JPS6361110A - High-speed three-dimensional measuring method and device - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a high-speed three-dimensional measurement method and apparatus for three-dimensionally measuring an object on a sample stage at high speed and forming three-dimensional information related to the object.

In particular, the present invention provides three-dimensional measurement of an object on a sample stage at high speed by irradiating the object with a controlled light beam and detecting the position of a laser light spot on the imaging plane, thereby obtaining three-dimensional information related to the object. The present invention relates to a high-speed three-dimensional measurement method and apparatus for forming.

[Prior Art and Problems to be Solved by the Invention] Many mechanical measurement devices require more information to be inspected when measuring different depths, especially when the gray scale or intensity contrast is poor and difficult to distinguish. Must be. In fact, this device is rarely used for two-dimensional use, and devices for three-dimensional use have been proposed. SMD (Surface Mounted Device) is a good example of a device where depth inspection is useful in determining the presence and orientation of parts. If you do not do this, changing the color, fabric, background, etc. of parts, etc.
Special techniques are usually required.

Even in standard circuit testing techniques, it is desirable to be able to test depth. For example, when an operator inspects an object using a microscope, both color and depth are used.

Techniques for detecting depth are passive when a controlled radiant light source is not required, and active when they utilize a beam of radiant energy. Passive techniques have little to do with the object to be measured or its context; computer imaging and psychophysics are the main research topics. For example, methods based on stereoscopic mismatch, camera motion, surface reflections, fineness, shadows, and light absorption have been investigated. These techniques are often associated with psychophysics. For example, the human visual system's ability to judge depth is thought to be a combination of these methods.

The disadvantage of passive techniques is that they require a large amount of computation to map depth. Methods based on 3D mismatch and Kamen motion are very powerful, but require consistency in a series of images. A method for structurally generating this number has not yet been developed for real-time computer equipment. However, several techniques for depth judgment have been developed, including techniques for displaying surface characteristics.

Active techniques overcome the inconsistency problem and directly measure depth by using energy beams and recording time of flight. (U.S. Patent 4,
212,534, etc.). Depth can also be measured by repositioning (triangulation and grid encoding), phase displacement of a laser beam relative to a reference beam (laser radar), shadow length and directional illumination).

This eliminates the need for numerous calculations to create a depth map, and information processing only requires calculation of three-dimensional characteristics, surface conditions, and image analysis. Devices with improved differentiation by intensity and color can use both area and intensity data.

Triangulation or optical constructs are dense, high-resolution (approximately 1 m
11 or more) and high-speed data (10Mllz) can be provided at low cost. This triangulation method is the oldest depth detection method, but it continues to make new developments. On the other hand, the laser/radar method is relatively new as a mechanical measurement method, and although it has advantages, the data rate is relatively slow (and the modulation frequency is wide up to the GHz band, so the cost required for high resolution is high). Triangulation methods, on the other hand, have the advantage of being relatively simple and inherently have high resolution capabilities.

As an improvement to the basic triangulation method, single and compound stripes (
grid pattern) projection, striped scanning, flying spot
There's a scanner. An example of a three-dimensional measuring device using structured light is US Pat. No. 4,105,925. The device has a linearly arranged sensor arranged to measure the light beam only when the reference surface is illuminated.

If an object is present, the beam will be destroyed. The object is scanned by a linear sensor so that two series of images are obtained,
You can know the existence and direction of the object.

Commercially available three-dimensional observation devices for measuring height include microprocessor-based laser beam segmentation devices. This device has a data rate of approximately 30kllz, with a data rate of 48
It is possible to create 60 fields of x, y, z coordinates of 0. Each data string can field the entire image. A 1/2 inch square object has an x of 1 mm,
When measuring at y,z resolution, the maximum speed at which 1 (11)% of parts flowing on a conveyor belt can be inspected is approximately 30 seconds per inch.

This single-striped device is ideal for calibration and partial inspection, and is suitable for stable images that do not change frequently.

Rather than image processing or 1 (11)% inspection, these devices
Rather, it is mostly used for calibration. To obtain adequate sampling and dense three-dimensional information, the stripes must be scanned across the object and observed by an array of cameras or an array of sensors that can limit the data rate.

One method of obtaining data in a coordinate measuring device is to replace the line scan or column sensors commonly used in coordinate measuring devices with transverse effect photodiodes, as disclosed in U.S. Pat. No. 4,375,921. There is.

This method improves ambiguity in multi-striped devices and also provides relatively large measurable range changes. Therefore, the entire surface can be detected, and there is no need to divide the detector area as in the case of a multiple striped device.

However, in many of the above devices, the bandwidth of the amplifier circuit is IMH.
2 or less. Dual detector device (three chambers) is 30MHz
However, basic triangulation methods are not effective in low light conditions, especially when examining large fields of view. These devices are also sensitive to mottling and geometric distortions.

U.S. Patents 4,068,955 and 4,192,612
The thickness measuring device in the above issue obtains data for measuring the distance or thickness of objects at a distance by means of well-known trigonometry, the light beam being directed by a beam splitting mirror opposite the surface of the object to be measured. It is being Approximately measure the thickness of the object by appropriate trigonometry to ensure the relative angles of incidence and reflection with respect to the object surface.

US Pat. No. 4,472,056 discloses an apparatus for detecting the shapes of three-dimensional products and parts, such as the soldering area and mounting parts of printed circuit boards, and the fluctuations of LSI bonding processes. This device has a scanning section that forms a bright line image using an imaging lens in the height direction of the object, and a unidirectional angry imaging device that automatically scans the bright line image using a row of image sensing elements perpendicular to the scanning direction of the rotating section. It consists of

Since this device has a very limited readout time, three-dimensional points must be measured with many photodetectors.

The device of US Pat. No. 4,355,904 measures depth using a pair of photodetectors, such as photodiodes, and a partially transparent filter. Its centroid is analog
Calculated by divider.

The method and apparatus of U.S. Pat. No. 4,553,844 scans a speckle beam over an object in one direction and detects the speckle image transverse to that direction.

The device of US Pat. No. 4,645,917 is a scanning aperture flying spot profiler. The sensor is a photomultiplier or an avalanche diode.

The method of US Pat. No. 4,349,277 is a parallax method of wavelength labeling based on optical triangulation.

A single processing step is involved to calculate the normalized signal, independent of surface reflections and large variations.

The device of U.S. Pat.
The surface profile is measured using a photomultiplier tube. As a noise countermeasure, amplitude modulation is applied to the laser beam. As a measure against background optical noise, a filter network is used in the photomultiplier.

In addition, as related technology, U.S. Patent No. 4.053,2
No. 34, No. 4,065,201, No. 4,160.599, No. 4.201,475, No. 4,249,244, 4,
269.515, 4.411,528, 4,525
．． No. 858, No. 4,567.347, No. 4.569,07
There is No. 8.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a high-speed, high-sensitivity method using a photodetector and a signal processing circuit that has a wide dynamic range, is relatively simple and inexpensive, and generates three-dimensional information related to the object to be measured. An object of the present invention is to improve a method and apparatus for three-dimensionally measuring an object on a sample stage at high speed so as to be able to achieve the following.

It is also an object of the present invention to move objects on a sample stage at high speed, beyond many of the limitations of the prior art, by optically processing the reflected light signal by a pair of optical components that improve the quality of recovered data. An object of the present invention is to provide a method and apparatus for three-dimensional measurement.

Furthermore, it is an object of the present invention to provide a relatively inexpensive, compact,
Improved method and apparatus for high-speed three-dimensional measurement of objects on a sample stage that yields high-resolution three-dimensional information related to the object to be measured and that can be interfaced with standard video speed equipment. It is to be.

[Means for solving problems]

The first method of the present invention includes the steps of scanning a surface of an object with a controlled light beam at a predetermined angle to generate a reflected light signal, and converting the reflected light signal to a second angle by a pair of optical components. optically splitting the received signal into a first beam and a reference beam; measuring the radiant energies of the first beam and the reference beam, respectively; a step of forming a second electric signal, a step of normalizing the first and second electric signals and keeping them within a predetermined range, and determining a centroid value of the first ray from the normalized electric signal. This is a high-speed three-dimensional measurement method that three-dimensionally measures the object on the sample stage at high speed and forms three-dimensional information related to the object.

The first inventive device of the present invention includes a light source that scans a surface of an object with a controlled light beam at a first predetermined angle to generate a reflected light signal, and a pair of optical components that directs the reflected light signal to a second angle. a pair of optical components for spatially filtering and smoothing the received signal and optically splitting the received signal into a first beam and a reference beam; first and second measuring means for respectively measuring the radiant energy of the surface light beam and forming first and second electrical signals proportional to the measured values; and a signal processing means for calculating the centroid value of the first ray from the normalized electrical signal, and forming three-dimensional information related to the object on the sample stage. .

Preferably, the method according to the invention includes, before the measuring step, the step of imaging the first and second light beams at first and second adjacent positions.

In the apparatus according to the present invention, the light source is preferably a laser scanner, the pair of optical components preferably includes an optical filter and a position detection component, and the first and second measuring means are preferably a laser scanner. It is preferred to include a sensitive photodetector that converts the radiant energy of the vI light cotton into an electrical current.

[For production]

This configuration provides a wide dynamic range for the signal processing circuitry, improves the quality of recovered data, can interface with standard video speed equipment, is highly sensitive, and allows objects on the sample stage to be moved at high speeds. It becomes possible to measure three-dimensionally.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.

In FIG. 1, the main parts of a basic three-dimensional measuring device are indicated collectively at 10. These assembled parts IO are placed on the sample stage, and the length of the controlled light source such as a laser, the condensing lens, etc.
Assembly 12, scanner and beam deformation focusing optical system 2
Consists of 2. Incidentally, the scanner and the light beam forming/focusing optical system 14 are connected to the reflecting surface 1 of the object, which is generally indicated by the reference numeral 18.
This is for continuous irradiation of the laser beam 14 to the laser beam 14. The object 18 is placed on the base plane 20 of the sample stage.

FIG. 2 is a diagram related to FIG. 1 and conceptually illustrates basic triangulation, ie, structured light, including equations that describe the relationships between variables. Basically, the height Z of the object 18 is calculated by the illumination angle θ and the polarization angle θ4. In addition,
The reflected light beam is incident on the detection element 28 of the detector 24.

Typical conventional multiple sensing elements are linearly arranged or arrayed sensors or, in the case of a single sensing element, cross-effect photodiodes or pi cells (duplex ~\
diodes). A dual-barrel %% photodiode is used as a position sensing device,
It is possible to measure minute changes in height over a limited height range.

The above system uses transversal effect photodiodes (L E P
) allows the Z-axis to be measured very accurately.

However, the internal resistance of the diode generates a noise current that is large enough to become a major noise source in the measurement device. (The same diode detects the height by attenuating the signal current, and the LEPO
It is possible to calculate the centroid. ) This noise current includes noise in a large dark current in addition to the Johnson current due to resistance. For example, commercially available high quality LIEPs produce noise currents that are orders of magnitude higher than silicon photodiodes. Compared to photomultipliers or avalanche detectors, LEPs have slower sensitivity, but PM tubes and APDs have orders of magnitude better sensitivity. The resistance and capacitance of LEP devices impose limits on bandwidth. A preferred technique for detecting depth can derive the centroid of a light spot by attenuation of the light intensity. Furthermore, it is possible to derive the electrical/optical relationship of the cross-effect photodiode. However, high-speed detectors with high sensitivity can also be used.

The sensitivity of basic triangulation depends on the height to reference line ratio and decreases with steep illumination and measurement angles. Sensitivity also depends on the spacing between the detection element 28 and the effective focal length of the focusing lens 26 of the detector 24. Detection element 28 and lens 2
Increasing the interval between the numbers 6 and 6 decreases the sensitivity.

Synchronous scanning, telesend link scanning, reverse scanning (desc
Scanning methods such as anning) significantly alleviate the above-mentioned drawbacks. These scanning methods allow depth measurements to be made with high resolution even at large angles. These scanning methods can be used in position sensing and signal processing equipment without modification. The device of the present invention can improve performance by using these scanning methods.

In Figure 1, shadows and light absorption, ie non-measurable areas of the detector, reduce resolution. For accurate measurements of fairly flat surfaces such as solder tar, ink thickness, surface flatness, etc., an illumination angle of 30° is preferred and reduces light absorption effects. Since the size and shape of objects vary, such as circuit boards, an illumination angle of 10° to 156° is preferred.

The lack of detection and problems can be overcome by using multiple detectors or automatically synchronized scanning methods with a field of view within 10 degrees.

The laser 12 and scanner 22 of the present invention are flying
Preferably, a spot scanner is defined. Laser 12 is coupled to modulator 13 to convert the information to high frequency and improve the noise characteristics of the device. Modulator 13 is capable of performing one of a variety of modulations including sine wave, pulse magnitude, pulse position, and the like. Laser 1
2 is a solid-state laser diode, and a TTL signal (
It is preferable that the shuttering be performed using TTL modulation.

The laser signal can be encoded and perform separate signal processing functions in the on and off states. Typical power levels are 20-30 mW (Class III-B), suitable for mechanical measurement applications.

The scanner 22 includes a rotating polygon mirror (X scanning) and a galvanometer (
One of the following is used: a linear resonant scanner (X-scan) and a galvanometer (X-scan), or an acousto-optic device (X-scan) and a galvanometer (X-scan). Preferably, scanner 22 comprises an acousto-optical device. This is because it has no moving parts and the rescan time is fast.

The sensor can be used in conventional triangulation oblivion devices with a fairly large baseline-to-height ratio, as well as in included angle triangulation devices using hypercentric or autosynchronous scanning methods.

Another conventional method of inspecting large areas is linear scanning. In this method, the object or projection device is transformed and a galvanometer (X
scanning) is not required.

As shown in FIG. 3, an optical device 30 is used in the measuring device of the present invention, and includes an objective lens 31 that collects scattered light from a target, and an objective lens 31 that focuses the collected light on an intermediate imaging plane. It consists of a pair of optical components including two diffraction limiting lenses 32. Lenses 31 and 32 are well known, but each lens 32 is adapted to operate at a preferred conjugate angle and is interchangeable to accommodate various magnification ratios.

Apparatus 30 also includes a mask 33. Mask 33 forms a rectangular diameter aperture (spatial filter) located in the intermediate imaging plane to remove background noise (stray light) caused by secondary reflections from objects outside the preferred instantaneous field of view of the device. . The mask 33 can be a fixed aperture or an electromechanical shutter, preferably a liquid crystal binary spatial light modulator physically controlled by software. This is suitable for inspecting close, high-intensity targets (re-spilled solder, wiring adhesive, loops, bin grids, etc.) where complex reflections occur. When used with a self-synchronized scanner or a far-centered scanner (rotating mirror of a moving mechanical component), the mask 33 is an elongated strip that can collect only the light useful for Z measurements.

Spatial filters or strips can also be programmed with transmission patterns and selection op patterns that correlate with object height features. For example, measuring the height of a bright pin in a bright background can be made more reliable by having a strip corresponding to the height of the bottle. The combined reflection results in a signal return that is greater than the signal return due to the reflected light of useful light.

When using a conventional triangular scanner (a solid-state device with no moving parts and an area detector), the aperture 34 of the mask 33 is only as large as necessary to detect a particular height range; Programming is possible.

A diffuser 35 of fine granular ground glass adjoins the intermediate imaging, producing a relatively single and wide light spot. Discrepancies in the measurement positions of light spots due to local changes in the variable filter 36 or directional scattering are spatially averaged and are therefore reduced. This is similar to a low pass filter in electronic devices.

The variable filter 36 is utilized as a position dependent propagation device and is relatively insensitive in determining the center position of the light spot and narrowing and focusing the shape of the intensity distribution.

The variable filter 36 has a propagation function that is linear with respect to position, or functions as a filter with a linear density characteristic (logarithmic propagation with respect to position).

Nonlinear computational methods can compress and expand depth sensitivity throughout its range. In particular, filter 36 is used to sense small height changes near the reference line (furthest Z coordinate) without reducing depth sensitivity for taller objects. The main purpose of the filter 36 is to
As a variable density device, its nonlinear calculations are performed by relatively standard filters. On the other hand, if linearity is important during Z measurements, a linear propagation function must be used.

Apparatus 30 also includes a second lens system 37. Second lens system 3
7 propagates (relays) the image via the device 30 and reduces/enlarges the intermediate image. The cylindrical lens 38 converts light spots into light (
(by changing the aperture) to further spatially average the surface of the variable filter.

Beam splitter 39 includes a reference split beam or signal;
This results in another split beam that is propagated to filter 36. The latter divided beam is propagated to the filter 36 of the first channel 40, and the former reference divided beam is propagated to the second channel 41. The second channel 41 is used for an intensity reference to normalize the data and remove luminance effects from height measurements.

The device 30 produces both the reference beam and the propagating beam by a linear variable filter with a metal-coated front surface that produces spatial reflections, and provides position-dependent attenuation and splitting by a single optical component.

If the propagation-reflection ratio of the beam splitter is known exactly and is constant, the position can be determined by a simple division of the voltages obtained in the two channels (ratio detection). That is, Z=V1/V2. Moreover, the position can also be obtained by Z=V1/(V1+V2).

Note that the latter calculation is more preferable.

The folding mirror device has mirrors 42 and 43, and a detection unit 4
The light beam is propagated to the local plane of 4. a photodetector 45.46;
The amplifiers 47, 48 should be connected as short as possible to minimize stray capacitance due to high speed applications and avoid mismatch between signal channels. Constant polarization prisms can be used in place of mirrors 42 and 43 for ease of device adjustment. The length of the wiring must be kept short so that low level signals are not corrupted by noise.

The laser light signal propagated by filter 36 and the light signal reflected by mirror 43 are imaged onto predetermined compatible areas of a pair of photodetectors 45 and 46 in assembly 44 by conventional field lenses 49 and 50, respectively. be done.

Assembly 44 has a preamplifier 47 and 48 for the photodetector, respectively. Each photodetector consists of a small-area photodiode (less than 3m x 111cm) with small capacitance and very large shunt resistance, a photomultiplier, an avalanche photodiode, or a combination of a detection element and a pre-amplifier. Preferably, it is an enhanced detector. This type of photodiode preferably has a cut-off frequency of less than 3 (11) MHz, corresponding to a rise time of 1 nanosecond or less. Buri Amp operates at video speed.

Since a single sensitive detector is used for each channel 40 or 41, the sensitivity of assembly 44 is very high and L
The noise is very low compared to EP. In addition, photodetector/
Since the amplifier has a very large bandwidth gain, it is possible to obtain a large signal with relatively small changes in signal level.

By slightly modifying the device shown in Figure 3, two optical chambers can be created. It can be formed by removing the variable filter 36 and slightly shifting the photodetectors 45 and 46.

Two-chamber detectors are useful for obtaining high-resolution measurements when sensing only a small area (scanning on a substrate, flatness detection, etc.) and provide high-resolution over a relatively large field depth. , and is a supplementary means to the variable density filter 36 to directly measure the center position of the light spot.

The spatial average value obtained by the diffuser 35 is generally necessary to improve the accuracy of the two-chamber method. This is because the intensity distribution is not constant due to diffusion or geometric (angular) deviation.

As mentioned above, the signal indicating the height of the target is the ratio Z=V+/
(V+ +Vz), and the intensity information is 1=
V, +V, is determined. To determine Z, a commercially available analog divider is used to determine the velocity as it approaches the video frame rate. However, this type of divider allows the sum (denominator) of V, +V to vary over a small narrow range (typically 3:1) if accuracy is to be kept high. Typical imaging produces reflection variations larger than this range. For example, printed circuit board components and backgrounds produce V, +V2 signals that vary in magnitude by several orders of magnitude, exhibiting diffuse reflection variations of approximately 5% to 1 (11)%. Additionally, the cautery reflex produces larger changes, so
It is necessary to confirm that the resulting Z value is inaccurate. Note that the return of a very small photon-limited signal results in division by O.

As the devices are placed further apart, the fluctuations will be correspondingly larger and therefore the dynamic range will also be larger. In this case, real-time laser intensity modulation is required.

As previously mentioned, the dynamic range of the measurement equipment must be large enough to track diffuse reflections that vary from a fraction of a percent to 1 (11)% (e.g., black on a standard gray scale test chart). (must be substantially larger than the entire area from to white). Mainly in applications where wide range depth must be interpreted,
The device is expanded to measure weak reflections from distant, dark targets and strong reflections from nearby bright targets. This vision device (robot steering) requires a dynamic range greater than that required for inspection of small quasi-planar objects such as circuit boards.

This device can be modified to extend the dynamic range by feedback to increase or decrease the laser power from a point on the target depending on the intensity being measured.

However, in such a device, there is a reduction in data rate (by a factor of 2), and a reflection value is measured that corresponds to exactly the same physical point on the object as the Z measurement. However, with a fast rise time, I
Since high power laser diodes with feedback power supply outputs up to W are commercially available, it is possible to actually build a feedback circuit. The power output of the IW allows the device to measure objects that return approximately 50 times less signal than would be measured by the inexpensive 20 mW laser diode described above.

Additionally, dual low and medium power laser diodes that are precisely offset from each other are commercially available.
It is mainly applied to the inspection of optical discs using "read after write" technology. In many applications, medium-power lasers have good S/N and are in practical use. A major advantage is that there is no need for additional opto-mechanical hardware to maintain nearly perfect registers between each means.

Data from the low power laser diode is buffered and feedback circuit 76 limits pulse amplitude modulation of the high power laser diode.

In this sense, the feedback circuit has the ability to be modified in the future. Instead of the modulator 13, an acousto-optic modulator is preferred for time-varying amplitude laser modulation to maintain the stability of the laser source and can function as a high speed electro-optic shutter.

As shown in FIG. 5, the signal processing circuit 51 can expand and compress variable data to obtain an appropriate Z value, and can also generate special values to display inaccurate height information. .

The amplifiers 47 and 48 convert the signal currents 1+ and L of the photodetectors 45 and 46 into corresponding voltages, respectively. VI
The sum value +2 is converted into an analog signal by the summing circuit 54 and converted into a digital signal by the high speed non-linear data converter 56. The purpose of digitizing the data is that the analog divider of circuit 58 requires approximately 3
To provide a simple method for selecting a gain value (inverse function) in order to scale V and +V2 so that they fall within the range of :1. This method is similar to an AGC (automatic gain control) circuit, except that the AGC circuit often averages the signal during long forbidden periods and returns a gain control value that reduces the bandwidth of the device. . The output of the converter 56 is input to a gain selection logic circuit 57 and becomes an output signal without being fed back. The gain value selected by the logic circuit 57 is
A gain value is selected to program a series of high precision amplifiers 60 and 62 that divide the signals V, +V2 3:1.

As previously mentioned, modulation of the laser light source converts information to high frequencies that have better noise characteristics for the device. The circuit 51 connects the first and second
A noise suppression circuit is provided for each of channels 40 and 41.

When the laser source 12 is in the ON state, the first anti-aliasing circuit of each noise suppression circuit 64 or 66 (shown in FIG. 6)
The filter 68 smooths out signal changes (high frequency noise),
Remove out-of-band noise.

This high frequency noise changes rapidly compared to the known (modulated) signal. In the off state, this abrupt change is removed by the second anti-aliasing filter 68. The demodulation stage includes sample and hold circuits 70 and 72.
to remove low frequency noise. Low frequency noise changes slowly compared to the clock speed of the device. Therefore,
The average value is obtained by circuit 70 during the off period to create a reference voltage. The reference voltage is subtracted by circuit 74 from the on-voltage value obtained by circuit 72, removing the low frequency portion. Low frequency (1/F.

60 Hz, etc.) noise suppression is important to maintain a large dynamic range.

This noise suppression has the advantage of being able to provide a black base level for each pixel, which is not possible with conventional video devices. By using black as IQ, it is possible to eliminate the drift of DC offset in the signal processing network. As a result, signal fidelity is significantly improved. This method is
It also has an automatic calibration function. The laser output is controlled by circuit 13 so that it is referenced to white since the drift of the slow term in light level is very small. Since the black reference is performed for each image, the entire light range is calibrated on an image-by-image basis.

After this, the signal from circuit 51 is preferably amplified and coupled to A/D converter 78. A/
Instead of a D-converter, it can also interface to a larger test and comparison product, a conventional video frame grabber.

With respect to the feedback circuit 76 and the signal processing circuit 51, the intensity signal vI +■2 for a point on the object is quantified as before by the nonlinear data converter 52 of the circuit 51, which includes an additional output 80 for data values outside the previous range. be converted into These digital values are buffered and passed through a delay line in a feedback circuit 76 to provide the data necessary to control the modulator 13 and to pass the modulated signal V, V2 to the digital scalers 60 and 62, Noise suppression circuits 64 and 6
6, and within the range of the divider 58.

Nonlinear data converters can be scaled to arbitrarily wide ranges within practical conditions of cost, circuit board size, speed requirements, and laser power range. As a result, the previous (three-dimensional plus, Daleyscale apparatus)
is the same as When put into practice, it is possible to obtain intensity change lines and depth information.

The above method and apparatus have a number of advantages.

For example, measurements can be performed at an imaging speed that provides high resolution and sufficient three-dimensional information. The method and apparatus also provide depth sensing at an accurate video frame rate at low cost.

In addition to the standard triangulation techniques shown in Figures 1 and 2, the detection method can also be applied to other three-dimensional imaging geometries. For example, the research literature shows that shadows and absorption can be completely prevented by a simple method using approximately coaxial illumination beams, focusing devices, and CCD detection arrays. The device's detector is unable to respond to low levels of light and is limited in speed because the mask is wide and transparent, the required optics are inadequate, and the detector is limited in speed. However, with some modifications,
By simply increasing the physical size of the optical container,
The present invention is capable of high-speed three-dimensional sensing even at low levels of light.

The preferred embodiments of the present invention have been described in detail above. However, as will be appreciated by those skilled in the art, the invention is not limited to the above embodiments, as set forth in the claims.

〔Effect of the invention〕

The present invention provides the following effects.

The method and apparatus according to the invention are relatively inexpensive, are capable of measuring objects in full three dimensions at video speeds, and are capable of measuring objects at low levels of light using high-speed circuits with a wide dynamic range. You can.

The measuring device can be integrated into inspection/calibration equipment since it can obtain both range and intensity data. Furthermore, in the case of line scanning, even stationary large objects can be inspected with high resolution.

The present invention can solve the problems of the prior art as follows.

(1) The spatially smooth configuration reduces reading errors in position sensitive filters, including programmable masks for spot/line conversion.

(2) Programmable masks allow multiple 11
Scattered light can be removed.

(3) A laser light source can be modulated by a feedback type high-speed signal processing device with a wide dynamic range.

(4) Noise shot function can be obtained by the detector.

(5) As a result of modulating the laser light source, the noise bandwidth can be narrowed.

Furthermore, since the feedback circuit first obtains the buffered intensity data of one row, the dynamic range can be widened, and the laser beams of the subsequent rows can be amplitude-modulated.

Spatial registration can be maintained by two precisely offset laser diodes.

TTL modulation allows separate signal processing operations to be performed for each on/off period.

[Brief explanation of the drawing]

Figure 1 is a schematic diagram showing the basic triangulation method, that is, the concept of ray construction, Figure 2 is a diagram showing two mathematical expressions showing the relationship between the variables in Figure 1, and Figure 3 is a diagram showing the basic triangulation method, that is, the concept of ray construction. FIG. 4 is a schematic diagram showing a set of optical components, a detection unit, and a signal processing circuit used in the method and device of the invention, FIG. 4 is a block diagram of signal processing showing the method and device of the invention, and FIG. 5 is the same as FIG. 6 is a more detailed schematic diagram of the signal processing circuit of FIG. 5, and FIG. 6 is a detailed block diagram of the noise suppression circuit of FIG. 10...High-speed three-dimensional measuring device, 12...
・Laser and condensing lens assembly, 18...
・Object to be measured, 20... Sample stand (base plane), 22...
・Scanner and beam forming/focusing optical system, 26...
... Focus lens, 28 ... Detection element, 33 ... Mask, 36 ... Variable filter, 39 ... Beam splitter (light beam splitter)
, 42.43... mirror, 54... total circuit (summing circuit), 58...
...Divider, 70.72...Sample and hold circuit. Engraving of the drawing (no changes to the content) ~,!

Claims (31)

[Claims] (1) scanning a surface of an object with a controlled light beam at a first predetermined angle to generate a reflected light signal; and receiving the reflected light signal at a second angle by a pair of optical components; spatially filtering and smoothing the received signal; optically splitting the received signal into a first beam and a reference beam; and dividing the radiant energy of the first beam and the reference beam. measuring each and forming first and second electrical signals proportional to the measured values; normalizing the first and second electrical signals and keeping them within a predetermined range; and the normalized electrical signals. A high-speed three-dimensional measurement method, comprising the step of calculating a centroid value of the first ray from a signal, three-dimensionally measuring the object on a sample stage at high speed and forming three-dimensional information related to the object. (2) scanning a surface of an object with a controlled light beam at a first predetermined angle to generate a reflected light signal; and receiving the reflected light signal at a second angle by a pair of optical components; spatially filtering and smoothing the received signal; optically splitting the received signal into a first beam and a second beam; and dividing a portion of the first beam into the second beam. measuring the reflected energy of the propagating portion of the first beam and the second beam, respectively, and forming first and second electrical signals proportional to the measured values; The first and second electrical signals are normalized to fall within a predetermined range, and the centroid value of the first ray is calculated from the normalized electrical signals. A high-speed three-dimensional measurement method for three-dimensionally measuring an object at high speed and forming three-dimensional information related to the object. (3) The normalization step includes the step of summing the first and second signals to form a normalized signal that falls within a predetermined range, and the centroid is calculated from the normalized sum. The method described in Scope 1 and 2. (4) The first and second electrical signals are in analog format, and the normalizing step includes the step of digitizing the first and second electrical signals and keeping them within a predetermined range. The method described in Section 2. (5) Claim 4, which includes a step of modulating the controlled light and changing the frequency of the controlled light to a predetermined frequency band, and a step of demodulating the first and second electrical signals. The method described in section. (6) The method according to claim 5, wherein the demodulation step includes a step of removing noise from the first and second electrical signals by a propagation band filter having the predetermined frequency band. 7. The method of claim 1, wherein the spatially filtering and smoothing step utilizes a programmable mask that correlates to the height of the object at the observation point. (8) The spatially filtering and smoothing step includes a step of forming a light spot from the received optical signal, and a step of converting the light spot into a single linear light. The method described in section. 9. The method of claim 8, wherein said spatially filtering and smoothing step utilizes a spatial filter of said pair of optical components. (10) The method according to Claims 1 and 2, including the step of focusing the first and second split light beams on the predetermined adjacent positions before the measurement step. (11) scanning a surface of an object with a controlled and modulated light beam at a first predetermined angle to generate a reflected light signal; and receiving the reflected light signal at a second angle by a pair of optical components. spatially filtering and smoothing the received signal by the pair of optical components; optically splitting the received signal into a first beam and a reference beam; and the first beam. and forming first and second electrical signals proportional to the measured values; and normalizing and scaling the first and second electrical signals to within a predetermined range. demodulating the scaled first and second electrical signals; calculating the centroid value of the first ray from the demodulated signals; A high-speed three-dimensional measurement method for three-dimensionally measuring an object at high speed and forming three-dimensional information related to the object. (12) (a) scanning a surface of an object with a controlled light beam at a first predetermined angle to generate a reflected light signal; and (b)
(c) spatially filtering and smoothing the received optical signal by the pair of optical components; (d) receiving the reflected signal at a second corner by a pair of optical components; optically splitting the received signal into a first beam and a reference beam; (e) measuring the radiant energy of the first beam and the reference beam, respectively, and applying first and second electric charges proportional to the measured values; a step of forming a signal; a step of generating a feedback signal using the pair of electrical signals; a step of using the feedback signal to control modulation of a light source of the controlled light; a process for use in steps (a) to (e); a process for normalizing and scaling the second pair of electrical signals to fit them within a predetermined range; A high-speed three-dimensional measurement method, comprising the step of calculating a centroid value of a first split beam of a signal, three-dimensionally measuring the object on a sample stage at high speed and forming three-dimensional information related to the object. (13) a light source that scans a surface of an object with a controlled light beam at a first predetermined angle to generate a reflected light signal; and a light source that receives the reflected light signal at a second angle and spatially transmits the received signal. a pair of optical components for filtering, smoothing, and optically splitting the received signal into a first beam and a reference beam; and measuring the radiant energy of the first beam and the reference beam, respectively, and converting the received signal into a measured value. first and second measuring means for forming proportional first and second electrical signals; and normalizing said first and second electrical signals to within a predetermined range and determining from said normalized electrical signals said A high-speed three-dimensional measuring device, comprising a signal processing means for calculating a centroid value of a first ray, and three-dimensionally measuring the object on the sample stage at high speed to form three-dimensional information related to the object. (14) a light source that scans a surface of an object with a controlled light beam at a first predetermined angle to generate a reflected light signal; and a light source that receives the reflected light signal at a second angle and spatially a pair of optical components for filtering, smoothing and optically splitting the received signal into a first beam and a second beam, and propagating a portion of the first beam at the second corner by a propagation means; and first and second measuring means for measuring the radiant energy of the propagation portion of the first beam and the second beam, respectively, and forming first and second electrical signals proportional to the measured values; a signal processing means for normalizing the first and second electrical signals, keeping them within a predetermined range, and calculating a centroid value of the first ray from the normalized electrical signals; A high-speed three-dimensional measurement device that measures three-dimensional objects at high speed and forms three-dimensional information related to the objects. (15) A patent in which the signal processing means includes a summation circuit that sums the first and second signals to form a normalized signal that falls within a predetermined range, and calculates the centroid from the normalized sum. Apparatus according to claims 13 and 14. (16) Claim 13, wherein the first and second electrical signals are in analog format, and the signal processing means includes generating means for digitizing the first and second electrical signals.
15. Apparatus according to paragraphs 1 and 14. (17) the signal processing means is coupled to the generation means;
17. The apparatus of claim 16, further comprising scaling means utilizing said digital signal to fit said first and second electrical signals within a predetermined range. 18. The apparatus of claim 17, wherein said scaling means includes a pair of programmable amplifiers for selectively amplifying said first and second electrical signals in response to said digital signal. (19) comprising modulation means for modulating the controlled light and changing the frequency of the controlled light to a predetermined frequency band;
19. The apparatus of claim 18, wherein said signal processing means includes a demodulator for demodulating the first and second digital signals. 20. The apparatus of claim 19, wherein said demodulator includes a filter for removing noise from said first and second electrical signal means and having a propagation band having said predetermined frequency band. (21) The pair of optical components includes means for spatially filtering and smoothing the received light signal, the pair of optical components having a programmable mask that correlates with the height of the object at the observation point. Apparatus according to clause 13. (22) The pair of optical components includes a diffuser that forms a light spot from the received light signal, and a light spot-to-light converter that converts the light spot into one linear light. 14. The apparatus of claim 13 including spatially filtering and smoothing means. 23. The apparatus of claim 22, wherein said spatially filtering and smoothing means comprises a spatial filter of said pair of optical components. (24) The apparatus according to Claims 13 and 14, wherein the pair of optical components includes an imaging means for imaging the first and second divided light beams at the predetermined adjacent positions. 25. The apparatus of claims 13 and 14, wherein the first and second measuring means include a single photodetector that converts the radiant energy into an electric current. (26) The device according to claim 25, wherein each of the photodetectors is a semiconductor device. (27) The device according to claims 13 and 14, wherein the pair of optical components includes splitting means for optically splitting the received light signal into first and second split beams. (28) The apparatus according to claims 13 and 14, wherein the light source is a laser scanner. (29) The apparatus of claim 28, wherein the laser scanner is a flying spot laser scanner. (30) a light source for scanning a surface of an object at a first predetermined angle with a controlled, modulated beam of light to generate a reflected optical signal; and means for spatially filtering and smoothing the received signal; a pair of optical components receiving the reflected optical signal at a second corner, the splitting means optically splitting the received signal into a first beam and a reference beam; and a pair of optical components receiving the reflected optical signal at a second corner; first and second measuring means for respectively measuring energy and forming first and second electrical signals proportional to the measured values; and scaling means for scaling the first and second electrical signals to fit within a predetermined range. and a demodulator that demodulates the first and second electrical signals to reduce noise, and calculates a centroid value of the first beam from the demodulated signals; A high-speed three-dimensional measuring device, comprising: a signal processing means for normalizing the object; (31) a light source including first and second light sources that continuously scan controlled and modulated first and second light beams at a first predetermined angle to generate a reflected optical signal; a pair of splitting means for optically splitting the received signal into a pair of first beams and a pair of reference beams; an optical component; first and second measurements for measuring the radiant energy of the pair of first beams and the pair of reference beams, respectively, and forming a pair of first and a pair of second electrical signals proportional to the measured values; means for generating a feedback signal with a first electrical signal of the pair to control modulation of a second light source, improve noise suppression, and widen the dynamic range; a scaling means for scaling the second electrical signal to fit within a predetermined range; and a demodulator for demodulating the pair of first and the pair of second electrical signals to reduce noise; signal processing means for calculating the centroid value of the pair of second electric signals with respect to the first signal and normalizing the first and second electric signals; A high-speed three-dimensional measurement device that measures and forms stereoscopic information related to said object.