JP3051199B2 - Object 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 an object for measuring the position and shape of a work required when performing arc welding, sealing work, deburring work, etc. by a robot or an automatic device instead of a human. The present invention relates to a shape measuring device or a robot visual sensor.

[0002]

2. Description of the Related Art The state of the prior art will be described by taking arc welding as an example. In order to automate the arc welding, it is necessary to measure the shape of the target surface and detect the position of the welding site (welding line) in order to move the welding torch gripped by a robot or the like in accordance with the work target. As such a robot visual sensor, a non-contact type sensor has no interference with a welding target, and is excellent in application flexibility to various shapes and measurement speed. As a non-contact visual sensor, a method of projecting slit light onto a target, measuring the shape of the target surface by capturing the shape of the bright line appearing on the target surface with a TV camera, and using the spot light of the laser beam to project the target surface By scanning, the position of the moving bright spot on the target surface is detected by a position detector (P
Osition Sensitive Device,
Abbreviation PSD, for example, "Optics", Vol. 12, No. 5, page 367 to page 373, 1983, published by the Society of Applied Physics of Japan.
A method of obtaining a three-dimensional position of a luminescent spot based on the principle of triangulation is known.

Further, utilizing the fact that the fluctuation frequency of the arc light intensity is concentrated below a certain frequency (100 kHz), the intensity of the laser light is increased to a sufficiently high frequency (for example, several hundred kHz).
), A bright spot image is received by a one-dimensional array of photodiodes, and only a modulation frequency component is extracted from the photoelectric conversion output of each photodiode with a filter to separate and detect only the bright spot signal. (For example,
Robot, No. 1 issued by the Japan Industrial Robot Association. 5
4, page 58 to page 65, "Development of visual sensor for arc welding robot" by Yoshikazu Ito).

[0004]

These methods, such as the method of measuring the shape by capturing with a television camera as described above, are susceptible to interference by strong extraneous light such as arc light. In particular, PS
In the method using D, since the center of gravity of the light distribution of the bright spot image captured on a single light receiving surface is detected as the central image of the bright spot, the multiple reflection generated when the surface of the V-shaped groove is scanned by the laser beam. , Simultaneous measurement of a plurality of spot images prevents accurate measurement of the true bright spot position, and tends to fail to achieve accurate shape measurement.

The latter method of separating and detecting only the signal of the bright spot is excellent in the ability to remove the influence of the arc light, but is different from the generally manufactured and widely used photodetectors such as CCD sensors and PSDs. However, since a special detector having a specific structure for detecting the modulated light is used, there is a problem that the degree of freedom in designing and manufacturing is lacking. Further, since a non-storage type light receiving element is used, there is a disadvantage that light detection sensitivity is low.

SUMMARY OF THE INVENTION An object of the present invention is to establish a method of measuring a distance and a shape capable of eliminating the influence of ambient light such as arc light in view of the above-mentioned state of the art. It is an object of the present invention to realize an object shape measuring device which can be constituted by a storage type CCD line sensor.

[0007]

The principle of triangulation according to claim 1, wherein a laser beam is projected onto the surface of the welding object and scanned, and a bright spot image is detected by a charge storage type one-dimensional light receiving element array. in object shape measuring apparatus determined, measuring the surface shape of the welding object position coordinates of the bright spot image based on the read
The scanning period was set sufficiently shorter than the fluctuation period of the arc light
A one-dimensional light receiving element array , pulse light emitting means for emitting light with a non-emission period of at least one read scanning cycle in synchronization with the light receiving and reading scanning of the one-dimensional light receiving element array , and one read output of the light receiving element array Circuit means for delaying by a scanning period, and a read scan adjacent to a read output in a read scan period next to a read scan period in which pulse emission occurs.
One of the delay circuit means between the read output during the test period and
And a circuit means for performing a subtraction process by delaying by one read-out scanning cycle, thereby measuring the shape of a welding object in arc welding.
Form a constant unit, in read output by the subtraction processing
An object shape measuring apparatus is configured to remove components due to fluctuating arc light .

[0008]

By controlling the CCD line sensor to limit the photoelectron accumulation period to a period corresponding to the pulse width of the laser light pulse (using the electronic shutter function of a normal CCD sensor), the CCD line sensor can be controlled with respect to ambient light. The sensitivity to laser light can be suppressed and the sensitivity to laser light can be selectively increased.However, under such laser light irradiation conditions, the laser light pulse is synchronized with the line scan, so that the line scan output during laser light pulse irradiation and the laser If the difference between the line scan outputs during the period in which there is no light pulse is taken, if the line scan period is sufficiently short, for example, 10 μs or less corresponding to the upper limit 100 kHz of the fluctuation frequency spectrum component of the arc light, the continuous line scan output Since the ambient light component inside has almost no change, the ambient light component in the subtraction output is removed and the signal by the laser pulse light is removed. By the amount is detected.

[0009]

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, 1 is a shape measuring object, 2 is a laser diode, 3 is a laser diode driving circuit, 4 is a mirror type optical deflector M for deflecting laser light and scanning the measuring object with laser light, a galvanometer and a mirror. Can be realized by a known optical deflection unit such as a motor mounted on a motor shaft. Also, depending on the application, an ultrasonic optical deflector that can operate at high speed with a small deflection angle as a non-mechanical deflector can be substituted. Reference numeral 5 denotes an optical deflector driving circuit.
Reference numeral 6 denotes a lens for detecting the laser light projected on the target surface as a bright spot image. Reference numeral 7 denotes a light receiving element array for converting a bright spot image into an electric signal, a CCD line sensor being a typical example. Here, in order to obtain a specific description, 128
Assume a CCD line sensor having two effective light receiving elements. Each light receiving element generates a pixel signal representing a bright spot image. Reference numeral 8 denotes a line sensor reading circuit that generates various drive signals necessary for operating the CCD line sensor 7, supplies the generated drive signals to the line sensor 7, amplifies and outputs the read pixel signals. Here, as an example of the configuration of the CCD line sensor 7, in order to double the reading speed of the pixel signal, the odd-numbered and even-numbered light-receiving elements are simultaneously read out in parallel so that the odd-numbered and even-numbered light-receiving elements are simultaneously read out. The one having two CCD shift registers that receive the parallel transfer of charges from the light receiving element will be described. Therefore, the two signals 9 and 10 output from the line sensor reading circuit 8 are defined as a pixel signal of an odd-numbered light receiving element row and a pixel signal of an even-numbered light receiving element row, respectively. Here, it is assumed that the frequency of the pixel clock signal applied to the CCD line sensor for reading out the pixel signal is 20 MHz. Numeral 11 denotes a timing circuit for generating various signals necessary for the operation of the object shape measuring instrument of the present embodiment. A signal 12 is a pixel clock signal φ, and a signal 13 is a line scanning start for giving a repetition cycle of reading of the CCD line sensor. The signal ST and the signal 14 are a mirror advancing signal MP for setting the light deflection angle θ _i by the rotation of the mirror of the mirror type optical deflector 4, and the signal 15 is between the initial position and the final position of the light deflection scanning by the mirror. Signal MA indicating the effective light deflection period of signals 12, 13, 1
4 and 15 are synchronized with each other. Reference numeral 16 denotes a gate circuit that passes the mirror step signal MP only during the effective light deflection period indicated by the signal MA. Reference numeral 17 denotes a light emission signal (LP) 1 having a pulse width corresponding to a certain laser emission time after a certain delay time after receiving the mirror step signal MP passed through the gate circuit 16.
8 is a light emission signal generation circuit that outputs 8. The optical deflector driving circuit 5 receives the input of the mirror step signal MP14, increases the mirror angle of the mirror type optical deflector 4 by a certain angle, and deflects the laser light to a new irradiation position on the measurement object. An electric current is generated and supplied to the mirror type optical deflector 4. After the mirror type optical deflector 4 is set to a new light deflection angle, the laser diode drive circuit 3 receives the input of the light emission signal 18 and supplies a drive current pulse for giving the laser diode 2 a required light emission output. . Optical deflector drive circuit 5
Receives the total number of mirror step-up signals MP input during the duration of the effective light deflection period signal MA and sends a drive signal to the mirror type optical deflector 4 for returning to the initial position when the light deflection operation at the final position is completed. Send out. The line sensor reading circuit 8 receives the input of the pixel clock signal (φ) 12 and the line scanning start (ST) 13 and
Generates various drive signals necessary to operate the.

The circuit block 19 is a main part of the object to be measured according to the embodiment of the present invention.
In response to the input of the read output 9 of the odd-numbered light receiving element rows, the image signal of the bright spot image by laser projection on the measurement object is separated from extraneous light and various interference signals and detected, as described below. Operate. Reference numeral 19-1 denotes an analog / digital converter (AD converter) which converts an analog pixel signal 9 into an n-bit (n = 10, for example, n = 10) digital signal. I do. The converted output is held in register A of 19-2, and n shift registers 19-3-1 to 19-3 are provided.
−n is input and held in parallel for each bit. The number of stages of each shift register is determined by the line scan start signal (S
T) Pixel clock signal (φ) included in the generation cycle of 13
Twelve (the number 64 of odd-numbered light receiving elements in this embodiment is 64)
Shift register 19-3-
The output of the last stage of 1-19-3-n is supplied to the A / D converter 19
A pixel signal delayed by one line read time is given to the output of -1. This delayed pixel signal is
4 is held in the register B. Reference numeral 19-5 denotes a subtracter having n bits, and the content RA of the register A is set as a minuend and the content RB of the register B is set as a decrement.
To the register C. The above-described A / D converter and each register operate upon receiving the input of the pixel clock signal φ12. 19-7 is an n-digit register (register T), which is a reference value R to be compared with the content RC of the register C.
Holds T. Reference numeral 19-8 denotes a comparator or an encoder, which stores the contents RC of the register C and the contents R of the register T.
T is input and the output is, for example, a difference value or a binary code [when the difference value is positive (RC> RT)
"1" and "0" when the difference value is zero (RC = RT) or negative (RC <RT). Alternatively, a more general encoded output that can be arbitrarily set by a logic circuit design can be generated. The output of the comparator 19-8 is stored in the buffer memory 19-9 for each pixel.
The buffer memory 19-9 stores the odd-numbered effective pixels (64 in this embodiment) of the line sensor 7 and the measurement points per one deflection cycle of the laser beam by the mirror for one scan of the measurement target by the light deflection. , Ie, the mirror step signal M within the effective light deflection period input to the optical deflector drive circuit 5
It has a storage capacity capable of accumulating pixel data (the output of the comparator 19-8) equal to the product of the number of Ps (256 in this embodiment). Furthermore, since writing of pixel data occurring during the subsequent light deflection period is performed in parallel with reading of accumulated data during the immediately preceding light deflection period, a double memory (double buffer) having a double memory having the same storage capacity is prepared. Memory) configuration, 2 times each time light deflection is repeated.
One memory is used alternately for writing and reading data. Next, the write operation to the buffer memory 19-9 will be described. Reference numeral 19-10 denotes a gate pulse generation circuit which starts counting the pixel clock signal φ by, for example, generation of a mirror step signal MP which has passed through the gate circuit 16, and raises a gate pulse when the counted value reaches a certain value. When the count value reaches the next constant value, the gate pulse falls so that the CCD line sensor 7 is enabled during the read output of the CCD line sensor 7 immediately after the light emission signal LP that is generated with a certain time delay from the mirror step signal MP. The pixel signal (64 in this embodiment)
Gate pulse signal V indicating the occurrence section of the (odd-numbered pixels)
A is generated. Reference numeral 19-11 denotes a gate circuit, which selectively passes the pixel clock signal 12 in the effective pixel section in response to the input of the gate pulse signal VA. The effective pixel clock signal is counted by a 19-12 X address counter, and the value of the counter becomes a part of the memory address as a count value j indicating the pixel position of the bright spot image of the projected laser light on the measurement target. 19-13 is a θ address counter which is a gate circuit 16.
And outputs a count value i indicating the deflection angle of the laser beam as a part of the memory address. Reference numeral 19-14 denotes a write control circuit. When a valid pixel clock signal is input to the write enable input terminal W, a buffer memory 19 designated by the value of the X address counter 19-12 and the value of the θ address counter 19-13. The output of the comparator 19-8 corresponding to the effective pixel clock signal is written to the address -9. The X address counter 19-12 is reset by the scanning start signal ST at the beginning of the reading of the line sensor 7.
On the other hand, the θ address counter 19-13 outputs a count-up signal 19-15 when the total number (256 in the present embodiment) of the effective mirror step-up signal MP is counted, and outputs the count-up signal 19-15 to switch the buffer memory of the write control circuit 19-14. The signal is input to the signal terminal C, and writing is performed in the next light deflection period. The writing destination is switched to the other buffer memory, and the θ address counter itself is reset. The count-up signal 19-15 is also input to the interrupt processing circuit 21, and notifies the data processing device (CPU) 23 of the completion of the writing via the system bus 22. CPU2
Receiving the notification of the completion of the writing, the instruction 3 again instructs the reading control circuit 24 via the system bus 22 to read the pixel data from the buffer memory in which the writing has been completed. At this time, a circuit block 20 having the same configuration as that of the circuit block 19 is prepared for the pixel signal 10 in the same manner as the circuit block 19 is prepared for the pixel signal 9. The generated pixel data is stored in the double buffer memory 2 included in the circuit block 20.
Store it in 0-9. Therefore, the CPU 23 alternately issues a read command for the buffer memories 19-9 and 20-9 to the read control circuit 24 to read all pixel data (128 × 256 in this embodiment). For the pixel having a value greater than zero in the read pixel data, the projection position of the laser beam is determined from the X address value j specifying the array on the light receiving element row and the θ address value i specifying the deflection angle of the laser beam. the position coordinate value x _P and z _P of calculated and stored in the position coordinate memory 25. The stored content gives the shape (specifically, position coordinates) of the trajectory on the target surface scanned by the laser light. The obtained position coordinate data is output through an input / output interface circuit 26 to display the shape of the object on a CRT display device or to perform further shape recognition processing on the position coordinate data by the CPU 23, for example, for arc welding. Then, the position of the welded portion can be detected and the position data of the welded portion can be transferred to the welding robot via the input / output interface circuit 26. Reference numeral 27 denotes a system control interface circuit. At the start of measurement, the CPU 23 sends a register (T) 19-7 from the bus 22 via the circuit 27.
, And the system components such as the timing circuit 11 are set to the initial state.

Next, the timing relationship of signals in the configuration of the above embodiment will be described in detail with reference to FIG.
φ indicates a pixel clock signal, and ST indicates a line scanning start signal. TR generated in synchronization with the line scanning start signal ST is a transfer signal, which is applied to the gate electrode of the CCD line sensor 7 which controls the transfer of the accumulated charge of the bright spot image accumulated in the light receiving element to the CCD shift register. The timing at which the input charges accumulated in the light receiving element array are simultaneously transferred to the CCD shift register in parallel is given. After the transfer, the CCD shift register is driven by the pixel clock signal φ, and the pixel signal is read. After the transfer of the stored charge,
The light receiving element array is ready for the next charge accumulation. PR is a stored charge reset signal, which is input to a gate electrode for controlling charge storage and dissipation of the light receiving element, and prevents charge storage during the continuation of the signal (when in the ON state). Accordingly, when the stored charge reset signal PR is turned off, charges can be stored, and charge storage of a bright spot image is performed. That is, the stored charge reset signal PR realizes a so-called electronic shutter function. Such a CCD line sensor with an electronic shutter function is commercially available. LP
Is a light emission signal, and in this embodiment, a line scanning start signal S
T occurs every ten cycles, and occurs in synchronization with, and only during, the period in which the stored charge reset signal PR is off. Therefore, the laser light controlled by the light emission signal LP is used as a signal light as compared with the case where the external interference light irradiates the light receiving element surface for the entire time and the accumulated charge reset signal PR causes a loss of charge during the ON period. Stored charge reset signal PR
During the period when is turned off, the entire amount of light is effectively converted to electric charge, and there is no loss of light energy. Therefore, at this point, it is possible to improve the signal-to-noise ratio between the output of the bright spot image and the output of the extraneous interference light. VA is a gate pulse signal indicating a period in which the effective pixel signal occurs. RA is the register (A) 19
In order to make it visually easy to understand, the binary code is converted into an analog signal and displayed as an amplitude. RB is the content of the register (B) 19-4, which is also indicated by the signal amplitude value of analog display. RC is an analog display of the contents of the register (C) 19-6, and a signal RB delayed by one line cycle from the signal RA is subtracted by the subtractor 1.
This is the result of subtraction according to 9-5. Signal part 2 on RA
Reference numeral 8 denotes a signal in which an output signal of a bright spot image is superimposed on an electric output 29 due to extraneous light. Since the electrical output 30 of the signal RB in the corresponding time section is delayed by one line cycle, the output signal component of the bright spot image is not included and only the output due to extraneous interference light. Here, the pixel clock frequency is 20 MHz,
Since the number of light receiving elements in the odd-numbered (or even-numbered) array is 64, the line scan start signal cycle is set to 4 μs (> 50).
ns × 64 = 3.2 μs). Therefore,
Since the fluctuation frequency of most components of the external interference light such as arc light is sufficiently lower than 100 kHz (the fluctuation period is sufficiently longer than 10 μs), the difference between the electrical outputs 29 and 30 of the external interference light due to the delay time difference of one line period. Is small enough. Therefore, the output 31 in the section corresponding to the subtraction output RC is almost only the output signal of the bright spot image from which the extraneous interference light output is removed. The output in this section is written to the buffer memory 19-9 (or 20-9) only for the continuous section of the gate pulse signal VA.
Further, by such a delay subtraction operation, an unnecessary signal in a fixed pattern generated due to a sensitivity deviation of each light receiving element is also removed at the same time.

FIG. 3 shows the relationship between signals in one deflection section of the mirror type optical deflector 4. θ _i is the deflection angle of the laser beam with respect to the x-axis, and the positive (upward right) inclined portion indicates the deflection angle during the effective light deflection period. MA is an effective light deflection period signal indicating an effective light deflection period, and MP is a mirror stepping signal. LP is a light emission signal, and includes pulse signals of a number (256 points in this embodiment) equal to the number of measurement points during the effective light deflection period. C is a count-up signal of the θ address counter 19-13 indicating that the deflection scanning of all the measurement points in one deflection period has been completed, and is a buffer memory 19-9 (or 20-
9), resetting the θ address counter 19-13, and interrupting the CPU 23.

FIG. 4 shows a deflection angle θ _i = _i determined from the θ address value i and the X address value j of the pixel data of the bright spot image stored in the buffer memory 19-9 (or 20-9).
ai + θ _o (i = 1, 2,..., 256) (where a,
_θo is a constant) and the X-coordinate value x _j = k of the odd-numbered light receiving element
× (2j−1−64) (j = 1, 2,..., 64) (where k is a constant) or the X coordinate value x of the even-numbered light receiving element
_j = k × (2j−64) (j = 1, 2,..., 64)
(Where k is a constant) from the coordinates x _p , z of the laser irradiation point P
_The method of calculating _p is shown. 2-1 is a deflected laser beam, and 2-2 is a reflected or scattered light from the irradiation point P. z is a coordinate axis made coincident with the optical axis of the lens 6, and x
Is a coordinate axis orthogonal to the z-axis and passing through the center O of the lens 6, both being in the light deflection plane (paper plane). The rotation axis of the mirror of the mirror type optical deflector 4 is perpendicular to the paper surface, and d is the distance between the rotation axis of the mirror intersecting the x axis and the lens center O. 1
Is the distance between the lens center O and the light receiving surface of the light receiving element row 7. Note that instead of calculating the values of x _p and z _p by the CPU each time using the calculation formula shown in FIG. 4, the coordinates x _{p of} the bright spot corresponding to the address values i and j of the pixel data of the bright spot image are used.
, Z _p are calculated in advance and stored in a look-up table, and the address values i,
The coordinate values x _p and z _p can be obtained at high speed by referring to the lookup table every time j is obtained. FIG. 5 shows an embodiment in which position coordinate data is obtained by using a look-up table. Components common to the embodiment shown in FIG. 1 and not directly related to the description are omitted. Reference numeral 19-16 denotes a lookup table memory which receives the input of the X address value and the θ address value from the write control circuit 19-14, and receives the coordinate value x _{p of the} bright spot.
And z _p are output. 19-9' as address coordinates x _p and z _p shape memory, the comparator 19-8
Is stored. Shape memory 19-
9 'represents the shape of the surface to be measured scanned by the laser beam. Shape memory (address size) of memory 19-9' depends on the calculation accuracy of the coordinate values x _p and z _p (representing the precision of shape). Shape memory 19-
9 'has a double buffer configuration similarly to the buffer memory 19-9, and the write control circuit 1 is controlled by the count-up signal 19-15 of the θ address counter 19-13.
The operation state of each constituent memory is alternately switched between a write operation and a read operation through the control of 9-14. The modification of the embodiment based on FIG. 5 described above is, of course, according to the embodiment of FIG.
At the same time, a look-up table memory 20-16 and a shape memory 20-9 '(both not shown) are provided instead of the buffer memory 20-9. When the CPU 23 receives the interrupt of the count-up signal, it sends a read command to the read control circuit 24 to display the contents of the shape memories 19-9 'and 20-9', and to display the V-groove shape from the obtained shape data. The coordinate value of the V-groove of the arc welded portion is detected, the coordinate value is output to the outside, and transferred to a robot that holds the welding torch to perform the welding operation.

In practice, erroneous pixel data may be detected due to a force-caused disturbance, or pixel data may be lost without being detected because the bright spot image is extremely weak. 19-9 and 20
The contents of -9 or the shape memories 19-9 'and 20-9' often contain defects as data compared to the ideal state. Therefore, it is effective in many cases to remove unnecessary information by performing preprocessing by the CPU on pixel data in the buffer memory or the shape memory. At this time,
The appropriate pretreatment depends on the specific shape measurement target and the measurement conditions. The means for realizing the pre-processing depends on the specific configuration of the CPU 23, and can be realized by program control by a general-purpose processor, using circuit means when high-speed processing is required, or using various existing data processing. Technology is applicable. The same can be said for the shape recognition processing performed as necessary in the CPU 23.

In the embodiment shown in FIG. 1, a subtraction process is performed between outputs having a time difference of one line cycle of the CCD line sensor.
The example of the delay means for performing the / D conversion, digitally encoding, and obtaining the delayed output using the digital shift register has been described. As an alternative embodiment, the output of the CCD line sensor is delayed using an analog CCD shift register or an ultrasonic delay line as an analog signal, and the original output and the delayed output are input to the differential amplifier to subtract the output. You may get it. In this case, writing to the buffer memory is performed by A / D converting the subtraction output. Embodiments of various modified circuit means such as a memory configuration method, a coordinate calculation means based on the principle of triangulation, including an alternative embodiment of such a delay means, are based on the basic contents of the present invention. It can be easily devised by those skilled in the art, and is not limited to the embodiment shown in the drawings.

[0016]

As described in detail above, claim 1
According to the object shape measuring apparatus of the present invention, the output of the CCD line sensor is subtracted from the output delayed by one line scanning period, and the subtraction is performed based on the external interference light and the sensitivity deviation of the light receiving element. Unnecessary signal components can be removed, and the surface of the object to be measured has a low reflection coefficient, so that the light intensity is weak. Further, since laser light that emits pulses in synchronization with the operation of the CCD line sensor is used, the contribution of the background light to the output of the CCD line sensor is suppressed, and only the signal light due to the laser light is efficiently detected. Can be expected to improve the signal output to unnecessary signal output ratio. Furthermore, since a charge storage type light receiving element is used,
Compared to the conventional light modulation type interference light elimination method that requires a non-storage type light receiving element, there is an advantage that the detection sensitivity of a bright spot is high and the method can be widely applied to a measurement object with a low reflectance. The object shape measuring apparatus according to the present invention having such features is, in particular,
It is useful as a visual sensor for arc welding robots that need to operate in the strong disturbance of arc welding light.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention of an object shape measuring apparatus for measuring a shape of an object surface by detecting a bright point of a laser beam scanning a surface of a measurement object.

FIG. 2 is a diagram showing a timing relationship of main signals at the time of detecting a bright spot in the embodiment of FIG. 1;

FIG. 3 is a diagram showing a timing relationship of various signals related to a mirror deflecting operation and a laser light pulse emission operation during one optical polarization scanning by a mirror.

FIG. 4 is an explanatory diagram showing a configuration of an optical unit normally used in an object shape measuring apparatus and a principle of calculating position coordinates of a laser irradiation point.

FIG. 5 is a partial configuration diagram of an embodiment for calculating position coordinates using a table lookup memory.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Shape measurement object 2 Laser diode 3 Laser diode drive circuit 4 Mirror deflector M5 for deflecting laser light and scanning a measurement object with laser light 5 Optical deflector drive circuit 6 Laser light projected on the target surface Lens 7 for detecting a bright spot image 7 Light receiving element array for converting the bright spot image into an electric signal 8 Line sensor reading circuit 9, 10 Pixel signal 11 Timing circuit 12 Pixel clock signal φ 13 Line scanning start signal ST 14 Mirror Step signal MP 15 Signal MA indicating effective light deflection period MA 16 Gate circuit 17 Light emission signal generation circuit 18 Light emission signal (LP) 19 Circuit block 19-1 A / D converter 19-2 Register A 19-3-1 Shift register 19-3-2 Shift Register 19-3-n Shift Register 19-4 Register B 19-5 Decrease Unit 19-6 Register C 19-7 Register T 19-8 Comparator or encoder 19-9 Buffer memory 19-10 Gate pulse generation circuit 19-11 Gate circuit 19-12 X address counter 19-13 θ address counter 19-14 Write Control circuit 19-15 Count-up signal 20 Circuit block 20-9 Double buffer memory 21 Interrupt processing circuit 22 System bus 23 Data processing device (CPU) 24 Read control circuit 25 Position coordinate memory 26 Input / output interface circuit 27 System control interface circuit 28 Signal part on RA 29 Electric output due to extraneous disturbance light 30 Electric output 31 Output signal of bright spot image φ Pixel clock signal ST Line scan start signal TR Transfer signal PR Accumulated charge reset signal LP Light emission signal VA Gate pulse signal RA Contents of register (A) 19-2 Contents of RB register (B) 19-4 RC Contents of register (C) 19-6 θi Deflection angle of laser beam with respect to X axis MA Effective light deflection Period signal MP Mirror step signal LP Light emission signal C Count-up signal of address counter 19-13

Claims (1)

(57) [Claims] 1. A laser beam is projected onto the surface of a welding object to perform scanning, a bright spot image is detected by a charge storage type one-dimensional light receiving element array, and the position coordinates of the bright spot image are determined based on the principle of triangulation. In the object shape measuring device for measuring and measuring the surface shape of the welding object, the readout scanning period is longer than the fluctuation period of the arc light.
A one-dimensional photodetector array which is divided short set, a pulse emission means for emitting across the no-light emission period of more than 1 read scanning period in synchronization with the received and read scanning of the one-dimensional photodetector array, photodetector arrays and a circuit means for delaying the read output by one read scanning period, the next read scanning period of the read output and the next read scanning period that occurs in the pulse emission
One between the read output during the read scan period
Delayed by one read scanning cycle by the delay circuit means
It was put in arc welding by a circuit means for performing subtraction processing
An object shape measuring device comprising a welding object shape measuring device , wherein a component due to a fluctuating arc light in the readout output is removed by the subtraction processing.