JP3282331B2 - 3D shape measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a three-dimensional shape measuring apparatus for measuring a three-dimensional shape of a target object and displaying a measurement result on a screen.

[0002]

2. Description of the Related Art Three-dimensional information is used in various industrial fields such as measurement of natural objects and living bodies, and visual recognition of robots. If the shape of a stationary object is measured, it does not matter even if it takes several seconds for the measurement. However, in an apparatus that performs three-dimensional measurement of a moving object, such as visual recognition of a mobile robot, an apparatus that measures a three-dimensional shape at high speed and with high accuracy is required. Conventionally, among the means for recognizing the three-dimensional shape of an object, a light cutting method is often used as the most practical means. In the light cutting method, as shown in FIG. 1, a target object is irradiated with a slit-shaped laser beam S, and a slit image of the target object 1 corresponding to the slit light S is captured on an imaging surface of a camera. Then, the spatial coordinates of the point p on the target object corresponding to a certain point p ′ on the slit are represented by a plane S defined by the slit light.
, And the coordinates of the intersection of a point p ′ and a straight line L connecting the center point O of the lens of the imaging device. As described above, the spatial coordinates of the point group on the object surface corresponding to each point on the slit image are obtained from one image, and the horizontal movement of the slit light and the image input are repeated to obtain a three-dimensional image of the entire target object. Information can be obtained.

A CCD area sensor is often used as an image sensor. To read an image in this configuration, all pixels of the CCD are read in response to scanning of one line of slit light, and the entire surface of a target object is scanned. Takes several times the time required for scanning one line. But,
Since a high-speed input device is required to perform three-dimensional measurement of a moving object, various devices with high-speed input have been proposed.

The device proposed in Japanese Patent Publication No. 56-501704 is a device which divides a photoelectric conversion element into blocks, reads out the blocks in parallel, and sequentially reads out each block in series, thereby shortening the time required for reading the entire range and reducing the processing time. Speeding up. Also, JP-A-64-46605 discloses that
PSD (Position Sensitive Device) as image sensor
Devices using arrays have been proposed. In this device, P
Since the output signal representing the incident position information of the slit image from the SD sensor is output in real time, high-speed measurement is possible without requiring a waiting time for one frame. In other proposals, there is a device that lights up four times in different patterns using 15 light sources and combines the obtained four times of data to obtain an entire image. In this apparatus, since the control angle of the rotating mirror can be reduced, the whole image can be obtained at high speed.

[0005]

However, in the above-mentioned first conventional example, the blocks are read out in parallel, but each block is read out in series in order, and the data is read out to the area where the slit light is not incident. Therefore, it takes time for the number of blocks. In the second conventional example, the position of the incident slit light can be detected because the PSD array is used. However, this is not a problem when only the projected slit light is not affected by external light. However, when the influence of the external light enters, the photocurrent due to the external light becomes a noise component, so that the detected position is different from the position where only the slit light is used. In addition, since the PSD detects the position, the difference from the case of only external light cannot be used, and there is a problem that accurate measurement cannot be performed. The third conventional example is 15
Since a plurality of light sources are used, control and processing of these light sources are required, resulting in a problem that the configuration becomes complicated and the device becomes large.

SUMMARY OF THE INVENTION It is an object of the present invention to solve these problems and to provide a three-dimensional shape measuring apparatus which processes a three-dimensional shape of a target object at high speed and is effective for a moving object.

[0007]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention relates to a three-dimensional shape measuring apparatus for projecting slit light onto a target object, taking an image with a two-dimensional image sensor, and measuring the shape of the target object. The two-dimensional image sensor selectively reads a band image in a region where slit light is assumed to be incident.

[0008]

In the above arrangement, the target object is irradiated with slit light, and the whole area of the two-dimensional image pickup element is not read out at the time of image reading. And selects and reads only the band-shaped image.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. First, FIG. 2 is a schematic block diagram of the entire apparatus according to the present invention. This apparatus can be roughly divided into a light projecting optical system 2 for irradiating a laser beam output from a semiconductor laser 5 as a slit-shaped light beam to a target object 1, and a light receiving optical system for guiding the irradiated laser beam to image sensors 24 and 12. There is a system 3, and the light projecting optical system and the light receiving optical system are arranged on the same rotating base 4. In addition to the optical system, the optical system includes a signal processing system that processes a signal output from the sensor to generate a pitch shift image and a color image, and a recording device that records the generated image. The solid arrows shown in FIG. 2 indicate the flow of electrical signals of the image signal and the control signal, and the broken arrows indicate the flow of projected light. A detailed description of these optical systems will be described later.

To explain the outline of the signal processing system, an image obtained by the distance image sensor 12 is obtained by subtracting an image 18a when the slit light is projected and an image 18b when the slit light is not projected. Incident light centroid position calculation processing 1
9. Pitch shift information calculation processing 20 and pitch shift image generation processing 21 are performed. The pitch shift image obtained is NTS
The data is output to the output terminal 50 by the C conversion process 27 or transferred to the SCSI terminal 49 or the built-in recording device 22 as digital data for use. The image obtained by the color image sensor 24 is converted into an analog
, A color image generation process 26 is performed. The obtained color image is subjected to NTSC conversion processing 28 and output terminal 51
To the SCSI terminal 49 or the recording device 22 as digital.

FIG. 3 is a perspective view showing a schematic configuration of the entire apparatus. In this embodiment, the length of the slit light is 2 in the length direction.
A 256 × 324 distance image generation system having distance information of 56 points and 324 points in the slit scanning direction will be described as an example. The LCD monitor 41 displays a color image picked up by the color image sensor 24, three-dimensional data recorded in a recording device inside or outside the apparatus, various kinds of information, a selection menu, and the like. The cursor key 42, the select key 43, and the cancel key 44 are operation members for selecting an image, setting various modes from a menu, and the like.
A zoom button 45 changes the focal length of the light projecting / receiving optical system, and an MF button 46 performs manual focusing. A shutter button 47 captures a distance image when turned on in a shutter mode described later. As a recording device of the captured image, a magneto-optical disk (hereinafter, referred to as MO) or a mini disk (hereinafter, referred to as MD) built in the device.
). Terminal 4
9 is a terminal for digitally inputting / outputting a signal such as an image.
SI or the like. A pitch shift image output terminal 50 and a color image output terminal 51 are terminals for outputting an image as an analog signal, and output a video signal such as NTSC.

The light projecting optical system scans slit light long in the horizontal direction in the vertical direction. Light from the semiconductor laser 5 is rotated by a polygon mirror 7, a condenser lens 10,
The light is projected onto the target object via the light projecting zoom lens 11 and the like. The light receiving optical system picks up an image with a distance image sensor 12 and a color image sensor 24 arranged on a light receiving and imaging surface via a light receiving zoom lens 14, a beam splitter 15, and the like. The optical system and the imaging system will be described later in detail.

While the distance image sensor 12 accumulates one image, the slit light from the light projecting system is moved downward by one pixel pitch of the distance image sensor 12 by the polygon mirror 7 which is constantly rotating. Scanned. The range image sensor scans and outputs the stored image information and stores the next image. The distance information of 256 points in the length direction of the slit light can be calculated from the image output by this one output. Further, by repeating mirror scanning and image capturing 324 times, a distance image of 256 × 324 points is generated.

The range of the distance to the target object measured with respect to one slit light is limited to the closest distance to be measured and the longest distance to be measured. Therefore, the slit light is reflected by the object and enters the image pickup device. The range is limited to a certain range. This is because the light projecting system and the light receiving system are set apart from each other by the base line length (length 1). This is shown in FIG.
And the direction perpendicular to the image sensor surface for the distance image is defined as the Z axis. The position of the broken line indicated by d is the measurement reference plane, and the distance from the element surface is d.

For this reason, first, in the measuring device, the position of the center of gravity of the laser beam received in 256 lines from the input image is determined by the object distance output from the autofocus unit and the azimuth of the slit light to be projected, ie, the scanning start. The calculation is performed as a shift amount from the measurement reference plane determined based on the time from the measurement reference plane. The calculation of the pitch shift amount will be described with reference to FIG. 4. First, FIG. 4 shows a light amount distribution generated by slit light projected on the target object surface. The squares shown in the lower part of the figure indicate the area that each element of the range image sensor looks at, and the squares are numbered from 1, 2, 3, 4,... Since the slit light having an extremely narrow slit width is scanned by the rotation of the polygon mirror 7 by one pitch of the distance image sensor during one image accumulation, the light quantity distribution upon input of one image is equivalent to one pitch of the distance image sensor. Is a rectangular light quantity distribution having a width of

In order to calculate the distance information in the Z-axis direction for each pixel of the range image sensor, it is desirable that such a rectangular light quantity distribution having one pitch width is used. When the width of the light quantity distribution is equal to or more than one pitch, the measured distance information is obtained as a load average of the received light intensity over the adjacent area, and accurate distance information cannot be obtained.

It is assumed that a stepped object surface indicated by a halftone dot in FIG. 4 exists under such light amount distribution, and slit light is projected from a direction perpendicular to the object surface. An elongated rectangular parallelepiped indicates a slit light amount distribution, and a hatched area indicates a light projection slit image. Then, assuming that the light receiving system optical axis Oxp is provided in a direction inclined to the left from the light projecting system optical axis Oxa, the light receiving slit light amount distribution on the light receiving surface is as shown in FIG. Become. This received light quantity does not include the steady light component,
It is desirable to remove the stationary light component other than the laser light component. For that purpose, images in a state where the laser light is not irradiated and an image in the irradiated state are input, and the difference between the two is used. The cells shown below indicate the respective element regions of the range image sensor.

An optical filter having an anisotropy is provided on the front surface of the range image sensor so as to reduce the resolution in the width direction of the slit light without lowering the resolution in the length direction of the received slit light. This filter produces a Gaussian light quantity distribution as shown in FIG.
The light receiving position can be calculated at a resolution finer than the pixel pitch by obtaining the center of the light amount distribution from each sensor in each of the columns 1, 2, 3, 4,. . As described above, when detecting the slit light incident position, the width of the slit light incident on the sensor is not narrowed but a distribution having a width of about 5 to 6 pixels is obtained by using a filter. This is because if the width is smaller than the width of one pixel, only a position detection resolution equivalent to the pixel pitch can be obtained.

The first column barycentric position G1 is obtained from the light quantity distribution D1 incident on the first column, and similarly the second, third, fourth,...
, The center of gravity is calculated for each column by calculating the center of gravity position G2, G3, G4,. As shown in the figure, the optical axis of the light projecting system is a direction perpendicular to the object plane, but the optical axis of the light receiving system is a direction inclined to the left. Therefore, in the case of a target object having a step as shown in FIG. In the portion (third and fourth rows) higher than the center of gravity of the portion (first and second rows), the center of gravity is located at a position shifted to the right. Note that FIG. 5 shows only two types of distributions, a first column distribution D1 and a fourth column distribution D4, but the second column distribution D2 is the same as the first column distribution D1, The distribution D3 in the column is the same as the distribution D4 in the fourth column. FIG. 6 shows the relationship between the light amount distribution and the position of the center of gravity in a plan view. Since the distributions of the first and second columns are the same, the centroids G1 and G2 obtained are detected as the same position, and the third and fourth columns have the same distribution, so that the centroids G3 and G4 are detected as the same position.

As described above, 256 incident light positions are obtained from the strip image for one slit, and 324 positions are further obtained.
324 images are obtained by performing the same calculation for the slit projected in the azimuth direction, and a pitch shift image composed of 256 × 324 points is obtained. The obtained pitch shift image is an image of only the position information of the slit light, and calibration (correction) from a table of detailed data such as correction of lens aberration is required to obtain an accurate distance image from the image. In this method, the lens aberration estimated from the focal length f and the focus position d of the photographing lens is calculated and corrected, and the vertical and horizontal distortions of the camera are corrected. The same processing is performed for a color image. The data required at this time is information on various measurement lenses, that is, the focal length f and the focus position d. In the system of the present embodiment, calibration is performed on a computer system, and the measurement apparatus (shown in FIG. 3) can be connected via a terminal such as SCSI or shared with a recording medium such as MO. To

As described above, a color image and a pitch shift image are output as digital signals from a terminal such as SCSI or an analog video signal from an output terminal such as NTSC from the measuring apparatus main body. Output to the computer as a digital signal. Also, when recording on a recording medium using a drive device 48 such as an MO or MD built in the main body, an image and various data are recorded.

The captured pitch shift image and color image are transferred to a computer connected to a measuring device together with various types of photographing lens information, and the computer calculates information on the distance to the target object from the transferred pitch shift image and photographing lens information. Calibration and conversion are performed on the held distance image. After the calibration, for the pitch shift image, a conversion curve between the shift distance and the measurement distance stored for each of the focal length f and the focus position d, the vertical and horizontal positions in the screen, and the XY positions is derived, and the conversion curve is obtained. The pitch shift image is converted into a distance image on the basis of.

Conversion to a distance image is well known,
For details, see the IEICE Technical Meeting, PRU91-113.
[Geometric correction of images that do not require camera positioning] Onodera and Kanaya, IEICE Transactions D-II vol. J74
-D-II No.9 pp.1227-1235, '91 / 9 [High-precision calibration method of range finder based on three-dimensional model of optical system] is disclosed in Ueshiba, Yoshimi, Oshima and others.

Next, the measuring device according to the present invention will be described in detail. First, the optical system will be described. As shown in FIGS. 1 and 2, the slit light S is emitted from the slit light irradiating device (light projection optical system) 2 to the target object 1 at the time of distance image capturing. The slit light irradiating device 2 includes a light source, for example, a semiconductor laser 5, a condenser lens 6, an optical system of a polygon mirror 7, a cylindrical lens 8, a condenser lens 10, and a projection zoom lens 11. The polygon mirror 7 is not limited to this, and may be a rotating mirror such as a resonance mirror or a galvanometer mirror.

The cylindrical lens 8 is a lens that does not focus and forms a focal line parallel to the cylinder axis because the convex side is not spherical but cylindrical. The polygon mirror 7 is formed by a number of mirrors around a rotation axis, and scans light incident upon rotation in one direction for each mirror surface.

The structure of the light projecting optical system will be described with reference to FIG.
FIG. 7A is a side view of the light projecting system, and FIG. 7B is a top view. Note that, in FIG. 7B, some of the components that overlap when illustrated are omitted. In FIG. 7A, the slit light has a length in a direction perpendicular to the paper surface, and the solid line drawn from the semiconductor laser 5 to the condenser lens 10 indicates the optical path, and Indicates a drawing line for indicating a position where the slit light is re-imaged by a broken line. In FIG. 7B, the slit light has a length in the vertical direction of the drawing, and the solid light from the semiconductor laser 5 to the cylindrical lens 8 indicates the optical path, and the image after the condenser lens 10 is re-imaged. The drawing line for indicating the position to be displayed is indicated by a broken line. The one shown by the one-dot chain line between the cylindrical lens 8 and the condenser lens 10 is as follows.
FIG. 3 schematically shows a state in which a laser beam that has proceeded in a substantially point-like manner is converted into a slit light having a wide width by a cylindrical lens 8. 7 (a) and 7 (b)
Then, the slit light is re-imaged at a position indicated by a vertical straight line (two-dot chain line) at the left end.

The collimator lens 6 (condenser lens 6 in FIG. 2)
Has a lens power for condensing output light (e.g., emission wavelength of 670 nm) of the semiconductor laser 5 on the condenser lens. The laser light passing through the collimator lens 6 is deflected by the polygon mirror 7 in a direction perpendicular to the length direction of the slit light. This deflection makes it possible to scan the slit light on the object plane. The laser light deflected by the polygon mirror 7 first enters the fθ lens 29. The fθ lens 29 is a lens arranged to correct a non-linear component since the moving speed of the slit light on the object plane becomes non-linear with respect to a constant rotation speed of the polygon mirror 7. The collimator lens 30 disposed next is for directing the light beam incident on the condenser lens 10 from the direction scanned by the polygon mirror 7 to the direction perpendicular to the condenser lens, thereby improving the projection efficiency. The laser light is converted by the cylindrical lens 8 into slit light having a length in the horizontal direction (in the direction perpendicular to the paper surface in FIG. 7A), and the condenser lens 1
The light is condensed on the pupil plane 0, forms an image, and is projected on the object as slit light having a very narrow width.

Condenser lens 1 arranged on image plane (imaging plane) 10p of projection zoom lens 11
The slit light once imaged at 0 is transmitted to the projection zoom lens 1
1 and is projected on the target object. The size of the image forming surface is set to 1/2 inch, 1/3 inch or the like according to the size of the image sensor, and is set to 1/2 inch in this embodiment. The slit light has a length in the horizontal direction generated by the cylindrical lens 8, and is scanned at high speed in a direction perpendicular to the length direction of the projected slit light as the polygon mirror rotates. At this time, the focus position of the light projecting zoom lens is determined by the auto focus sensor 31 provided in the imaging system.
Are controlled by the AF driving system 17 at the same value and simultaneously with the imaging system according to the distance to the target object plane based on the signal from The auto focus sensor 31 is the same as that generally used in a still camera. Also, the focal length is controlled by the same value as that of the imaging system and at the same time based on an operation from the user or the system.

The polygon mirror 7 is connected to a light projection / scanning drive system 9 composed of a polygon mirror driving motor and a polygon mirror driving driver, and the rotation thereof is controlled by these components. Reference numeral 33 denotes a scanning start sensor using a photodiode arranged on the side of the condenser lens 10, which monitors the scanning start timing to determine whether laser scanning has reached a stable state.

This light projecting system has a zoom function, so that the magnification of the target object 1 can be adjusted to a required value. The zoom function includes two functions: a power zoom (PZ) in which the user can arbitrarily select an angle of view, and an auto zoom (AZ) that automatically adjusts to a preset angle of view. In response to zooming of the light receiving optical system 3, the light projecting optical system 2 is controlled by the AZ drive system 16 so that the angle of view always coincides, and performs zooming so that the optical magnifications are always equal. Referring to the schematic diagram of FIG. 8, the relationship of the slit light projection by zooming is as shown in equations (1) to (3). θ = α1 × 1 / f (1) φ = α2 × 1 / f (2) ψ = α3 × 1 / f (3) Based on one point of the light projecting system, θ is a pitch shift image of 256 points in a row The angle at which the extremely thin slit light moves during the integration of one image to obtain the angle, φ is the angle representing the length of the slit light on the target object, and ψ is the total of 324 slit light beams on the target object. At the scanning angle, the slit light scans from the position shown by the solid line in the direction of the arrow and moves to the position shown by the broken line. f represents the focal length of the light projecting lens. The width of the slit light itself is set as narrow as possible. α1, α2, α
3 is a proportionality coefficient, and these angles θ, φ, and ψ are proportional to the reciprocal of the focal length f.

An image pickup device (light receiving optical system) including a color image pickup system and a distance image pickup system at a position perpendicular to the slit light irradiation device (projection optical system) 2 and separated from the slit light irradiation device 2 by a base line length l. ) 3 is arranged on one rotating base 4, and the configuration of this light receiving optical system is shown in FIG. 9. The light receiving optical system 3 includes an imaging zoom lens 14, an autofocus unit 31, a beam splitter 15, various filters 61,
62, color CCD image sensor 24, range image sensor 1
2 is comprised.

The received light is split by a beam splitter 14 s disposed in the imaging system zoom lens 14, and a part of the light is guided to an autofocus unit 31. This A
The F unit 31 functions to measure the approximate distance to the object surface and adjust the focus of the light projecting system and the light receiving system lens. Used.

The other light beam split by the beam splitter 14s disposed in the imaging system zoom lens 14 is further split by the beam splitter 15 disposed after the imaging system zoom lens into two light beams of transmission and reflection. It is guided to the range image sensor 12 and the color image sensor 24. This beam splitter 15 is used for the distance image sensor 12.
A long wavelength component as a light beam incident on the laser beam, here a long wavelength component including a laser wavelength component (670 nm) of about 650
It has an optical property of transmitting wavelength components longer than nm and reflecting other wavelength components.

Since the reflected short-wavelength component has almost all wavelength components of visible light, it is impossible that the color information is disturbed in a normal case. The reflected light flux passes through a low-pass filter 61 such as a quartz filter for preventing false resolution, and forms an image on the single-plate color image sensor 24. The single-plate color image sensor 24 is a CCD image sensor used in a video camera or the like, and has an RG
This is an image sensor in which dye filters of B, or complementary color yellow Ye, cyan Cy, magenta Mg, and green G are arranged on a mosaic to extract color information. Note that green can be used instead of a luminance signal. FIG. 10 shows a wavelength band of light received by the complementary color image sensor. The color image pickup device receives light in a wavelength region reflected by a beam splitter having a characteristic represented by a reflectance h. The curves show yellow Ye, cyan C
This is the spectral sensitivity of each pixel with a dye filter of y, magenta Mg, and green G.

On the other hand, the luminous flux of the long wavelength component transmitted by the beam splitter 15 passes through a filter of an infrared ray (Infrared Ray, hereinafter referred to as IR) cut to extract only a laser beam component (wavelength 670 nm). And a distance image sensor 12 which passes through a low-pass filter such as a quartz filter.
Image on top. In the configuration diagram of the light receiving optical system shown in FIG. 9, both the IR cut and low pass characteristics are indicated by one filter 62. FIG. 11 shows a wavelength band (shaded area) of light received by the range image sensor 12. The wavelength region shorter than the wavelength of the laser light is controlled by the beam splitter 15 (the transmittance indicated by the solid line), and the longer wavelength region is controlled by the IR cut filter
(The transmittance indicated by the broken line).

The low-pass filter 62 used here is different from the above-described purpose of preventing false resolution for a color image, and is used to detect the position of the received laser beam at a resolution finer than the imaging element pitch when calculating the distance data. It is intended to have a complementary function. Therefore, unlike the isotropic optical characteristics of the low-pass filter 61 at the time of a color image, the resolution does not decrease in the length direction of the received slit light, and decreases in the width direction of the slit light. It is desirable to have an optical property having anisotropy that causes the anisotropy.
As a means for realizing such optical characteristics, a low-pass filter using diffraction such as a single-layer quartz crystal filter or a grating can be used. However, this low-pass filter is not indispensable in the system configuration, and an analog filter for the sensor output later, Alternatively, the function can be substituted by a digital filter after digital conversion of the sensor output.

Next, scanning of the image pickup devices 12 and 24 will be described. 12p and 24p shown side by side with the image pickup devices 12 and 24 in FIG. 9 are image pickup devices in which 12 and 24 are shown in a plan view for explanation. In the CCD image pickup device, scanning in the vertical direction is generally slower than scanning speed in the length direction (horizontal direction) of the horizontal registers 12h and 24h. Therefore, a color image obtained by the image sensor 24 (24p) is usually subjected to analog signal processing in accordance with the output of a horizontal transfer line of a CCD that performs high-speed scanning, and the NTSC is sequentially performed.
And output the image to a monitor. When attempting to output a pitch-shifted image to the same monitor, it is necessary that the distance image data be generated in the same direction as the horizontal scanning direction of the color image and in the same position order, without having to store position information. It is desirable from the viewpoint of simplification of functions required for the memory as well as reduction of the memory capacity.

Accordingly, it is desirable that the length direction of the slit for projecting the distance data is in the high-speed scanning direction of the color image image sensor, that is, generally in the horizontal scanning direction. Further, it is desirable that the scanning direction in which the slit is projected is the vertical scanning direction of the color image. That is, it is desirable that the projected slit light be scanned from top to bottom.

Therefore, a three-dimensional shape measuring apparatus using a light-section method, in which the object to be measured as in this embodiment is captured by two types of image input sensors, a color image sensor and a distance image sensor, is used.
In a two-dimensional input camera, the slit light having a slit length in the horizontal scanning direction of the color image pickup device is projected, and the slit is scanned in the same direction as the vertical scanning direction of the color image pickup device, thereby reducing memory. , Leading to a reduction in memory requirements. In addition, the direction in which the band image is read from the range image sensor at high speed with respect to such slit light is limited. The horizontal direction in which the range image sensor can perform high-speed scanning is parallel to the horizontal scanning direction of the color image sensor. It needs to be in the right direction. In other words, the positional relationship and the scanning direction relationship between these image sensors and the incident slit light are as shown in FIG.

In the optical system having the above-described configuration of the image pickup section, the following two configurations are newly required.

First, a color image and an image for generating a distance image are input from the same lens. Must be performed independently of each other. When measuring a close measurement target in a dark place, the brightness for a distance is high, but the brightness for a color image is low. Conversely, when measuring a distant measurement target in a bright place, the brightness for a distance is low, but the brightness for a color image is high. Therefore, in this light-receiving zoom lens, the control by the aperture, which is an exposure adjusting means in a general lens, is not performed, and is always kept in an open state.

The exposure control of a color image is adjusted by an accumulation time by an electronic shutter function of an FIT-CCD or the like generally used as a color image sensor. Normally, an electronic shutter function of a FIT-CCD or the like used as a color image sensor enables control of an accumulation time of 1/60 to 1/10000 second. In order to secure a wider dynamic range, under the conditions of bright outdoor use, an ND filter that has a function of reducing the amount of transmitted light without changing the component of the incident light immediately before the color image sensor is inserted. By reducing the amount of light, it is possible to use the device in a higher brightness state without reducing the amount of light incident on the range image sensor.

On the other hand, the exposure control for the distance image is performed by controlling the number of lasers to be projected, controlling the current supplied to the laser, and controlling the insertion of an ND filter at an arbitrary optical position from the laser to the output lens. The output level is adjusted by adjusting the intensity or the amplifier gain supplied to the output signal. In this control, the laser intensity control value is determined from the distance information Daf to the measurement object obtained from the AF control unit and the focal length f of the lens under the measurement conditions. FIG. 12 shows an example of the control map.

Normally, the output of the distance image sensor is inversely proportional to the square of the distance information Daf to the object to be measured. Further, when the focal length f becomes shorter, the area required for illumination increases, so that the output signal of the range image sensor becomes smaller. Therefore, in the apparatus of this embodiment, the output level of the distance image calculation data is controlled by the number of lasers according to the focal length.
In the example shown in FIG. 3, three lasers are used up to a focal length f of 36.7 mm, and one laser is used above the focal length f. Further, the focal length f
And an image gain β (= Daf / f) calculated from distance information Daf to the measurement target determined by the output of the AF sensor and an amplifier gain given in the analog preprocessing circuit to the distance image sensor output. It is also controlled by changing it. In the illustrated example, when β = 35 to 50, 、, β
= 1 for 50 to 75, 2 for β = 75 to 100, β = 1
In the case of 00 to 200, the gain of the amplifier is set as 4, for example. In addition, in the case of using a laser light amount exceeding the safety standard defined in JIS in measurement in a telephoto system having a short focal length at a short distance, an ND filter is inserted at an arbitrary optical position from the laser to the output lens. Light amount control is also an effective means.

However, if a good measurement result cannot be obtained even when the measurement is performed with the control value obtained by the above control, a laser light intensity adjustment key is provided to adjust the laser intensity by operating the key, or the sensor accumulation time is adjusted. It is also possible to use a means such as changing the number. As another means, there is a method of performing laser pre-scanning based on the estimated laser intensity control value obtained from the distance information and the estimated reflectance of the measurement target obtained from the color image. The maximum output value of the distance image sensor at the time of pre-scanning is calculated, and this value is A / A
The laser intensity and the image sensor accumulation time which fall within the dynamic range of the D conversion and are sufficient for the distance information calculation in the subsequent stage are calculated, and the distance image is captured based on the calculated control values. . In addition, if there is an auxiliary lighting device for AF, the amount of light reflected by the irradiation of the auxiliary lighting is detected by the AF sensor at the center of the screen where the measurement target is considered to be present, and the detected reflection is detected. It is also possible to calculate the laser intensity and the image sensor accumulation time on the basis of the light amount and to capture the distance image.

The other is that parallax is generated by projecting and receiving light from different viewpoints (positions) (FIG. 13).
(See reference.) A means to solve this is needed. When light is projected and received by the same lens system having the same value of the image plane size and the focal length, the field of view coincides only at a specific distance (indicated by a large arrow OBJ1). If there is no object at the coincident distance, the three-dimensional shape measurement of the unprojected area is performed, and an unmeasurable area occurs. For example, when trying to measure an object at a position where the fields of view do not match as represented by a small arrow OBJ2 in FIG. 13, a deviation occurs between the light projecting range and the light receiving range. However, a problem occurs in that the light receiving system performs scanning.

The above problem can be solved by the following configuration. (1) The elevation angle of the optical axis of the light projecting system is steplessly changed according to the distance to the target (see FIG. 14). The focal length of the light emitting and receiving system is set to the same value, and the elevation range of the optical axis (indicated by a broken line) of the light emitting system is changed according to the distance to the target object by the autofocus measurement. It corresponds to. That is, since the influence of parallax increases as the distance becomes shorter, the elevation angle is increased and the scanning range is set to S1, and the elevation angle is decreased and the scanning range is set to S2 as the distance increases. The change of the optical axis of the light projecting system is changed by mechanical means.

(2) The elevation angle of the optical axis of the light projecting system is continuously changed in accordance with the distance to the object by an optical means for providing a variable refractive index prism immediately after the light projecting lens system projects light. However, the focal lengths f of the light projecting / receiving systems are set equal. By moving a prism having a curvature in and out according to the autofocus measurement distance, the refractive index changes, and the elevation angle of the optical axis of the light projecting system changes.

(3) Using an optical system having the same image plane size, the focal length fa of the light projecting system is controlled to be smaller than the focal length fp of the light receiving system. Alternatively, an optical system having a large image plane size is used for the light projecting system, and control is performed so that the focal lengths fa and fp of the light projecting / receiving system become the same value. With such means, the scanning range of the light projecting system is given a margin (about 1.5 times that of the light receiving system) with respect to the scanning range of the light receiving system, and the distance to the target object is divided into a plurality of zones. Then, the elevation angle of the optical axis of the light projecting system is changed stepwise according to each zone. For example, in the example shown in FIG. 15, the distance to the target is divided into two zones, the far side is defined as zone Z1, and the near side is defined as zone Z2. Then, in the case of the zone Z1, the elevation angle of the optical axis of the light projecting system is changed by a predetermined angle.

(4) As in (3) above, the scanning range of the light projecting system has a margin (about 1.5 times the light receiving system), the elevation angle of the optical axis of the light projecting system is fixed, and the focal length f varies. To set a limit on the closest measurement distance. In the example shown in FIG.
At a distance shorter than the distance at which J2 is located, there is an area where the light receiving range is outside the light projecting range.

In the cases of (1) and (2) (FIG. 14), it is assumed that the fields of view are completely coincident. You can start. On the other hand, in the cases of (3) and (4), as shown in FIGS. 15 and 16, the fields of view do not coincide with each other, and the laser scanning area of the light projecting system is large, so that a useless area is generated. Therefore, the time required to scan this useless area is calculated from the autofocus calculation reference distance.
Since the scan proceeds from top to bottom, there is a waste area at the beginning,
After the calculation time elapses from the start of scanning, the microcomputer starts to take in data from the range image sensor. In this case, since the scanning range is wide, (1) and (2)
In this case, the laser scanning time is required to be about 1.5 times as long as that of the case (1), and the input time of the three-dimensional shape is prolonged accordingly.

Further, since the light projecting system and the light receiving system have different elevation angles, the laser does not move at a strict uniform speed on the object plane perpendicular to the optical axis of the light receiving system. However, since the elevation angle itself is a small angle, it does not cause much problem, and by having a conversion table for converting the vertical position information scanned by the sensor and the pitch shift amount into the distance information. Three-dimensional measurement having almost isotropic properties is possible.

Next, the imaging system will be described in detail.
If the distance range to the object to be measured is limited with respect to the direction in which one slit is projected, the position of the slit on the sensor that receives the reflected light from the object is limited within a certain range. You. This is shown in FIG.

The longest distance measured is Df, and the shortest distance measured is Dn. Now, when the cutting plane by the slit light emitted from the light projecting system is the slit A, the range of the imaging device surface that receives the slit light reflected on the object surface is the distance between the intersection PAn of the measured latest distance Dn and the slit A. The point at which the three-dimensional position is projected on the image sensor is the lowermost point in the figure, and the three-dimensional position of the intersection PAf between the farthest distance Df and the slit A is the image sensor centering on the principal point of the image pickup lens. The point projected upward is limited to the closed section Ar on the image sensor having the uppermost point in the drawing. Similarly, in the case of the slit light B while keeping the positional relationship between the light projecting system and the light receiving system as it is, the projected point of the intersection PBn of the measured latest distance Dn and the slit B is defined as the lowest point in the figure, and the measured farthest distance The point where the intersection point PBf of Df and the slit B is projected is limited to a closed section Br on the image sensor having the uppermost point in the drawing.

As described above, in order to generate the distance data of 256 points in one column by projecting one slit light, the whole area of the image pickup device is not scanned but only the necessary range corresponding to the slit light is scanned. And the processing speed can be increased.

Further, in order to increase the speed of an apparatus for generating three-dimensional shape data, a function of outputting a strip image of only this area, for example, only an element image of 256 × 16 pixels, is required. A high-speed driving solid-state imaging device capable of selectively reading such a band-like region will be described below.
Various types of solid-state imaging devices can be used. The first is MO
This is a configuration in which a read start address setting function of an image sensor having an XY address scanning method such as S or CMD is added (FIG. 18). Second is a read transfer path (generally a horizontal register) in an analog transfer system such as a CCD image sensor.
This is a configuration in which a charge discharging function is added in parallel at the time of charge transfer to the device (FIGS. 19 and 20). The third configuration is that a block divided into strips is set in advance regardless of the scanning method, an output function is provided for each block, and the parallel output is used (FIG. 21).

FIG. 18 shows a configuration of an XY address scanning type sensor as a first configuration. Normally, each pixel is scanned by switches arranged in a matrix of a vertical scanning circuit 61 and a horizontal scanning circuit 62. The vertical scanning circuit 61 and the horizontal scanning circuit 62 are constituted by digital shift registers.
One row (256 pixels) can be scanned by inputting a 256 shift signal of a horizontal shift signal to one shift signal input of a vertical scan. In this embodiment, a strip-shaped random access read is realized by providing a scan start set register 63 for supplying a scan start set signal, which is a register initial value, to the vertical scanning circuit 61. By inputting the signals sgn1 and sgn2 representing the scan start position to the scan start set register 63, the position of the band image to be read is instructed.

When the number of pixels increases, the number of bits of the scan start set signal increases, and the number of input pins increases. Therefore, it is desirable to arrange the scan start set signal decoder 64. Then, the content of the scan start set register 63 is transferred in parallel to the vertical scanning circuit 61 at the start of reading, so that the scan start position (row) is set. Thereafter, signals from a desired row are obtained by performing horizontal scanning 256 times. Subsequently, one shift signal input in the vertical scanning and 256 shift signals in the horizontal direction are input, the signal of the next row is read, and by repeating this, an image of a desired band area is read. By performing the above operation, scanning of only the desired band-shaped area is realized, and the required scanning is completed in a time (the number of readout rows / the number of all-area rows) much shorter than the time required to scan the entire area.

The area that has been read once is reset and the next accumulation is started, but the accumulation of electric charge continues in the area that has not been read. At that time, there is no problem if the next reading is the same area, but when reading different areas, image information having different accumulation times will be mixed.
In the case of a three-dimensional measuring device using light cutting, it is necessary to read while shifting the band-like region that needs to be read together with scanning of the laser slit. The image of the area that is read redundantly is the image of the integration time from the previous reading to the current reading, but the integration of the image of the area that is newly read due to the shift of the reading area is an image that has been continued for a long time. Would. Therefore, in the present embodiment, this read-out band area is set as an area including both the area required this time and the area required next time. By doing so, the integration required for the area required for the next input is always cleared at the previous time, so that it is possible to avoid the problem of taking in an image composed of pixels having different integration times.

Next, FIG. 19 shows a configuration in the case of interline transfer of the CCD image pickup device as a second configuration, and FIG. 20 shows a configuration in the case of frame transfer. In the CCD image pickup device of this embodiment, an integration clear gate I for discharging charges to an overflow drain OD is provided in parallel with a transfer gate TG for transferring charges in parallel to the horizontal register 66.
By installing the CG, strip-shaped random access reading is realized.

In the case of the interline transfer, normally, when the image storage of the entire area is completed, the accumulated charges of all the pixels are transferred from the light receiving section to the transfer area in parallel. The scanning of the charge generated in each pixel is performed by a vertical register and a transfer gate T.
One shift signal is input to G, and the electric charges in the vertical registers are shifted downward one by one in parallel, and the electric charges in the lowermost vertical register are read out to the horizontal register 66. Thereafter, by supplying a 256 shift signal input of a horizontal shift signal, one row of charges can be scanned. This operation is repeated by the number of rows (340 rows) to read the entire area.

In this embodiment, the charges generated in unnecessary rows in the stage of scanning the charges generated in each pixel are supplied in parallel by supplying one shift signal input to the vertical register and the integration clear gate ICG. Overflow drain OD
Discharge to For a row that needs to be read, one shift signal is input to the vertical register and the transfer gate TG in the same manner as in the normal case, and each electric charge in the vertical register is shifted downward by one stage in parallel, while the lower vertical register Is read into the horizontal register 66, and thereafter, a row of charges is scanned by supplying a 256 shift signal input of a horizontal shift signal. In this manner, a random access function for each row is realized, and the necessary scanning is completed in a time (the number of readout rows / the number of all rows in the area) which is much shorter than the time for scanning the entire area of the image sensor. .

In the case of the frame transfer shown in FIG. 20, the structure is larger than that of the element in the case of the interline transfer. The upper side is a photoelectric conversion area and the lower side is a storage area. Generally, the accumulation area has the same number of pixels as the photoelectric conversion unit. In a normal operation, when image storage in all areas is completed, stored charges of all pixels are transferred in parallel from the photoelectric conversion area to the storage area by vertical transfer pulses for the number of rows. After the transfer, the scanning of the electric charge generated in each pixel is read out to the horizontal register 66 under the control of the vertical register and the transfer gate TG, as in the case of the interline transfer. Can be scanned.

In this embodiment, the charges accumulated in the pixels in the whole area are transferred in parallel from the photoelectric conversion area to the accumulation area by the number of vertical transfer pulses corresponding to the number of rows, and are generated in each pixel when the charges are transferred to the horizontal register 66. Unnecessary row charges are discharged in parallel to the overflow drain OD only by inputting one shift signal to the vertical register and the integration clear gate ICG at the stage of scanning of the charge thus performed. Further, only the number of rows required to be read each time (for example, 16 rows) is prepared, and the signal of the first pixel row that does not need to be read is synchronized with the vertical transfer pulse when vertically transferring from the photoelectric conversion area to the storage area. The charge may be discharged by opening the integration clear gate ICG, and reading from the horizontal register 66 may be performed when only the charges of the pixel rows that need to be read are transferred to the accumulation region. In this manner, a random access function is realized for each row, and the required scanning is completed in a time (the number of readout rows / the total number of rows in the entire area) much shorter than the time required to scan the entire area.

Next, FIG. 21 shows, as a third example, the configuration of a sensor that divides a plurality of blocks and outputs each block.
Here, the case of an XY address scanning type sensor will be described as an example, but it is easy to adopt the same configuration in an analog transfer system such as a CCD image sensor. In the present embodiment, a block configuration divided into a predetermined number of rows required for reading is adopted, and signals of each block are scanned in parallel and output. With respect to this parallel read output, an output is selected by operating the multiplexer 65 according to the area to be read, and is finally output. By performing such reading, random access is realized for each row although the output order is different, and the reading time is also reduced by the number of block divisions. FIG. 22 shows the relationship between the output form of a band-like image which is randomly read by this block-divided configuration and the block switching signal of the multiplexer.
Shown in Numerals 1 to 6 in the figure correspond to the line numbers in FIG.

FIG. 21 shows an example in which the data is divided into two very simple blocks (B1 and B2) and three arbitrary rows are read out. Will be described. The sensor outputs two lines internally, lines 1 to
3 and a block B2 output (FIG. 22B) for outputting lines 4 to 6, which are sent to the multiplexer as analog signals and selected and output by the selection signal Sel. Is done. By the operation of the multiplexer 65, the block B is output as the sensor output Out.
When one output is selected, the output of the block B1 becomes the sensor output as it is, and the output of the band-shaped images of lines 1, 2, and 3 is sequentially output (FIG. 22C). In addition, when block B2 output is selected as the sensor output, line 4,
5 and 6 are read out (FIG. 22F).

On the other hand, while the first and fourth lines are being output as the block outputs, the block B2 is selected and the line 4 is output, and then the multiplexer 65 is switched to the block B1 selection, whereby the sensor output is changed to the lines 4, 2, and 3. 3 are sequentially output, and the band-shaped images on lines 2, 3, and 4 are read out (FIG. 22D). For the sensor output, block B2 is selected up to the first two lines and line 4,
Then, when the block B1 is selected and the line 3 is output, the band-shaped images on the lines 3, 4, and 5 are read (FIG. 22E). In the figure, black triangles indicate that the output has been switched from block B2 to block B1. By switching the block selection signal during scanning in this way, it is possible to selectively read a strip image at an arbitrary position having the same size as the divided block, although the output order is different.

As described above, the three types of distance image sensors that can be randomly accessed in units of rows can be applied in this embodiment to shorten the input time to the three-dimensional shape measuring apparatus.

Next, the electronic circuit will be described. FIG. 23 is a block diagram showing the configuration of the entire electronic circuit. The main body of the measuring apparatus according to the present embodiment includes the lens driving circuits 71 and 7 for the light emitting and receiving system.
2. AF circuit 73, electric pan head drive circuit 76, input / output 7
A microcomputer CPU1 for controlling the image sensors 5, 74, etc .; image sensor drive circuits 13, 23;
7, 78, timer 79, SCSI controller 80,
Memory controller MC, pitch shift image processing circuit 83
And the like. Microcomputer C for controlling lens, input / output, etc.
The microcomputer receives signals from the control panel 75, such as a power supply operation and a key operation of a shooting mode, under the control of the PU1.
A control signal is transmitted to the PU 2, the light receiving system lens driving unit 71, the light projecting system lens driving unit 72, the AF driving unit 73, the display image generation unit 74, and the like, and controls zoom, focus, imaging, and the like.

For a color image, a color image sensor 24 is used.
And a block of the sensor drive circuit 23, the analog preprocessing circuit 81, and the image memory 84. For the distance image, there are a block of a distance image sensor 12, a sensor drive circuit 13, an analog pre-processing circuit 82, a pitch shift image processing circuit 83, and a pitch shift image memory 85.

When the power is turned on, the color image sensor 24,
The color image pickup system of the color image sensor drive circuit 23 and the color image analog pre-processing circuit 81 is driven, and the color image picked up to function as a monitor is supplied to the display image generation unit 74 and displayed on the display 41. These color image pickup system circuits are the same circuit systems as those known in a conventional video camera or the like. On the other hand, a sensor, a laser, and the like for capturing a distance image are initialized only when the power is turned on, and are not driven.
Since only 8 takes a long time until the mirror rotates at a constant speed, driving is started. In this state, the user sets the field of view by the power zoom operation while referring to the color image on the monitor screen 41, and performs release preparation for image input. Release operation is performed and release signal is generated.
The distance image sensor 12, the distance image sensor drive circuit 13, the distance image imaging system of the distance image analog pre-processing circuit 82, and the laser drive circuit 77 are driven by the transmission, and the pitch shift image memory 85 and the color image memory 84 respectively. Image information is captured.

The color image is supplied to the monitor device as an analog signal, but the color image frame memory 85 is subjected to A / D conversion by the A / D converter AD1 to input image information as digital information. I do. These processes are the same as those known in digital video, digital still video, and the like.

For the distance image, the microcomputer CPU 2 waits for a scanning start signal of the laser slit light transmitted from the scanning start sensor 33 shown in FIG. Thereafter, the focal length f, the base line length l, and the distance d to the measurement reference plane described above are obtained.
Wait for the useless time Td required for useless scanning due to the above. After measuring the above dead time Td from this scanning start signal,
Driving of the range image sensor 12 and its driving circuit 13 is started, and data acquisition is started. These timing operations are performed by a timer 79.

When the driving of the sensor is started, the slit laser having a length in the horizontal direction starts scanning from the top of the light receiving system scanning range downward. At the same time, the image integration of the range image sensor is started, and when the slit light is scanned by the polygon mirror 7 with the angular variation corresponding to one pixel of the range image sensor, the image integration unit changes to the accumulation unit. High-speed vertical transfer to is performed. Next, the range image sensor drive unit 13 is controlled so as to capture the image in the uppermost row at the center of the band-shaped region, and the image is read.
At the same time as the vertical transfer of the image from the integration unit to the storage unit is completed, the image is sequentially read from the storage unit as the output of the distance image sensor, and the integration of the charge according to the input light amount is performed by the integration unit for the next image read. Processing will be performed.

When the reading of one band-shaped image is completed in this way, the laser slit light is again scanned at an angular variation corresponding to one pixel of the distance image sensor, and high-speed vertical transfer from the image integration unit to the storage unit is performed. Done. The band-shaped area of the sensor for the current reading distance image is set at a position shifted by one pitch below the band-shaped area of the distance image sensor from the previous reading, and the image is read.

A series of this operation is continuously performed, and the input of the band-shaped image is sequentially repeated to obtain 324 images. During this time,
Since the polygon mirror rotates at a constant speed, a strip-shaped image for slit light having different light-cut surfaces is sequentially input. The output of this range image sensor is correlated 2
After processing such as double sampling, offset, and dark output is performed by the analog pre-processing circuit 82 for the distance image, it is converted to digital by the A / D converter AD2 and transmitted to the pitch-shift image processing circuit 83 as digital data.

The pitch shift image processing circuit 83 performs a center-of-gravity calculation for converting one band-shaped image data (256 × 16 pixels) into 256 points of the center of the received laser beam by a light-receiving center-of-gravity position calculating circuit (described later) shown in FIG. The calculated pitch shift amount is stored in the pitch shift image memory 85. By repeating this 324 times, 256 × 32
4 are obtained.

With the above processing, the respective images are stored in the pitch shift image memory 85 and the color image memory 84. These two images are transferred to the SCSI terminal 4 by the SCSI controller 80 via the memory controller MC under the control of the microcomputer CPU2 which controls the memory.
9. Output as digital data to the built-in MO device 22 or the like, or convert to analog by D / A converter DA1 and
It can be output to the LCD monitor 41 and the NTSC output terminals 50 and 51 as a TSC signal.

When outputting from the SCSI terminal 49,
In the output according to the SI standard, it takes several seconds to complete transmission of an output image of one set of color / pitch shift images. Therefore, a color image is recorded by a video device as a color NTSC signal usually used in a video device, a pitch-shifted image is treated as a luminance signal of an NTSC signal, and a monochrome image is output as a pitch-shifted image NTSC signal, whereby color / pitch shift in a moving image is performed. Image output becomes possible. With a high-speed image processing device, such real-time video images can be input to a computer,
As another means, the NTSC signal can be input to a computer by recording the NTSC signal by connecting it to a normal video device and then processing the grayscale image (pitch shift image) while advancing the frame during reproduction. . By using the color and pitch shift images of the moving object input to the computer in this way, it is possible to utilize the moving object in fields such as motion analysis.

It is also possible to perform control by providing, as an external device of the system, a rotary gantry controller 76 for controlling operations such as panning and tilting of the electric pan head 4 in which the present measuring apparatus is installed. This control operation will be described later.

FIG. 24 shows a detailed configuration of the light-receiving center-of-gravity position calculating circuit of the pitch shift image processing circuit 83. This circuit has a hardware configuration for calculating the center of gravity from information of five points out of 16 points of one band of image data. The signal from the range image sensor 12 is extracted only by the analog pre-processing circuit 82 for the effective pixels, A / D converted by the A / D converter AD1, and input to the circuit from the input terminal input on the left end of FIG. This input signal is transmitted to four registers 101a to 101a.
01d, 256 × 8-bit FIFO (FirstIn First
Out), 256 × 4 lines are stored and one line that is directly input, and a total of five lines are used for the calculation. The registers 103 and 104 have the same 256 ×
This is an 8-bit register. The register 109 is 256 × 5b
It is a FIFO register. These registers 10
The reason why two of 3, 104 and 109 are prepared for the same application is that the processing by the selection circuits 106 and 108, the comparison circuit 107, and the like requires several clock pulses, so the memory capacity is increased to two. Are alternately used for odd numbers (O) and even numbers (E), and which is used is controlled by clock pulses RCLK_O and RCLK_E.

The calculation of the center of gravity of the light-receiving laser is performed by the following equation based on the data of five points on five lines. First, since the amount of received light near the position of the center of gravity becomes maximum, the first row (I = 1 to 2)
The center of gravity of 56) is 、 (I, n) = D (I, n + 2) + D (I, n + 1) + D (I, n) + D (I, n-1) + D ( I, n-2) Find n = N (I) where (4) is maximum for each I. It is considered that the center of gravity is present in the vicinity of the N (I) th column, and then the complement of only the load average point Δ (I, N (I)) is obtained by the following equation. Δ (I, N (I)) = {2 * D (I, N (I) +2) + D (I, N (I) +1) -D (I, N (I) -1) -2 * D (I, N (I) -2)} / Σ (I, N (I)) (5) Finally, W (I) = N (I) + Δ (I, N (I)) (6) Set the center of gravity to be obtained.

Here, D (I, n) indicates data on the I-th row and the n-th column. One column has 256 pieces of data. D (I, n-1) is stored in the register 101a, and D (I, n-1) is stored in the register 101b.
n), D (I, n + 1) in the register 101c, D
The data of (I, n + 2) is stored and used for the calculation. Σ (I, n)
(Equation (4)) is calculated by the adder circuit 、 and stored in the register 104. Then, the result of the next operation is calculated as MAX (Σ (I,
n)) (comparison circuit 107), and if larger, the contents of this register 104 are updated and calculated at the same time.
{2 * D (I, n + 2) + D (I, n + 1) -D (I, n-1) -2 * D (I, n-2)} (= (5) Numerator = R1) in the register 103,
The column number n is updated and stored in the register 109. The data of D (I, n + 2) and D (I, n-2) is calculated by the shift circuit 10
The processing of (× 2) is performed by shifting one bit to the left by 2. After that, the addition circuit (+) and the subtraction circuit (-) perform the operation to calculate R1 at the point A, which is stored in the register 103.

For the column number n, the clock pulse PCLK is counted by the 5-bit binary counter 110. As a result of the comparison by the comparison circuit 107, when the maximum value is updated, the counter value at that time is stored in the register 109. Capture and store. In this embodiment, since the number n is 1 to 16, 5 bits are sufficient for the register 109 and the binary counter 110.

By repeating this operation for one band image, N (な る (I, n)) required to calculate the above equation is obtained.
(I), Σ (I, N (I)) and R1 are stored in register 10 respectively.
9, 104, and 103. Assuming that Σ (I, N (I)) = R2, the division circuit (÷) performs the operation of R1 / R2, and then the addition circuit (+) performs R1 / R2 + N (I) = Δ (I, N (I )) + N (I)
= W (I) is calculated, and the value of W (I) for 256 columns is finally output from the output terminal output at the right end.

The 256 W (I) s are stored in the pitch shift image memory 85, and this process is repeated for 324 strip images, so that the pitch shift information of 256 × 324 points is stored in the pitch shift image memory 85. A pitch shift image composed of W (I) is formed.

Next, the operation of the apparatus will be described in detail with reference to a flowchart. FIG. 25 is a flowchart of a main routine executed when the main switch is turned on. First, in step # 1, the CPU, the memory,
Initializes devices such as SCSI, MO, display, and control panel. Next, in step # 3,
The operation mode is determined. The operation modes include a camera mode for performing three-dimensional measurement and a replay mode for reading three-dimensional data from the memory device 22 such as an MO and displaying the data on a built-in display, and can be selected by operating a switch. Alternatively, one of the modes may be set by default. If the mode is the replay mode, the process proceeds to step # 5, and the process in the replay mode described later is performed. If the mode is the camera mode, the process proceeds to step # 6, and the processing in the camera mode described later is performed. When the camera mode ends, the polygon mirror is stopped in step # 7, and the AF / PZ is reset in step # 8 to return the lens to the initial position. Step #
At 9, the image sensor and the sensor drive circuit are stopped,
Returning to step # 3, the operation mode is determined.

Next, the operation in the camera mode will be described with reference to a flowchart shown in FIG. When the camera mode is selected, each device is initialized in step # 11, and a color image sensor is activated in step # 13 to supply a color image to the monitor display 41. The image is driven by driving an auto-focus sensor 31 arranged in the light-receiving zoom lens to receive light in an optimum focus state at all times to obtain an optimum color image. Next, in step # 15, the drive of the polygon mirror, which requires a long time to stabilize, is started first in order to prepare for the distance image pickup, and the process of the AF / PZ subroutine is performed in step # 17. Next, in step # 19, the process waits until the polygon mirror is stabilized. At the time when the polygon mirror is stabilized, the shutter mode is entered in step # 21, and the shutter mode subroutine is executed. In step # 23, the data transfer mode is entered, and the data transfer mode subroutine is executed. In step # 25, it is determined whether or not the camera mode has ended. If the camera mode has ended, the process proceeds to step # 27 and returns. If not, the process returns to step # 21.

AF / PZ Subroutine FIG. 26
(B) will be described. In step # 31, the lens position of the light emitting / receiving system is reset, the laser scanning range is reset in step # 33, and the process returns in step # 35.

Next, the operation in the shutter mode will be described with reference to a flowchart shown in FIG. In this state, the user changes the position, posture, and zooming of the measuring device to set the composition, while the device sets the shutter release button 4
7 is pressed and a release signal is output. First,
In step # 41, the AF / AE subroutine is executed to perform focusing and photometry. This subroutine will be described later. Next, in step # 43, select key 43 and A
Check the status of F / AE. Here, first, it is determined whether or not the select key has been pressed. If it is pressed ([Select]), the flow returns to step # 79 to exit the shutter mode. This corresponds to releasing the shutter release button down to the first stage in a general single-lens reflex camera. AF / A unless select key is pressed
The state of the E processing is checked, and if the AF / AE processing is in operation, the process returns to step # 41 and the AF / AE processing is repeatedly performed. If the AF / AE processing has been completed, the process proceeds to the next step # 45. That is, during the above-described processing period, the light receiving, the focusing of the projection system zoom lens, and the photometry are repeatedly performed, and control is performed so that the in-focus state is always maintained.

When the AF / AE is completed, the shutter button 47 is unlocked and the driving of focusing and zooming is inhibited (AF / PZ lock) in step # 45 in preparation for photographing. Then, in a step # 47, it is determined whether or not the shutter button 47 is pressed. If the shutter button has been pressed, the process proceeds to step # 55.
If it has not been pressed, the flow proceeds to step # 51 to determine whether a predetermined time has elapsed, and if not, the flow returns to step # 47 to determine whether or not the shutter button has been pressed.
If the predetermined time has elapsed, the operation of the shutter button is locked in step # 53, and the process returns to step # 41.

In step # 55, the laser beam is emitted, and in step # 57, the time until the laser beam starts to oscillate normally and the time until the preparation of the polygon mirror is completed are waited. When the preparation is completed, sensor driving is started in step # 59. First, in step # 61, the process waits until receiving an output from the scanning start sensor attached to the side of the condenser lens.
At 63, the drive of the distance image sensor is started after waiting for the dead time Td. The dead time Td is calculated from the focal length f, the base length l, and the distance d to the measurement reference plane. In step # 65, the input band image position of the distance image sensor is set to the initial position, and the operation of capturing the pitch shift image and the color image is started. At the same time, the center of gravity of the input light is calculated.
In step # 67, it is determined whether the scanning is completed. If not completed, the flow returns to step # 65 to repeat image capture. While shifting the input strip image position from the initial position by one pitch in accordance with the scanning of the laser slit light, 324 strip images are captured.

When the sensor driving is started, the color image sensor is driven again in step # 69 following the start of driving of the distance image sensor, and the color image read in step # 71 is stored in the color image memory. take in. The driving of the two image sensors and the taking into the memory are automatically and simultaneously processed by the hardware configuration. Thereafter, the process proceeds to step # 73.

When the capture of the pitch shift image and the color image is completed, the emission of the laser beam is stopped in step # 73, the prohibition of the zoom drive and the focus drive is released in step # 75, and the captured image is selected in step # 77. Is displayed according to the set mode, and the process returns in step # 79.

Next, the flowchart of the AF / AE subroutine in step # 41 will be described with reference to FIG. 28. First, in step # 91, the lens drive amount is calculated based on the information of the AF sensor 31, and the focus lens is calculated based on the calculation result. Is driven (step # 93). Step #
At step 95, a scanning start laser position is set, and step #
At 97, the laser power is controlled. In step # 99, photometry (AE) is performed, and the process returns in step # 101.

Next, the operation of the data transfer mode will be described with reference to a flowchart shown in FIG. First, the display mode is determined in step # 111. A flag check is performed to determine whether the image to be displayed is a pitch-shifted image or a color image displayed in shades, whichever is in a selected state.
In this display mode, for example, a color image display is set by default, and can be selected by key operation. When there is no key operation, or when a color image is selected by key operation, a color image is displayed in step # 113. If the pitch shift image display is selected, a pitch shift image is displayed in step # 115.
After displaying the image, it is determined in step # 117 whether the display mode has been changed. If it has been changed, the flow returns to step # 111 to display an image according to the selected mode. If the display mode has not been changed, step #
Go to 119.

In step # 119, it is determined whether data transfer is necessary. If the data transfer is not necessary, the process proceeds to step # 133, where a color image is displayed. If data transfer is necessary, a data header is created in step # 121. In step # 123, it is determined whether the mode is the SCSI output mode. If the SCSI output mode has been selected, external output data is created in step # 125, and data transfer is performed in step # 131. SCSI
If it is not the output mode, it will be recorded with the built-in recording device,
In step # 127, data for the internal MO drive is created, and in step # 129, the data transfer instruction to the MO is C
Sent from PU2 to the SCSI controller, step #
At 131, data transfer is performed. Then, step # 133
To display a color image, and the flow returns to step # 135. These data transfer destinations can be selected by key operation.

Next, the replay mode will be described.
In step # 3, the switch for switching to the replay mode is checked. If the switch has not been made, the apparatus enters a standby state (camera mode) for performing the next image input. If the switch has been made, the apparatus shifts to the replay mode. .

This replay mode is different from the above-described camera mode, in that the image data already recorded in the built-in recording device such as the MO is replayed to check again, or the SCSI data is renewed.
This is a mode for outputting to an external device via a terminal. The operation of this replay mode will be described with reference to a flowchart of FIG.

First, in step # 151, a list of images recorded in the MO is displayed. In step # 153,
The user performs reconfirmation display or selection of image data to be transferred to the outside from the list display screen. Next step # 1
At 55, the selected image data is loaded from the built-in MO into the color image / pitch shift image memories 84 and 85, respectively, for the color / pitch shift image, and step # 157.
Then, a flag check is performed to determine whether the image to be displayed is a color image or a pitch shift image. Step # 1 according to the selected display mode
At 59, a color image is displayed, or at step # 161, a pitch shift image is displayed. After displaying the image, step #
At 163, it is determined whether or not the display mode has been changed. If the display mode has been changed, the process returns to step # 157 and the display is performed again.

If there is no change, it is determined at step # 165 whether or not the next image data is to be displayed. If the next image display is not performed, it is determined in step # 167 whether or not the image data read from the MO to the memory is to be transferred to the outside. If not, the process jumps to step # 173. In the case of transfer, data for external output is created in step # 169, and data transfer is performed in step # 171. Then, in step # 173, it is determined whether to display the next image data, and if it is to be displayed, the process returns to step # 151.
If not, the process returns at step # 175.

Next, FIG. 31 shows a state transition diagram by each key operation of these series of operations. In this figure, up, down, left and right triangles indicating operations indicate operations of the cursor keys 42 in FIG. 3, and [Shutter], [Select], and [Cancel] indicate a shutter button 47, a select key 43, and a cancel key, respectively. 44 is shown. Although the present embodiment has a clock function, the time can be automatically assigned to the file name of the image file to be recorded.

On the menu screen, selection and execution can be performed by operating the left / right cursor keys and the select key 43 from the selectable state of the reproduction display, the list display, and the clock function. . The clock function allows time setting, and the list display function allows file operations such as file name change, deletion, and selection of display files. In the reproduction display function, a color image display is set by default, and the display can be switched between a pitch shift image display and a character screen display with the left and right cursor keys. In each display mode, the previous image and the next image can be displayed by operating the up and down cursor keys. In the character screen display, file operations such as file name change and deletion can be performed with the select key.

When the user operates the cancel key on the menu screen, the apparatus enters a shooting standby state (camera mode), and the user can return to the menu by pressing the select key. When the shutter button is pressed in the state of waiting for photographing, photographing is possible, the image is taken into the memory, and after photographing, the state can be returned to the state of waiting for photographing with the cancel key. When the select key is operated from the state immediately after shooting, the image can be recorded and the image captured in the memory is transferred to the storage device. At this time, the color image is set by default, and the pitch shift image /
A color image can be selected.

Next, a description will be given of a high-accuracy input by the three-dimensional shape measuring apparatus by divisional capture. When the distance between the light projecting system and the light receiving system, that is, the base line length l and the focal length f, and the distance d to the measurement object are determined, the three-dimensional resolution and accuracy are determined. Therefore, high-precision measurement is achieved by setting the focal length f large and measuring. In other words, the measurement accuracy increases as the distance increases. However, although it is possible to obtain a three-dimensional image with high measurement accuracy, the visual field is narrowed as the focal length f increases.

Therefore, the resolution at which the focal length f is to be measured,
The value is set to a value corresponding to the accuracy, and the rotary gantry 4 such as an electric camera platform is operated to divide the field of view into a plurality of regions, and each of the divided regions is measured. This is to reconstruct a single image. With such a function, it is possible to realize a three-dimensional shape measuring apparatus with variable resolution. By making use of this function, the environment can be measured by performing three-dimensional measurement on the entire surrounding space. A specific example will be described below, and the operation will be described.
The example shown in FIG. 32 is a simplified explanatory diagram, and the light projecting system 2 and the light receiving system 3 are arranged in a horizontal positional relationship, which is different from the example shown in FIG. In this arrangement, the slit light has a length in the vertical direction and needs to scan in the left-right direction.

FIG. 32 shows a state when the image pasting function is used.
FIG. 33 is a flowchart showing the operation of the image pasting function. FIG. 34 shows a display state when this function is used, and there is a display section indicating measurement accuracy below the image display section.

First, as shown in FIG. 32 (a), a wide angle which allows the user to take an image of the target object 1 within the visual field range by an operation by the user.
The zoom drive system 16 is driven so as to be in the wide state (focal length f0) and the visual field range is set (step # 20).
1). At this time, the resolution in the Z-axis direction (refer to FIG. 17, the uneven direction of the object) is represented by a bar display below the image as shown in FIG. This Z-axis direction resolution ΔZ
In the case where the base line length is fixed as in this system, the distance d to the object to be measured and the focal length f at the time of measurement simply have the following relationship. ΔZ = K × d × (df) / f (7) where K is a coefficient for estimating the resolution in the Z-axis direction,
It is determined by the sensor pitch and the like. In addition, the above zooming operation is performed by the system computer from the SCSI.
A command can be transmitted via a terminal, and operation settings such as a zoom operation and a release operation by remote control can be performed.

If the user determines that the measurement is performed with the accuracy and resolution that satisfy the above setting operation (NO in step # 203), the measurement is started by the user's release operation (step # 205). ), And the result is displayed on the display (step # 207). In this display, as shown in FIG. 34A, the inputted pitch shift image or color image and the Z-axis direction measurement resolution obtained by taking in the image are displayed as bars below the image.
As a result, when measurement with higher measurement accuracy is not required (NO in the determination in step # 209), the user is asked to determine whether or not to complete the measurement and to write the data to the storage medium. Processing is performed to complete the operation.

When the user determines that the measurement is not performed with satisfactory accuracy (YE in the determination in step # 203).
S) Or, by displaying the pitch shift image captured by the first release operation or displaying the Z-axis direction measurement resolution, the user can change the accuracy by key operation to set the desired Z-axis direction resolution and accuracy, and instruct re-measurement. Can be performed (YES in step # 209).

When the precision setting key is input, the system determines the current state, that is, the focal length f0 in a state where the entire view of the measurement target is captured, and the approximate distance d to the measurement target obtained from the AF sensor. The memory is stored to store the visual field range (step # 210). Further, the focal length f1 to be set is calculated from the input desired measurement resolution in the Z-axis direction and the approximate distance d using the above equation (7) (step # 21).
1).

When the focal length f1 is calculated, zooming is automatically performed on the focal length f1 (step # 21).
3) The number of frames to be divided and input based on the stored visual field range to be measured, the approximate distance d, and the focal length f1, the calculation of the pan and tilt angles corresponding thereto, and the pan and tilt of the pan / tilt rotation base to determine the visual field position. Make settings (Step # 2
15), measurement is performed in each divided input frame (step # 217). The image to be divided and input at the time of the image stitching function is to be stitched later and reconstructed into one image.
It is set to include the overlap that should be the margin.

Information indicating the obtained pitch shift image, color image, and captured view directions in the X and Y directions (for example,
The pan / tilt decode angle values or the order of capture in the X and Y directions, the lens focal length, and the measured distance information are stored in an internal MO storage device (step # 21).
9). At this time, the directory information such as the file name and the file size is not written into the memory, and the directory information can be written after the user's confirmation at the end to temporarily store the information.

Next, in the field of view slightly overlapped with the above field of view, the field of view is controlled by panning and tilting operations according to the calculated pan and tilt angles, and an image of an adjacent area is input. The whole area is input by repeating (NO in step # 221, see FIG. 32B).

When all areas have been input (step #
If the determination in step 221 is YES), the camera posture and the focal length are returned to the initial values before increasing the measurement accuracy (step # 22).
3) Complete the operation, wait for the user to judge the writing, write the directory information in the case of the write instruction, and terminate without writing the directory information in the case of the instruction not to write. The information stored in the memory up to the next is erased.

When the measurement is performed in advance and the measurement is performed again as in the above operation, the distance measurement to the target object and the measurement of the distance distribution within the measurement angle of view are completed by the first measurement. are doing. Therefore, re-measurement is not performed for a region having a large difference in the distance from the target object, that is, a divided input frame including only a peripheral region (background) different from the measurement target because of the bonding. It is also possible to perform re-measurement of only the frame. FIG.
In the example shown in (b), the halftone dot region including the target object is a region where the re-measurement is performed, and the other region is a region where the target object is not included and the re-measurement is not performed.

As described above, high-speed three-dimensional measurement can be performed, and three-dimensional shape measurement in which the resolution can be freely set by repeating partial input based on the three-dimensional measurement and performing bonding work. Becomes possible.

In such a bonding measurement, the measurement is performed with a uniform resolution over the entire screen, but the shape, color information is complicated for the eyes, mouth, and nose like a human face. Although high-resolution data is required, a measurement target such as a cheek or a forehead that needs sufficient measurement at a low resolution may be considered. For such a measurement target, efficient data input can be realized by performing data bonding by a partial zooming operation. This partial zooming bonding function is realized by the following operation.

FIG. 35 is a flowchart showing the operation of the partial zooming and bonding function. First, similarly to the case of the uniform resolution bonding in step # 251, the visual field setting for capturing the entire area of the measurement target is performed, and step # 25
In step 3, a partial zoom input mode is selected. When the selection is made, the currently set focal length f0 and pan / tilt decode angle values are stored in memory (step # 25).
5) The measurement is started in the state of the focal length f0, and image input is performed as general image data (step # 257). The resulting pitch shift image, color image, and information indicating the view direction of the captured X and Y directions of the image (for example,
Pan and tilt decode angle values), lens focal length, and measured distance information are stored in an internal storage device (step #).
259). Subsequently, at step # 261, the maximum focal length fma
After zooming to x, the above-described schematic image data is analyzed, and it is determined whether or not to perform re-measurement for each divided input frame input after zooming.

When the zooming is performed and the measurement is performed at the maximum focal length fmax, the rough data is divided into frame sizes that can be input. In step # 263, the pan / tilt positions X and Y are set to the initial start positions Xs and Ys. The pan / tilt is controlled to the X and Y positions set in step # 265. Next, in step # 267, X ± ΔX,
The color information R of the initial input color image in the area of Y ± ΔY,
Statistical processing is performed on the G and B values to calculate standard deviations σR, σG, and σB for each area. In step # 269, it is determined whether all the calculated standard deviations σR, σG, and σB are less than the set predetermined values. Indicates that the area is a uniform area, and the process proceeds to step # 271 without performing the zooming measurement. Conversely, standard deviation σ that is equal to or greater than a predetermined value
If there is R, σG, σB, it is determined that the area has complex color information, and zooming measurement is performed (step #).
275).

At step # 271, X ± ΔX, Y ± ΔY
The standard deviation σd is calculated from the information of the initial input distance value d in the area of, and in step # 273, it is determined whether the calculated value of the standard deviation σd is smaller than a set predetermined value. Since the small area is a flat area with little change in shape, zooming measurement is not performed, and step # 279 is performed.
Proceed to. Conversely, if the value is equal to or more than the predetermined value, it is determined that the region has a complicated shape (distance information) and zooming measurement is performed (step # 275).

After the zooming measurement in step # 275, the obtained pitch-shifted image, color image, information indicating the viewing direction in the X and Y directions (for example, pan and tilt decode angle values) from which the image was captured, Information such as lens focal length and measured distance information is stored in a storage device such as an internal MO (step # 277). Thereafter, the process proceeds to step # 279.

Next, in step # 279, the pan / tilt position X is changed by 2ΔX. X in step # 281
It is determined whether or not the directional scanning has been completed, and if not, the process returns to step # 265. Step # if completed
At 283, the pan / tilt position Y is changed by 2ΔY.
At step # 285, it is determined whether or not all the scans have been completed.
If it has not been completed, the process returns to step # 265. If completed, the process proceeds to step # 287, and this routine ends.

As described above, the rough image data and the partial detailed image information whose position can be determined can be input, and the partial detailed image data at the position can be attached to the rough image data to obtain the shape and color information. Efficient three-dimensional input according to complexity can be realized.

[0125]

As described above, when reading out an image picked up by the two-dimensional image pickup device, the whole area is not read out, but the strip image of the area where the projected slit light is assumed to be incident is read out. Therefore, the image can be read in a very short time as compared with the case of reading the entire area. As a result, the shape of the target object can be measured at high speed even if the target is a moving object, and conversely, the shape of the target object can be measured at high speed even when the measurement device itself is installed on a moving object. Can be.

[Brief description of the drawings]

FIG. 1 is an explanatory view showing the principle of a light cutting method.

FIG. 2 is a schematic block diagram of the entire apparatus according to the present invention.

FIG. 3 is a perspective view showing a schematic configuration of the entire apparatus according to the present invention.

FIG. 4 is an explanatory diagram of a light amount distribution generated on a target object plane.

FIG. 5 is an explanatory diagram of a light amount distribution generated on a light receiving surface of an image sensor.

FIG. 6 is an explanatory diagram of a light amount distribution generated on a light receiving surface of an image sensor.

FIG. 7 is a sectional view showing a configuration of a light projecting optical system.

FIG. 8 is an explanatory view of a projection slit light.

FIG. 9 is a sectional view showing a configuration of a light receiving optical system.

FIG. 10 is an explanatory diagram showing characteristics of an incident wavelength of the color image sensor.

FIG. 11 is an explanatory diagram showing a light receiving wavelength of a range image sensor.

FIG. 12 is an explanatory diagram illustrating an example of output control of a distance sensor.

FIG. 13 is an explanatory diagram of parallax between a light projecting system and a light receiving system.

FIG. 14 is an explanatory diagram of stepless elevation angle control.

FIG. 15 is an explanatory diagram of multi-stage elevation angle control.

FIG. 16 is an explanatory diagram of the elevation angle fixed closest distance control.

FIG. 17 is an explanatory diagram of an incident range and a scanning range of reflected light incident on an image sensor.

FIG. 18 is an explanatory diagram of an XY address scanning type sensor.

FIG. 19 is an explanatory diagram of an analog transfer type sensor (at the time of interline transfer).

FIG. 20 is an explanatory diagram of an analog transfer type sensor (at the time of a frame transfer method).

FIG. 21 is an explanatory diagram of a block-divided sensor.

FIG. 22 is an explanatory diagram of row random access of the block division sensor.

FIG. 23 is a block diagram of a circuit configuration of the entire apparatus.

FIG. 24 is a circuit diagram for calculating a light receiving center of gravity position.

FIG. 25 is a flowchart showing the operation of a main routine.

FIG. 26 is a flowchart showing an operation in a camera mode.

FIG. 27 is a flowchart showing an operation in a shutter mode.

FIG. 28 is a flowchart of an AF / AE subroutine.

FIG. 29 is a flowchart showing an operation in a data transfer mode;

FIG. 30 is a flowchart showing an operation in a replay mode.

FIG. 31 is an operation state transition diagram of the measurement device.

FIG. 32 is an explanatory diagram of an image pasting function.

FIG. 33 is a flowchart showing the operation of the image pasting function;

FIG. 34 is an explanatory diagram of a display state of the image pasting function.

FIG. 35 is a flowchart showing the operation of the partial zooming and bonding function;

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 target object 2 light projecting optical system 3 light receiving optical system 12 two-dimensional image sensor

──────────────────────────────────────────────────続 き Continued on the front page (58) Fields investigated (Int. Cl. ⁷ , DB name) G01B 11/00-11/30 H04N 5/335 G06T 1/00

Claims (4)

(57) [Claims] 1. A three-dimensional shape measuring apparatus for projecting slit light onto a target object, capturing an image with a two-dimensional image sensor, and measuring the shape of the target object. A three-dimensional shape measuring apparatus characterized by selectively reading a band image of a region. 2. The method according to claim 1, wherein the two-dimensional image pickup device is divided into a plurality of blocks, the blocks are read in parallel for each block, and the signals read in parallel are selected and output. 3D shape measuring device. 3. The three-dimensional shape measuring apparatus according to claim 1, wherein said two-dimensional image pickup device has a one-row parallel discharging means and selects and outputs a band-shaped image. 4. The apparatus according to claim 1, wherein said two-dimensional image pickup device is an XY address scanning type two-dimensional image pickup device, has a read start line setting means, and selects and outputs a band image. 3D shape measuring device.