase can be obtained independently for each point from the luminance change, that is, without using information on the luminance change of the surrounding pixels. Therefore, it can be said that this is an effective technique for measuring the shape of an object having steps or discontinuities. Here, first, the phase can be obtained from N (3 or more) luminances obtained by phase-shifting the most commonly used phase shift amount α every π / 2 (90 degrees).

  When the phase shift amount Ψ = α (= 2πk / N, k = 0,..., N−1) is added to the expression of the luminance distribution of the grating shown in Expression (1), Expression (2) is obtained.

FIG. 18 shows an example of the relationship between the phase shift amount α and the luminance change I at a point having the initial phase θ (lattice phase when α = 0) (when N = 4). In FIG. 18, the luminance for each phase shift amount of π / 2 is set to I ₀ , I ₁ , I ₂ , and I ₃ . When the initial phase at a certain point is θ, if the number of phase shifts is N and the luminance when the phase shift amount is 2πk / N is I _k (k = 1, 2,..., N−1), ) And the phase value θ can be obtained from this relational expression. In the following formula, (x, y) is omitted. Increasing the number N of phase shifts can reduce the influence of noise.

(All space table technique)
The all-space table forming technique uses the reference plane to obtain the correspondence between the phase θ of the grating projected by the projector and the three-dimensional coordinates (x, y, z) in advance for each pixel of the camera and form a table. In addition, it is a technique in which three-dimensional coordinates can be obtained immediately without calculation by referring to a table from the phase of the grating projected onto the object during measurement. In this technique, the correspondence between the phase of the grating and the three-dimensional coordinates is obtained for each pixel of the camera, thereby eliminating the effects of distortion of the lens and the distortion of the brightness of the projection grating, and at the same time, the three-dimensional shape. Can be obtained.

FIG. 19 shows the principle of measuring the shape by the calibration process using the whole space table forming technique using a large number of reference planes. A reference plane installed perpendicular to the z-axis is translated little by little in the z-axis direction. The camera 13 and the projector 15 are fixed above the reference plane. A one-dimensional grid is projected from the projector 15 onto the reference plane. The phase of the projected grating can be easily calculated by the phase shift method. It is assumed that a certain pixel of the camera 13 is shooting a point on the straight line L shown in FIG. The pixel is, the reference plane _{R 0, R 1, R 2} , ···, according to _{R N,} respectively point _{P 0, P 1, P 2} , ···, I will shoot _{P N.} The phases θ ₀ , θ ₁ , θ ₂ ,..., Θ _{N at the} respective points can be obtained by a phase shift method from a lattice image captured by shifting the phase of the lattice projected by the projector 15.

FIG. 20 shows the relationship between the imaging line L of one pixel and the x, y, z coordinates on the reference plane. For points P ₀ , P ₁ , P ₂ ,..., P _N on each reference plane captured by a certain pixel, an x coordinate and a y coordinate are obtained from a two-dimensional pattern fixed on the reference plane. To obtain the z coordinate from the position of the reference plane. The two-dimensional pattern fixed on the reference plane needs to be a pattern that can obtain the x coordinate and the y coordinate from the pattern. In this example, a liquid crystal panel in which a light diffusing plate is attached to the surface of a reference surface is used as a technique for obtaining phase values in the x and y directions from a two-dimensional lattice pattern. Therefore, the lattice pattern can be displayed on the reference plane, and the phase distribution can be obtained from the captured image by the phase shift method (for further details, refer to Patent Document 3).

In this way, three-dimensional spatial coordinates (x, y, z coordinates) with respect to the phase θ of the projection grating are obtained for each pixel for each reference plane position. The phase θ of the projection grating can be obtained only at the position of the reference plane. However, if necessary, the three-dimensional spatial coordinates for all phases can be obtained accurately by reducing the interval between the reference planes and interpolating between them. be able to. By this calibration technique, the phases θ ₀ , θ ₁ , θ ₂ ,..., Θ _N and the x coordinates x ₀ , x ₁ , x ₂ ,..., X _N , y coordinates y ₀ , y ₁ , y ₂ , ···, _{y N} and z-coordinate _{z 0, z 1, z 2} , ···, relationship _{z N} is obtained respectively for each pixel. Next, based on this correspondence relationship, a table or approximate expression of x coordinate, y coordinate and z coordinate corresponding to the equally-spaced phase θ is created in advance and stored in the memory as an “all space table”. deep. When an object to be actually measured is placed in the space of the entire space table, if the phase of each pixel is obtained, the object (x, y, z) of the object viewed by the pixel can be simply referred to the entire space table. Coordinates can be obtained with high accuracy. FIG. 21 is a graph showing the relationship between the phase θ in a certain pixel and the height z that is the measurement position. Thereby, it is possible to immediately obtain the spatial coordinates (x, y, z coordinates) of the point from the phase obtained by photographing the measurement target object 21. For example, in the case of the point P in FIG. 19, z _P can be obtained by using a table or an approximate expression showing the correspondence relationship between the phase θ _P and the z coordinate shown in FIG. 20.

(Advantages of the whole space table technique)
(1) A coordinate value can be obtained with high accuracy without being affected by distortion due to distortion of the lens of the camera 13. With respect to the camera 13, it is possible to accurately measure the shape even when an optical system such as a fish-eye lens that is clearly distorted is used or measurement is performed through the wall of a container such as an aquarium.
(2) Measurement is possible even when the pitch of the grating is slightly unequal. For example, the shape can be measured by projecting while rotating a radial or spiral grid.
(3) Measurement is possible even if the luminance distribution of the grating is not a little cosine wave-like. Further, even when the rectangular wave-shaped grating is defocused and projected as a grating having a luminance distribution close to a cosine wave, the projector 10 has a slightly non-linear relationship between the input value and the display density as in the liquid crystal panel. Even if the grid projection is performed using a simple object, the shape can be measured with high accuracy.
(4) The calculation time for shape measurement can be greatly shortened.
(5) Coordinate data obtained by photographing with a plurality of cameras from different directions using the same reference plane can be easily synthesized because it can be expressed in the same coordinate system. Therefore, all-around measurement can be easily realized.

  That is, in the all-space table forming technique, a table or an approximate expression indicating the correspondence relationship between the phase and the three-dimensional space coordinates is created at the time of calibration, and a three-dimensional space coordinate distribution is obtained by referring to the table or the approximate expression at the time of shape measurement. Therefore, even if the projected grid is distorted, if the optical system is not changed during calibration and shape measurement, the influence of the distortion and the like is canceled and correct three-dimensional spatial coordinates can be obtained. In addition, by using the total space tabulation technique, when measuring 3D shapes, it is only necessary to refer to a table or approximate expression that shows the correspondence between the phase and spatial coordinates created in advance. It is possible to obtain a three-dimensional spatial coordinate distribution at a high speed without performing a complicated matrix calculation for obtaining. Note that the control unit 50 can be configured to perform the above-described calibration processing before shape measurement, and a table or an approximate expression indicating the correspondence between the phase and the spatial coordinates is a predetermined storage area of the memory 52 in the control unit 50. The control unit 50 can be configured to be stored in the memory and read out during shape measurement.

(Light source switching phase shift method)
As described above, the phase shift method disclosed in Patent Documents 1 to 3 performs phase shift by moving the grating by a piezo stage or the like. In this case, speeding up is difficult and expensive. On the other hand, in the light source switching phase shift method disclosed in Patent Document 4, a large number of linear light sources (for example, linear LED light sources) are used while the grating is fixed without moving the grating to shift the phase of the grating. The phase shift is performed by moving the shadow by switching. This eliminates the need for a moving device, and the measurement process can be speeded up by using a linear LED light source. However, in the light source switching phase shift method, the phase shift amount is proportional to the distance from the light source, and is different from the conventional phase shift method, which is a constant phase shift amount that equally divides the 2π phase. For this reason, Patent Document 4 shows that in the light source switching type phase shift method, it is effective to use the above-described all-space table formation technique together in order to cancel the error depending on the distance from the light source. .

  By using the total space table forming technique together with the phase shift method of the light source switching method, a 1: 1 correspondence relationship between the calculated phase and height is preliminarily displayed as a total space table by a calibration process before shape measurement. When the shape of the measurement target object 21 is measured based on the 1: 1 correspondence between the phase and the height even if there is an error between the phase shift amount and the theoretical value. If the same optical system is used, these errors are canceled out.

  For example, in the light source switching phase shift method, when only one of three LED light sources is turned on, a grid shadow is projected onto the measurement target object. When the light sources are switched sequentially and turned on, the shadow of the lattice changes. In other words, focusing on one point on the measurement target object, the phase of the grating shifts by switching the light source. However, the phase shift amount changes depending on the distance z from the light source. If the distance z is constant, the phase shift amount is constant. However, if the phase shift is performed integer N times, the total phase shift amount is not always 2π.

  In the case of the light source switching phase shift method, the phase analysis can be performed only in a special case where the total phase shift amount is an integer multiple of 2π, such as N = 5, 7 in the equation (3), and in other cases, the analysis is performed using only the formula. Although it is difficult to do this, it can be analyzed by combining it with an all-space tabulation technique. That is, in the all-space table forming technique, the position of the reference plane is associated with the phase and stored in the memory 52, and even if the phase value is slightly deviated, the phase value that is output using the equation (3) 3D coordinates can be obtained. Therefore, as long as the same optical system is used for calibration and measurement and the same phase value is used, the error when the total phase shift amount is not an integral multiple of 2π is canceled.

  That is, when the luminance value obtained by the light source switching phase shift method is substituted into the equation (3) to obtain the phase value, the phase value is different from the actual phase value and has an error (this actual and The different phase values will be called “skeptical phases”.) When the all-space table forming technique is used, if the relationship between the suspicious phase and the coordinates is a monotone function, this error is canceled and a correct coordinate value is obtained. By selecting an integer N that is closest to the number of shifts N (decimal number) when the phase is shifted to 2π by the equidistant phase shift method, the coordinate z in the widest possible range can be analyzed.

  FIG. 22 is a diagram showing the configuration of the shape measuring apparatus 1a based on the light source switching phase shift method and the operating principle thereof. Components similar to those in FIG. 1 are denoted by the same reference numerals. The shape measuring device 1a shown in FIG. 22 differs from the shape measuring device 1 shown in FIG. 1 in the arrangement of the grating plate 12, but the phase analysis process in the control unit 50 is the same.

Light source array 11 includes five light sources _{L -2,} L _-1, has a configuration that is linearly arranged to L 0, _{L 1} and _{L 2} at regular intervals. For example, as shown in FIG. 23A, each light source can be configured as a linear light source by arranging a plurality of LED chips 11c in the Y-axis direction and providing a directional lens 11d as necessary. Further, as shown in FIG. 23 (b), each light source can be configured as a single point light source by integrating a plurality of LED chips.

In the following description, with respect to the shape measuring apparatus 1a shown in FIG. 22, as shown in FIG. 23 (a), each light source is arranged at equal intervals in the X-axis direction, and five light sources L ₋₂ , L ₋₁ , L Assume that the light source array 11 and the grating plate 12 constituting ₀ , L ₁ and L ₂ are arranged in the Z-axis direction so as to be parallel in the X- and Y-axis planes.

In FIG. 22, for the linear light source as shown in FIG. 23A, the nth linear light source is L _n (n = −2, −1, 0, 1, 2). Consider that only the n-th line light source L _n is sequentially turned on while changing n, and the grid is projected onto the measurement target object 21. At this time, the transmittance distribution of the lattice line at z = d has a cosine wave shape (which can be analyzed even if it is not a cosine wave, but here it is considered as a cosine wave that can be theoretically handled easily), and the light source L _n The luminance distribution of the shadow of the grid irradiated with is expressed by the following equation.

Where φ is the phase of the one-dimensional lattice, a _g is the amplitude, b _g is the background luminance, x _g is the x coordinate on the lattice plane, e is the reference point (lattice origin) E (φ = 0) on the lattice plane. This is the distance between the point C. Of the central positions of the light transmission regions 12b constituting the one-dimensional grating of the grating plate 12, the one having the shortest distance from the Z-axis is defined as a reference point E (φ = 0), and the grating surface and Z The intersection with the axis is C.

This grating, the luminance distribution of the shadow of the grating when illuminated by a light source L _n is expressed by the following equation.

  When this equation (5) is replaced using equations (6) to (9), equation (10) is obtained.

Thus, when the light source was on L _n, the position S (x, y, z) the relationship between the luminance I _n and phase φ and the phase shift amount Ψ in can be formulated. Note that when the reflectance r of the measurement target object 21 is taken into consideration, a and b in the equation (10) may be multiplied by the reflectance r, but are omitted here for the sake of simplicity.

(Relationship between phase, phase shift amount and height)
When the phase shift amount Ψ is obtained, the following expression is obtained from the expression (9), and the height z is obtained. In this case, the isophase lines are contour lines.

The following equation is obtained from equation (8).

  When the phase shift amount ψ and the phase φ are obtained from the equations (10) and (12), the height z can be obtained regardless of the position of the camera.

  In this light source switching phase shift method, since the phase shift amount is not constant and changes with distance, the phase shift equation that divides 2π in Equation (3) into N equal parts cannot be used. Find the phase. The suspicious phase has an error with respect to the true phase, but if the range is limited, the relationship between the suspicious phase and the coordinates (x, y, z) is stored and held as an all space table by combining with the all space table forming technique. These errors will be canceled by setting them in advance. In this case, the shape can be analyzed without directly using the equations (4) to (9).

(Advantages of light source switching phase shift method)
The advantages of the light source switching phase shift method are many as follows.
(1) This is a full-field measurement method, and the entire measurement field can be analyzed at once.
(2) Since the phase analysis is performed, the resolution becomes high without being affected by the reflectance of the object.
(3) When the analysis is performed using the phase and the phase shift amount, the shape can be measured regardless of the position of the camera 13, and the camera can be easily arranged.
(4) If the analysis is performed using the phase and the phase shift amount, the dynamic range becomes wide and the resolution becomes high. In the light source switching phase shift method, the phase shift can be accurately performed by switching the light source, so that the phase connection can be easily performed, and measurement with high accuracy and a wide dynamic range is possible.
(5) The measurement range can be easily determined in accordance with the measurement application by changing the pitch of the LED chip, the distance from the grid, and the pitch of the grid.
(6) Since the brightness and power can be changed by changing the number of LED chips used as the light source, it is possible to measure a large object and a small object, and it is possible to analyze both at high speed and at low speed. In addition, the shape of the moving object can be measured.
(7) Since a phase shift device such as an expensive piezo stage is not used, the cost is low, and there is no part that physically moves, so that there is stability, a compact configuration is possible, and it is suitable for mass production.
(8) With respect to the light source, since light of a specific wavelength can be selected, if it is combined with an optical filter, it can be used in a place with noise light. If infrared or ultraviolet chips are used, measurement can be performed without affecting the operator or other measuring instruments.
(9) There is no moving part, it can be controlled by an electric signal, can be blinked at high speed, and high-speed photography is possible.
(10) A plurality of shape measuring devices 1 can be combined to measure the entire circumference, measure the inner circumference, and the like.
(11) Since it is not always necessary to use a lens other than the camera 13, it is possible to arrange devices that are not influenced by lens aberrations and that are close to theory, and that the lattice projection side uses the theoretical formula to analyze with high accuracy. Can do.

(Application from Shadow Moire Method)
In the shape measuring apparatus 1 of the present embodiment, the distance d is approximately the same as the height z (however, d <z) by application from the shadow moire method, but the relationship of Expression (10) does not change. Since the lattice is placed near the measurement target object 21, the lattice pitch can be made fine, and the accuracy is high.

  However, since the distance between the camera 13 and the grating plate 12 is large and the grating of the grating plate 12 is installed near the measurement target object 21 and is within the field of view of the camera 13, for example, the camera 13 captures the shadow of the grating from the gap of the grating. Will do. For this reason, moire fringes created by the lattice and the shadow of the lattice are observed. In this case, even if the grating of the grating plate 12 is physically shifted, the shadow of the grating is also shifted by the same amount, so that a phase shift of moire fringes cannot be obtained.

  Further, if both the lattice of the lattice plate 12 and the shadow of the lattice are reflected, measurement in the area where the lattice is reflected in the camera 13 cannot be performed. In order to solve this, in the technique assumed from the prior art, the image is taken by the camera 13 while moving the grating at a high frequency. However, this technique requires a mechanism for operating the grating with high accuracy and high frequency, resulting in an increase in apparatus cost.

  Therefore, as an aspect of the shape measuring apparatus 1 according to the present invention shown in FIG. 1, the phase shift is performed by switching the light sources of the light source array 11. FIG. 24 shows a shape measuring device 1b for the convenience of explanation of the operation principle of the shape measuring device 1 according to the present invention shown in FIG. Similar components are denoted by the same reference numerals.

(Phase analysis by light source switching phase shift shadow grating method)
As shown in FIG. 24, a grating plane is placed at a position d away from the light source plane, an object is placed behind the grating, and the phase Φ of the projected grating is obtained. The phase of the one-dimensional grating is obtained by the light source switching phase shift method at each pixel of the camera at the same position of z = 0 as the light source surface.

From the one-dimensional lattice equation (4), the phase of the U point at the coordinates x _G = x _u is as follows.

luminance I _n at the location S (x, z) by the n-th light source L _n is represented by the formula (5), the phase at this time is given by the following formula, is seen that the phase shift by switching the light source.

  By photographing the shadow of this lattice, the phase analysis can be performed using the suspicious phase by the whole space table forming technique.

(Phase shift analysis by light source switching phase shift shadow measurement method)
As shown in FIG. 24, in the light source switching phase shift shadow measurement method, the grating plate 12 is installed at a position separated by a distance d from the light source surface using the LED, and the one-dimensional grating of the grating plate 12 is placed on the measurement target object 21. Finds the phase Φ of the shadow of the grating projected onto. In each pixel of the camera 13, the height z or the height h from the grating plate 12 is determined by the phase difference (Φ _U −Φ _S ) between the phase Φ _U of the one-dimensional grating and the phase Φ _S of the shadow of the grating. Is required.

In other words, moire fringes formed by both the one-dimensional grating and the grating shadow projected onto the measurement target object 21 are observed, and the phase of the moire fringe is the phase of the one-dimensional grating Φ _G and the grating shadow. which they are the difference between the phase Φ _R. This calculation formula will be described below.

Here, the lens center V of the camera 13 is placed at the position of coordinates (v, 0, 0) on the x-axis. That is, the camera lens center V is separated from the origin by a distance v. Assume that one pixel of the camera 13 looks at a point U on the lattice plane or a point S on the measurement target object 21. As a whole field of view of the camera 13, the shadows of the one-dimensional lattice and the projected lattice on the measurement target object 21 are each seen in half. Moire fringes are observed due to this interference. The phase of the point S on these two objects is Φ _S , and the phase on the one-dimensional lattice plane is Φ _U. At this time, the phases of Φ _S and Φ _U are the phases of the lattices at the points G and U, respectively, and the phase of moire fringes is obtained from the phase difference. That is, the moire phase Φ _{M at} x is expressed by the following equation from the equations (13) and (14).

  On the other hand, Equation (16) is obtained from the similarity of two triangles formed by a perpendicular line from the point S on the x axis.

  Expression (16) is obtained from Expression (16).

By substituting equation (17) into equation (15), the moire phase Φ _M becomes the following equation.

  As can be seen from equation (18), it is a function of only z regardless of x and y. It can also be seen that moire fringes are phase-shifted at equal intervals by switching the light source.

  Here, it is assumed that the phase shift is performed so that the phase shift amount ψ becomes the expression (19).

  That is, if the phase shift is performed according to the relationship of the following equation (20), the total phase shift amount when the phase shift is performed N times as shown in the equation (21) is −2πM.

  In the case of the light source switching phase shift shadow measurement method, if one pixel is set to be one pitch of the grid, it is possible to capture only the moire fringes, not the grid, and easily measure the object 21 to be measured. Can be analyzed. Alternatively, if the N pixels are set to one pitch of the lattice, moiré fringes can be observed by taking the average of the N pixels.

(Phase analysis by light source switching phase shift shadow grating method using two cameras)
Normally, in the shadow moire method for measuring the shape of the measurement target object 21 through a grid, the image data of the part in which the grid is reflected cannot be used for analysis, and information on that part is lost. However, in the shape measuring apparatus 1 according to the present invention, when the camera 13 is translated in the x-axis direction by a half pitch of the grating, a portion that has not been seen so far can be photographed. Furthermore, if two or more cameras are prepared, the efficiency can be improved.

  As described above with reference to FIG. 7, phase-shifted N images can be taken by switching the light sources using two cameras according to the light source switching phase shift method. It is possible to analyze by the light source switching phase shift method using the indicated luminance value.

  The place where accurate equal division phase shift can be performed is limited to the place z where M in the equation (20) is an integer. In the vicinity, the amount of phase shift is slightly different depending on the height, so that it becomes possible to analyze it by combining it with an all-space table forming technique using a suspicious phase.

(Phase analysis by light source switching phase shift shadow moire method using two cameras)
By switching the light source in the configuration shown in FIG. 7, it is possible to perform phase shift without moving the grating, as shown by the equation (19).

On the other hand, the lattice shadow seen from the gap of the one-dimensional lattice shows data for a half period. Assume that this half cycle is shot with N pixels. If the shape does not change greatly in this half cycle, the data in the half cycle can be regarded as phase-shifted data by the equal division phase shift method. In the phase analysis method such as the equation (3), the phase value is obtained using N pieces of data that are phase-shifted by one equally divided phase. Similarly, even if N pieces of data that are newly phase-shifted by half a period are used, The phase value can be obtained. For example, when N = 4, assuming that the luminance values obtained by phase shifting are I _{0 to} I ₃ , the initial phase Φ is obtained by the following equation.

  Thus, the phase value of moire fringes can be obtained quantitatively. If the two cameras 13-1 and 13-2 shown in FIG. 7 are used, the phase values at all the pixels are obtained. In addition, as described above, the moire fringes themselves may be detected by obtaining the luminance average value of N pixels continuous in the x-axis direction corresponding to one pitch of the lattice.

In summary, with respect to the shape measuring apparatus 1 of the present embodiment, specific components include the following types.
(1) Regarding the light source array 11, there are one light source (no light source switching) and plural light sources (with light source switching).
(2) When the control unit 50 is controlled using the shadow grid method and when the control unit 50 is controlled using the shadow moire method.
(3) As a measurement mode, there is a case where there is a relative movement and a case where there is no relative movement.
(4) When the camera 13 is configured with one unit and with a plurality of units.

A configuration example of the shape measuring apparatus 1 configured by combining the above-described components is as follows.
(1) A configuration in which one light source is used, control using the shadow grid method is performed, relative movement is performed, and one or more cameras are provided.
(2) A configuration in which a plurality of light sources are used, control using the shadow grid method is performed, there is no relative movement, and one or more cameras are provided.
(3) A configuration in which a plurality of light sources are used, control using the shadow grid method is performed, there is relative movement, and one or more cameras are provided.
(4) A configuration in which one light source is used, control using the shadow moire method is performed, relative movement is performed, and one or a plurality of cameras are provided.
(5) A configuration in which a plurality of light sources are used, control using the shadow moire method is performed, there is no relative movement, and one or more cameras are provided.
(6) A configuration in which a plurality of light sources are used, control using the shadow moire method is performed, there is relative movement, and one or more cameras are provided.

  When performing control using the shadow grid method, N pixels are allocated to 1/2 grid pitch, or N pixels are allocated to 1 grid pitch, so that a precise shape is not necessarily combined with the total space table forming technique. Measurement is possible. On the other hand, when the pixel size is larger than one grid pitch, only moire fringes can be observed. In this case, it is set as the measurement mode which performs control using the shadow moire method. That is, even when the pixel size is smaller than 1/2 grid pitch, when the pixel size is smaller than 1 grid pitch, or when 1/2 grid pitch or 1 grid pitch is not N pixels, it is almost 1 Moire fringes can be obtained by adaptively smoothing each pixel value according to the width of the pitch, an integral multiple thereof, or the application, and a measurement mode in which control using the shadow moire method is performed can be obtained.

  In addition, when phase shift is performed by switching light sources, N pixels are not strictly assigned to 1/2 lattice pitch or 1 lattice pitch by installing so as to perform phase shift by 2π or π by switching light sources N times. Can also perform phase analysis. In particular, the error can be canceled by combining with the total space table forming technique.

  In addition, the pixel row related to the camera 13 does not necessarily have to be perpendicular to the grid line (ie, the y-axis direction) (ie, the x-axis direction), and N pixels are assigned to 1/2 grid pitch or 1 grid pitch. Any mode may be used. Similarly, with respect to relative movement, the moving direction need not be perpendicular to the grid line (ie, the y-axis direction) (ie, the x-axis direction). In particular, when the moving direction is oblique with respect to the grid line, a phase change also occurs in the y-axis direction, and phase analysis processing separated in the x and y-axis directions becomes easy. In particular, even if a slight error occurs in the parallelism, verticality, and equal spacing of the measurement pixel rows of the light source array 11, the grating plate 12, and the camera 13, even if a slight error occurs, it is combined with the total space table forming technique. By adopting the measurement mode, errors due to these are canceled and the phase analysis process can be performed accurately.

  Although the above embodiments have been described with specific examples, the present invention can be applied to various embodiments. For example, instead of moving the measurement target object, when the shape measurement apparatus is moved, or when both the measurement target object and the shape measurement apparatus are moved, the pitch of the light source and the grating may be determined from the relative speed. The control unit 50 can be configured as a computer. When the control unit 50 is configured as a computer, the computer stores a program for realizing each component of the control unit 50 in the memory 52. . The central processing unit (CPU) provided in the computer reads a program or processing data in which processing contents for realizing the function of each component are described from the memory 52 as appropriate, and each component of the control unit 50 Can be realized on a computer. Here, it goes without saying that the function of each component may be realized by a part of hardware.

  According to the present invention, since the entire measurement target object can be measured with higher accuracy, it is useful for precise shape measurement of the measurement target object and inspection such as scratches / damage.

1, 1a, 1b Shape measuring device 10 Projector 11 Light source array 12 Grating plate 12a Grating light shielding area 12b Grating light transmission area 13 Camera 13-1 First camera 13-2 Second camera 14 Reference plate 21 Measurement target object 50 Control unit 51 Control unit 52 Memory 511 Measurement range setting unit 512 Light source determination unit 513 Camera photographing processing unit 514 Pixel-based luminance calculation unit 515 Phase calculation unit 516 Three-dimensional spatial coordinate calculation unit

Claims (11)

A shape measuring device that measures the shape of an object to be measured without contact,
A light source array comprising at least one point light source or linear light source;
A lattice plate that forms a one-dimensional lattice by light transmission regions composed of straight lines arranged at equal intervals;
A camera that captures an object to be measured on which a shadow of a lattice by the one-dimensional lattice is projected by turning on a light source in the light source array;
The phase of the shadow of the grid projected onto the measurement target object by the light source or the phase of the moire fringe composed of the shadow of the one-dimensional grid and the grid is calculated for each pixel of the camera, and the calculated phase is used as the basis. And a control unit for obtaining a three-dimensional spatial coordinate related to the measurement target object,
The grid plate is disposed within the field of view of the camera;
When the phase of the shadow of the lattice is calculated, the captured image by the camera is assigned a predetermined number of pixels in a predetermined range in an area other than the lattice portion of the lattice plate, and includes the one-dimensional lattice and the shadow of the lattice. When calculating the phase of moire fringes, a set of one-dimensional lattice and a predetermined range including the shadow of the lattice are set to assign a predetermined number of pixels,
The control unit calculates a phase of a shadow of the grid or a phase of a moire fringe composed of the one-dimensional grid and the shadow of the grid based on the pixel value of the predetermined number of pixels. . The light source array is composed of at least three point light sources or linear light sources arranged at equal intervals corresponding to a plurality of types of phase shifts.
The control unit determines a light source for performing phase shift of a shadow of the projected grid among the at least three light sources, sequentially lights the determined light sources, and is projected onto the measurement target object, respectively. The camera is controlled to capture the shadow of the lattice or the moire fringe, and the phase of the shadow of the lattice or the phase of the moire fringe projected on the measurement target object by the determined light sources is determined for each pixel of the camera. The shape measuring apparatus according to claim 1, further comprising means for calculating each of the two. The light source array consists of one point light source or a linear light source,
The shape measuring device is in an arrangement relationship that moves relative to the measurement target object at a predetermined speed on a plane parallel to the grid plate,
The control unit controls the camera to capture a shadow of the grating or the moire fringes projected onto the measurement target object in synchronization with the predetermined speed, and the grating projected onto the measurement target object The shape measuring apparatus according to claim 1, further comprising means for calculating a phase of a shadow of the image or a phase of the moire fringe for each pixel of the camera. The light source array is composed of at least three point light sources or linear light sources arranged at equal intervals corresponding to a plurality of types of phase shifts.
The shape measuring device is in an arrangement relationship that moves relative to the measurement target object at a predetermined speed on a plane parallel to the grid plate,
The control unit determines a light source that performs phase shift of shadows of the projected grid among the at least three light sources, and sequentially turns on each of the determined light sources and performs the measurement. The camera is controlled so as to photograph the shadow of the lattice or the moire fringe projected on the target object, and the phase of the shadow of the lattice projected on the measurement target object by the determined light sources or the moire fringe The shape measuring apparatus according to claim 1, further comprising means for calculating the phase of each of the pixels of the camera. The camera is composed of a plurality of cameras that shoot a measurement target object on which a shadow of a lattice by the one-dimensional lattice is projected by turning on a light source in the light source array with different lines of sight,
The control unit controls the plurality of cameras so that shadows of the lattice or the moire fringes projected on the measurement target object are photographed at different positions, and based on images photographed by the plurality of cameras. The phase of the shadow of the lattice or the phase of the moire fringe formed by the shadow of the lattice projected on the measurement target object is calculated for each pixel of the camera, respectively. The shape measuring apparatus according to one item.   The control unit obtains a three-dimensional spatial coordinate relating to the measurement target object based on a table or an approximate expression in which the phase of the one-dimensional lattice and the three-dimensional spatial coordinate prepared in advance for each pixel are associated one-to-one. The shape measuring apparatus according to claim 1, wherein the shape measuring apparatus includes:   Regarding the pixel row at the time of photographing by the camera, the number N of pixels that are parallel to the plane of the grating plate and perpendicular to the straight line of the light transmission region of the one-dimensional grating (N is an integer of 1 or more) 7. The apparatus according to claim 1, further comprising means for calculating a luminance value of the moire fringes from an average value of luminances of the continuous N pixels in correspondence with one pitch of a one-dimensional lattice on the lattice plate. The shape measuring device according to claim 1. The camera is composed of a plurality of cameras that shoot a measurement target object on which a shadow of a lattice by the one-dimensional lattice is projected by turning on a light source in the light source array with different lines of sight,
With respect to the pixel row in the measurement range by each of the plurality of cameras, the number N of pixels that are parallel to the plane of the grid plate and continuous in the direction perpendicular to the straight line of the light transmission region of the one-dimensional grid (N is an integer of 1 or more) ) Corresponding to ½ pitch of the one-dimensional lattice in the lattice plate, and at the first pixel position in the continuous pixel number N, the shadow of the lattice or the average value of the luminance of the continuous N pixels. 6. The shape measuring apparatus according to claim 5, further comprising means for calculating a luminance value of the moire fringes. The line of sight of the camera is arranged so that the half cycle of the phase of the shadow of the grid projected onto the measurement target object in a region other than the grid portion of the grid plate is a plurality of pixels,
The shape measuring apparatus according to claim 8, wherein the control unit includes means for calculating a phase of a shadow of the lattice from each luminance value obtained by phase shifting by the number of pixels. A portion of the one-dimensional lattice is transparent;
The shape measuring apparatus according to claim 1, wherein the line of sight of the camera is arranged so as to photograph a shadow of the lattice from the transparent portion. A light source array composed of at least one point light source or a linear light source, a lattice plate that forms a one-dimensional lattice by light transmission regions composed of straight lines arranged at equal intervals, and the one-dimensional lattice formed by turning on the light source in the light source array A shape measurement method for measuring the shape of a measurement target object in a non-contact manner by a shape measurement device including a camera that captures a measurement target object on which a grid shadow is projected, and a control unit,
The control procedure of the control unit is:
When calculating the phase of the grid shadow with respect to the image taken by the camera, a predetermined number of pixels are assigned to a predetermined range in an area other than the grid portion of the grid plate, and the grid includes the one-dimensional grid and the grid shadow. A step of setting a measurement range by assigning a predetermined number of pixels to a predetermined range including a set of the one-dimensional lattice and a shadow of the lattice when calculating the phase of moire fringes;
Controlling photographing of the camera so as to photograph a measurement target object on which a shadow of a lattice by the one-dimensional lattice is projected by turning on a light source in the light source array;
A luminance value is calculated from a pixel value in the measurement range from the captured image, and from the luminance value, the phase of the shadow of the grid projected onto the measurement target object by the light source or the shadow of the one-dimensional grid and the grid Calculating the phase of the moiré fringes for each pixel of the camera;
Based on the calculated phase, obtaining three-dimensional spatial coordinates related to the measurement target object;
A shape measuring method comprising: