JPH11230725A - Three dimensional measuring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To store data that is associated a location in measuring view with a location in memory space as a virtual screen and simplify controlling memory that stores data. SOLUTION: A three dimensional measuring device 1 projects a light beam L so as to raster-scan a virtual surface VS, and outputs a specified data DD corresponding to an incident angle of the light beam L reflected at the time when it passes a measuring object, namely, each sampling partition sp which is obtained by subdividing the virtual surface VS into a main-scan direction X and a sub-scan direction Y. The device is provided with a memory 60 for storing the specified data DD is provided, and a memory control means to access the memory 60 with a position data Xg in the main-scan direction and Yg in the sub-scan direction as an address for each sampling partition sp to write the specified data of each sampling partition sp.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a three-dimensional measuring device for projecting a light beam onto an object to measure the shape of the object in a non-contact manner.

[0002]

2. Description of the Related Art A non-contact type three-dimensional measuring device (three-dimensional camera) called a range finder can perform higher-speed measurement than a contact-type three-dimensional measuring device.
It is used for data input to the D system, body measurement, visual recognition of robots, and the like.

An optical projection method is known as a measurement method suitable for a range finder. This method is a method of optically scanning an object to obtain a distance image (three-dimensional image) based on the principle of triangulation. Active measurement in which a beam of reference light is projected to perform raster scanning of the object. Is a kind of method. The raster scanning includes, for example, a mode in which main scanning in one direction is performed from left to right, and a mode in which scanning from left to right and scanning in the opposite direction are alternately performed (reciprocating main scanning). The distance image is a set of pixels indicating three-dimensional positions of a plurality of parts on the object. The calculation for obtaining the distance image from the shooting information is performed by a range finder or an external information processing device (such as a computer system).

Generally, measurement information obtained by a range finder is input to an information processing apparatus online or offline using a storage medium, and undergoes predetermined processing such as analysis, processing, storage, and display.

[0005]

As the above-mentioned range finder, for example, a range image is displayed by connecting to a display device, and a change in the position and shape of the moving object is monitored by periodically repeating the measurement. Display 3
It can be used as a three-dimensional video camera. In this case, it is necessary to temporarily store the distance image or the measurement data for one frame on which the distance image is generated in a memory, and to read the data every display frame period. As the memory control at that time, a mode in which data is written in a chronological order in a fixed sampling cycle in parallel with the raster scanning can be adopted. The address pointer is incremented at the sampling period.

However, in such a memory control, rearrangement (rewriting) for adapting to line-sequential display is performed.
And control of the read address. In particular, when the measurement is speeded up by the reciprocating main scanning, the address specification at the time of reading becomes complicated.

SUMMARY OF THE INVENTION It is an object of the present invention to store data by associating a position in a visual field of measurement with a position in a memory space as a virtual screen, and to simplify control of a memory for storing data.

[0008]

According to a first aspect of the present invention, a light beam is projected so as to perform raster scanning toward a virtual plane, and the virtual plane is subdivided in a main scanning direction and a sub-scanning direction. What is claimed is: 1. A three-dimensional measuring device that outputs specific data according to an incident angle of a light beam reflected by a measurement target at a time when passing through each sampling section, wherein a memory storing the specific data, A memory control unit that accesses the memory using the position data in the main scanning direction and the position data in the sub-scanning direction as addresses, and writes the specific data in each of the sampling sections.

According to a second aspect of the present invention, a light beam is projected so as to perform raster scanning toward a virtual plane, and passes through each sampling section obtained by subdividing the virtual plane in a main scanning direction and a sub-scanning direction. Output specific data according to the incident angle of the light beam reflected by the measurement target at the time 3
A dimension measurement device, a memory for storing position data of each sampling section in the sub-scanning direction, and accessing the memory using the position data of the sampling section in the main scanning direction and the specific data as addresses; Memory control means for writing the position data of the sampling section in the sub-scanning direction.

In the invention according to claim 3, the specific data is position data of a spot of the light beam incident on a light receiving surface. The invention according to claim 4, wherein the specific data is distance data calculated based on a position of a spot of the light beam incident on a light receiving surface.

[0011]

FIG. 1 is a diagram showing an outline of a three-dimensional measuring apparatus 1 according to the present invention. The three-dimensional measuring apparatus 1 performs light scanning so as to perform raster scanning toward the virtual plane VS.
, A light receiving system 20 that receives the light beam L reflected by the object Q to be measured, and a frame memory 60 that stores specific data DD corresponding to the measured value.

The light projecting system 10 includes a semiconductor laser (LD) 11 as a light source and a galvano mirror 12 as a main scanning unit.
X and a galvanomirror 12Y serving as a sub-scanning unit. Each of the galvanometer mirrors 12X and 12Y includes a mirror that reflects the light beam L and an electromagnetic mechanism that rotates the mirror. The electromagnetic mechanism corrects the count value of the clock SPCLK in a look-up table format, and
An A-converted drive voltage is supplied. The lookup table stores, for example, conversion data for changing the rotation speed of the mirror so that the scanning speed on the virtual plane VS is constant. The main scanning is of a reciprocating type in which the direction of beam deflection is reversed every line. The sub-scan is performed intermittently every main scan of one line. In the main scanning, since the beam deflection speed is higher than that in the sub-scanning, a deviation between the control target value indicated by the driving voltage and the actual rotation angle position is likely to occur.
Therefore, in particular, the galvanomirror 12X is provided with a rotation angle sensor for accurately grasping the position of the light spot on the virtual plane VS. Note that the following description may be made on the assumption that the main scanning direction (X direction) is a horizontal direction and the sub scanning direction (Y direction) is a vertical direction.

The light receiving system 20 includes an imaging lens 21, a prism 22 for separating visible light and the light beam L, a CCD image pickup device 23 for outputting a color photographed image for monitoring,
And a light receiving device 25 for detecting the incident angle of the light beam L. The light receiving device 25 is a position detection type detector (PSD) that outputs an analog signal corresponding to a spot position of light incident on the light receiving surface. By using the PSD, scanning can be speeded up as much as charge accumulation is not required, as compared with the case of using a CCD imaging device. The light receiving system 20 and the above-mentioned light projecting system 10 are arranged at a certain distance in the Y direction, and the arrangement relationship between them is known. Therefore, if the incident angle of the light beam L incident on the prism 22 in the Y direction is known, the light beam L
Can be obtained by applying a well-known triangulation method. Light beam L
Corresponds to the distance between the center of the light receiving surface of the light receiving device 25 and the light receiving spot. If the output of the light receiving device 25 is periodically sampled during the scanning period, the object Q is set for each sampling section (in principle, a point) sp obtained by subdividing the virtual plane VS in the X direction and the Y direction.
(A position in a direction orthogonal to the virtual plane VS) can be measured. That is, a distance image having the sampling section sp as a pixel can be obtained.

In this embodiment, the light receiving device 25
Is written into the frame memory 60 as the specific data DD. An important matter here is to use the position data Xg, Yg in the X direction and the Y direction of each sampling section sp for addressing the frame memory 60. Thereby, the detection data Yp
Are simply written in the order of occurrence, the pixel array on the virtual screen, which is the address space of the frame memory 60, matches the pixel array on the virtual plane VS. Therefore, no inconvenience arises even if data is read from the frame memory 60 by specifying an address so as to perform raster scanning in the one-way main scanning format. In simple writing, the pixel arrangement direction is switched for each line. Therefore, it is necessary to rearrange the pixels before reading or to specify a complicated address at the time of reading.

The detection data Yp written in the frame memory 60 is read out for displaying a distance image, and a look-up table (LUT) 71 and a D / A converter 72 are provided.
Is output to a display (not shown) as an NTSC video signal. The LUT 71 stores conversion data corresponding to performing a triangulation calculation for obtaining a distance image and performing correction based on calibration on the result. The calibration measures, for example, a plane. Reading from the frame memory 60 is performed for each frame period of video image display. Detection data Y
The distance image based on p is three-dimensional information of the object Q viewed from the light projecting system 10.

FIG. 2 is a block diagram of a main part of the control system.
The three-dimensional measuring device 1 is a CP having a microprocessor.
A controller 52 for performing scanning control and data input / output control is provided together with U51. The controller 52 is a semiconductor device (for example, a gate array) in which a plurality of circuit modules are integrated. The LUT 33 and the D / A converter 34 are involved in the control of the galvanomirror 12X by the controller 52, and the LU is used in the control of the galvanomirror 12Y.
The T31 and the D / A converter 32 are involved. The rotation angle sensor signal (0 to 5 volts) of the galvanomirror 12X is A /
After being converted into 12-bit data by the D converter 35,
The controller 52 receives the position data Xg via the UT 36.
Is input to

The controller 52 receives the detection data Yp from the LUT 39. The input of the LUT 39 includes two types of detection signals Sigma,
ΔY is quantized by D / A converters 37 and 38, respectively. The value of the detection signal Sigma, ΔY is expressed by the following equation.

Sigma = X1 + X2 + Y1 + Y2 ΔY = (X2 + Y2) − (X1 + Y1) X1: Output signal of first electrode in X direction (photocurrent) X2: Output signal of second electrode in X direction Y1: Output signal of first electrode in Y direction Output signal Y2: Output signal of second electrode in X direction FIG. 3 is a diagram showing an output image size.

The number of horizontal pixels of the image is "128". In consideration of the time required for reversing the driving direction of the galvanomirror 12X for main scanning, a margin for 16 pixels is provided at each end of the image. Further, sampling is performed twice per pixel in order to prevent image omission described below. Therefore, one line scanning time H is equal to the sampling clock SP.
320 cycles of CLK [320 = (128 + 16 × 2) ×
2]. In the main scanning control, the driving signal is generated by counting the sampling clock SPCLK. Since it is a reciprocating type, the sampling clock SPC
The counter is reset every 640 cycles of LK.

The number of vertical pixels is "96". The retrace period (mirror return period) in the Y direction is 4H. Therefore,
The scanning time V for one screen is 100H. In the sub-scanning, the driving signal is generated by counting the sampling clock SPCLK as in the main scanning.

FIG. 4 is a diagram for explaining the sub-scan timing adjustment. Galvano mirror 1 as shown in FIG.
There is a phase difference of about 30 clock cycles between the drive signal for 2X (solid line) and the detection signal (broken line) indicating the actual rotation angle. That is, the mirror operation is delayed with respect to the drive request. For this reason, when the sub-scan is performed at the time when the count of the main scan for one line is completed (count value = 319), the scan spot reaches before the end of each line as shown in FIG. Since a locus that moves to the next line is drawn, proper scanning cannot be performed. Therefore, the controller 5
2 is provided with a dedicated register (YCUE register), and when the counter value of the main scan matches the value of this register, activates the counter of the sub-scan so that the timing of the sub-scan can be adjusted. ing.
With this configuration, the scanning state can be easily optimized according to the scanning range and the scanning speed changed by zooming or the like.

FIG. 5 is a diagram showing a method for preventing pixel omission. As described above, the position data X, which is the monitor information of the galvanometer mirror 12X, is used to specify the address of the frame memory 60.
g, a fixed number (12 in this example)
If the same number of samplings (one per pixel) are performed to obtain the data of 8), pixel omission in which data is not written to an address at a certain pixel position is likely to occur. The cause may be uneven rotation of the galvanomirror 12X, noise, or the like. FIG. 5 shows an example in which scanning for two or more pixels is performed during a sampling period due to rotation unevenness. That is, during the period from the time t2 to the time t3, the main scanning advances from the pixel position 68 to the pixel position 70, and the amount of the advance is twice the normal amount. In this situation, one sampling per pixel causes pixel omission at the pixel position 69. In the present embodiment, since the number of times of sampling per pixel is 2, sampling is performed even at the time point t2 'of the pixel position 69, and pixel omission is prevented. If the pixel position is the same in a plurality of samplings, the data is overwritten on the same address, so that the data written last becomes effective as measurement information. Three or more samplings may be performed per pixel to further reduce pixel omission.

FIG. 6 is a block diagram showing a functional configuration of the controller 52. The controller 52 has a write control unit 510, a memory control unit 520, and a display control unit 530. The write control unit 510 includes the X counter 51
1, a Y counter 512, and a comparator 513. The memory controller 520 includes the address controller 52
1, a data controller 522, a memory status register 523, and a control signal generation circuit 524.

The X counter 511 and the Y counter 512 receive the sampling clock SPCLK from the frequency divider 541.
Is entered. X counter 511 count value (0 to 6
39) is used for drive control of main scanning. Y counter 5
The twelve count values (0 to 99) are used for drive control and address designation in sub-scanning. Comparator 513 and YCUE
The register 542 is provided for adjusting the sub-scan timing described above. The YCUE register 542 has C
An optimum value according to the measurement condition is set from the PU 51.
The address decoder 543 is a circuit for switching between accessing the frame memory 60 by directly specifying an address from the CPU 51 or accessing by specifying an address from the write control unit 510. Note that the CPU 51
The data specified by the control address from and read out from the frame memory 60 is transferred to the CPU 51 via the memory control unit 520.

In writing to the frame memory 60, the address controller 521 sets the Y counter 512
The address is specified by the position data Yg from the position data Xg and the position data Xg from the galvanomirror 12X. (W) in the figure indicates that it is for writing. In the read operation, the address controller 521 operates the display control unit 53.
The address is specified by the position data Xg and Yg from 0. (R) in the figure indicates that it is for reading. For example, position data Xg, Yg is 7 bits, if the address as a frame memory 60 (Add.) Were used device is a 16-bit A ₀ to A ₁₅ is, A ₀ to A ₆ position data Xg the assignment, the a ₇ to a ₁₃ position data Y
g, and the remaining A ₁₄ and A ₁₅ are assigned to the bank designation.

The data controller 522 is responsible for writing and reading the detection data Yp as the specific data DD. The memory status register 523 stores the states of the four banks A, B, C, and D in the frame memory 60.

The display control unit 530 includes a clock generation unit 53
Read address (X) based on various synchronization signals from
g, Yg) is generated and given to the address controller 520. Further, it outputs detection data Yp from data controller 522 to a display (not shown) together with a predetermined synchronization signal.

FIG. 7 is a diagram showing an example of properly using the banks of the frame memory 60. In the three-dimensional measuring apparatus 1, the frame memory 60 has four banks A, B, C, and D.
The banks A, B, C, and D are used in a rotation format in the order of writing, displaying (reading), waiting (idle), and clearing. For example, when writing to bank C, while reading data from bank B to which the data of the previous measurement has been written before,
The data in bank D is erased. At this time, the bank A is not accessed at all. At the time of the next measurement, the current data is written to the bank D, and the previous data is read from the bank C. Such memory control enables a parallel processing operation of sequentially outputting data obtained by each measurement while periodically repeating the measurement, thereby displaying a change in the position of the moving object. However, in this embodiment, writing of data (that is, scanning for one screen) and reading for display are asynchronous. The display frame period is shorter than the scanning time for one screen. Therefore, until the writing of the latest data (frame) for one screen is completed,
The previous frame is repeatedly read and displayed. When scanning of one screen is completed in a frame cycle (for example, 1/30 second), display can be updated in a frame cycle by synchronizing writing and reading.

FIGS. 8 to 15 show modified examples of the use of the frame memory 60. FIG. In the example of FIG. 8, the detection data Yp is stored in the lookup table 71 without being stored, converted into the distance data Dz, and the distance data Dz is written into the frame memory 60. For address designation, position data Xg and Yg are used as in the example of FIG. When outputting data to the display, the frame memory 6
The distance data Dz read from 0 is directly input to the D / A converter 72 and converted into a video signal.

In the example of FIG. 9, PS is used for addressing in the X direction.
The position data Xp obtained by A / D converting the detection signal in the X direction from D25 is used. Other configurations are the same as those in the example of FIG.

In the example of FIG. 10, the distance data Dz is written into the frame memory 60 as in the example of FIG. 8, and the position data Xp is used for addressing in the X direction as in the example of FIG.

In the example of FIG. 11, the detected data Yp is used for addressing in the Y direction, and the position data Yg is written in the frame memory 60. That is, the depth of the object Q viewed from the light receiving system 20 is stored. This configuration is preferable depending on the application, but has a disadvantage in that if effective detection data Yp cannot be obtained due to an insufficient amount of incident light or noise mixed in, pixel omission occurs.

In the example of FIG. 12, the distance data Dz is written into the frame memory 60 as in the examples of FIGS.
p is used.

The example of FIG. 13 uses the detection data Yp for addressing in the Y direction, and uses the position data Xp for addressing in the X direction as in the example of FIG. Other configurations are the same as those in the example of FIG.

The example of FIG. 14 writes the distance data Dz into the frame memory 60 as in the example of FIG. 12, uses the position data Xp to specify the address in the X direction, and uses the detection data Yp to specify the address in the Y direction. It is.

In the example shown in FIG. 15, one of the position data Yg and the detection data Yp is written into the frame memory 60, and the other is used for addressing in the Y direction. Is made possible. Position data Xg for addressing in the X direction
Is used. When outputting data to the display, one of the two types of conversion data stored in the look-up table 71i is selected and used in accordance with the type of data written in the frame memory 60.

[0037]

According to the first to fourth aspects of the present invention,
Data is stored in association with the position in the field of view of the measurement and the position in the memory space as the virtual screen, and control of the memory for storing the data can be simplified.

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of a three-dimensional measuring apparatus according to the present invention.

FIG. 2 is a block diagram of a main part of a control system.

FIG. 3 is a diagram showing an output image size.

FIG. 4 is a diagram for explaining timing adjustment of sub scanning.

FIG. 5 is a diagram showing a method for preventing pixel omission.

FIG. 6 is a block diagram illustrating a functional configuration of a controller.

FIG. 7 is a diagram showing an example of the proper use of a bank of a frame memory.

FIG. 8 is a diagram illustrating a modified example of a usage form of a frame memory.

FIG. 9 is a diagram showing a modified example of a usage form of a frame memory.

FIG. 10 is a diagram showing a modified example of a usage form of a frame memory.

FIG. 11 is a diagram showing a modified example of a usage form of a frame memory.

FIG. 12 is a diagram illustrating a modified example of a usage form of a frame memory.

FIG. 13 is a diagram illustrating a modified example of a usage form of a frame memory.

FIG. 14 is a diagram illustrating a modified example of a usage form of a frame memory.

FIG. 15 is a diagram showing a modified example of a usage form of a frame memory.

[Explanation of symbols]

 Reference Signs List 1 3D measuring device VS virtual plane L light beam X main scanning direction Y sub-scanning direction sp sampling section Q measurement target DD specific data Yp detection data (spot position data) Dz distance data 24 cylindrical lens (optical member) 25 photoelectric conversion Device 52 Controller (Memory control means)

Claims (4)

[Claims] A light beam is projected so as to perform raster scanning toward a virtual plane, and is reflected by a measurement target at a point in time when the virtual plane passes through each of the sampling sections subdivided in a main scanning direction and a sub-scanning direction. A three-dimensional measuring device that outputs specific data according to the incident angle of the light beam, wherein a memory that stores the specific data, and position data in the main scanning direction and position data in the sub-scanning direction of each sampling section. A memory control unit that accesses the memory using the address as an address and writes the specific data in each sampling section. 2. A light beam is projected so as to perform raster scanning toward a virtual plane, and is reflected by a measurement object at a point in time when the virtual plane passes through each of the sampling sections subdivided in a main scanning direction and a sub-scanning direction. A three-dimensional measuring device that outputs specific data according to the incident angle of the light beam, wherein a memory stores position data in the sub-scanning direction of each sampling section; and a position of each sampling section in the main scanning direction. A three-dimensional measuring apparatus, comprising: a memory control unit that accesses the memory using the data and the specific data as addresses and writes position data in the sub-scanning direction of each sampling section. 3. The three-dimensional measuring apparatus according to claim 1, wherein the specific data is position data of a spot of the light beam incident on a light receiving surface. 4. The three-dimensional measuring apparatus according to claim 1, wherein the specific data is distance data calculated based on a position of a spot of the light beam incident on a light receiving surface.