JPH08136222A - Method and device for three-dimensional measurement - Google Patents

Abstract

PURPOSE: To measure the position of an object accurately accurately without receiving effects of the minuteness of the picture elements of an input image and the color of the surface of the object. CONSTITUTION: The specified pattern is formed in a work station 3 and moved. A liquid crystal projector 2 emits the light on the surface of an object 1 and projects the above described pattern. A CCD camera 5 picks up the image of the object, whose pattern is projected, at every specified time and forms a plurality of the images. These images are inputted into the work station 3. In the work station 3, the specified picture element on each image is noticed, and the brightness of the picture element is extracted. After the correlation value of the obtained time-series data and the above described pattern is obtained for every images, the comparison of the respective correlation values is performed, and the image, whose brightness becomes the maximum among the above described images, is specified. The position of the object is computed based on the time when the specified image is obtained and the position of the above described image.

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 method and apparatus for measuring the three-dimensional position of an object and recognizing the shape and the like. The present invention relates to a three-dimensional measurement method and apparatus for executing the above-described position measurement and recognition of a shape by processing an image obtained by imaging a specific projected pattern.

[0002]

2. Description of the Related Art A conventional three-dimensional measuring device, as shown in FIG. 17, includes a light projecting system 30 including a laser light source 31 and a reflecting mirror 32, and a light receiving system 33 including an image pickup device 34 and an image processing device 35. Composed. Reflecting mirror 32 of the light projecting system 30
And the image pickup device 34 of the light receiving system 33 are located at the same height and apart from each other by a predetermined distance L, and are respectively arranged toward the object 38.

The light projecting system 30 reflects the laser light 36 projected from the laser light source 31 in the horizontal direction by the reflecting mirror 32, projects the reflected light 37 toward the object 38, and projects the reflected light 37 toward the object 38. A bright point S ₀ is generated on the surface. The reflecting mirror 32
Is connected to a swinging mechanism (not shown). The swinging mechanism swings the reflecting mirror 32 in the horizontal direction to scan the reflected light 37 and move the bright spot S ₀ in the horizontal direction. Let The reflecting mirror 32 is provided with an angle detecting means such as an encoder so that the scanning angle θ (t ₀ ) of the reflected light 37 at a certain time t ₀ , that is, the projection direction with respect to the direction of the image pickup device 34 can be known. Has become.

In the light receiving system 33, the object 38 is picked up by the image pickup device 34, and the image is taken into the image processing device 35. In the figure, reference numeral 39 denotes an input image displayed on the monitor screen of the image processing device 35. In the image processing device 35, the image S ₀ ′ of the bright point S ₀ exists at which pixel position on the input image 39. Ask what to do. The image processing device 35 determines the coordinates (X) of the pixel position where the image S ₀ ′ of the bright point S ₀ exists.
₀ , Y ₀ ), the line-of-sight direction Φ (t ₀ ) of the bright point S ₀ viewed from the imaging device 34 is obtained, and the scanning angle θ (t ₀ ) and the line-of-sight direction Φ (t ₀ ) of the bright point S ₀ are obtained. , And the distance L between the reflecting mirror 32 and the image pickup device 34, the bright point S ₀ is calculated according to the principle of triangulation.
The coordinates (X, Y, Z) of the spatial position of are calculated.

[0005]

According to the conventional three-dimensional measuring method, the image S ₀ ′ of the bright spot S ₀ in the input image 39 is obtained.
Depending on whether the brightness is maximized at any pixel location of the coordinates (X ₀ of the pixel position where the image S ₀ of the bright point S ₀ 'is present,
Y ₀ ) is obtained, but in such a method, the measurement accuracy depends on the fineness of the pixel, and therefore the image S ₀ ′ of the bright spot exists with an accuracy higher than the fineness of the pixel. Since the pixel position cannot be obtained, it is difficult to measure the position of the object with high accuracy.

Further, when a plurality of color areas are present on the surface of the object 38, when the bright spot S ₀ is generated at the boundary of the color areas or the like, the reflectance of each color area is different, so that the maximum brightness is obtained. There is a possibility that the pixel position at which the pixel position is shifted from the center position of the image S ₀ ′ of the bright spot. Therefore, the pixel position where the image S ₀ ′ of the bright point S ₀ exists cannot be accurately obtained, and it is difficult to accurately measure the position of the object.

The present invention has been made in view of the above problems, and the position of an object can be accurately and accurately measured without being affected by the fineness of pixels of an input image or the color of the surface of the object. An object of the present invention is to provide a three-dimensional measuring method and an apparatus thereof.

[0008]

To achieve the above object, the present invention projects a specific pattern by projecting light onto the surface of an object and moving the projected pattern in a predetermined direction. Time-series data obtained by capturing a target object at predetermined time intervals to generate a plurality of images, focusing on a specific pixel on each image, and extracting the brightness of the pixel for each image, and After obtaining the correlation value with the pattern for each image, the correlation value is compared to identify the image with the maximum brightness of the pixel, and the time when the identified image was obtained and the pixel position. The position of the object is calculated from this.

The invention according to claims 2 to 4 is a three-dimensional measuring method for increasing the measurement accuracy, and in the invention according to claim 2, the pattern is a stripe pattern in which bright and dark repeat a predetermined code having a strong autocorrelation. It is a coded pattern. Further, in the invention of claim 3, as the pattern, M having a strong autocorrelation is used.
The sequence code is used, and in the invention of claim 4, a chirp signal having a strong autocorrelation is used.

According to a fifth aspect of the present invention, there is provided a three-dimensional measuring method for further increasing the measuring accuracy, wherein the pattern is modulated and projected into a pattern having a peak value at the boundary of the striped pattern.

According to a sixth aspect of the present invention, there is provided a three-dimensional measurement method for increasing the measurement accuracy in each of the vertical and horizontal directions, wherein the pattern is a first pattern in which a predetermined code is encoded in a vertical striped pattern in which light and dark are repeated. And a second pattern in which a predetermined code is encoded in a horizontal striped pattern in which light and dark are repeated.

In the invention of claim 7, the first and second patterns have vertical stripe patterns and horizontal stripe patterns with the same color, and in the invention of claim 8, the measurement accuracy in each of the vertical and horizontal directions is improved. In addition, in order to speed up the measurement, the first and second patterns have vertical stripe patterns and horizontal stripe patterns in different colors.

According to a ninth aspect of the present invention, there is provided a three-dimensional measuring apparatus for carrying out the above-mentioned three-dimensional measuring method, which comprises a light projecting means for projecting a specific pattern by projecting light onto the surface of an object. Pattern generating means for generating a pattern to be projected by the means and moving the pattern in a predetermined direction; image capturing means for capturing a plurality of images by capturing an image of the object onto which the pattern is projected at predetermined intervals; Focusing on the storage means for sequentially storing each image obtained by the image pickup means and the specific pixel on each image stored in the storage means, the brightness of the pixel is extracted for each image. An arithmetic unit that obtains the correlation value between the time-series data and the pattern for each image is compared with each correlation value obtained by the arithmetic unit to identify the image in which the brightness of the pixel is maximum, and to identify the image. When the captured image was obtained It is obtained by a position calculating means for calculating the position of the object from said pixel position with.

According to a tenth aspect of the present invention, there is provided a three-dimensional measuring apparatus having improved measurement accuracy, wherein the arithmetic means temporally divides each image and interpolates the brightness of the pixel in each divided image. Then, the correlation value between the time series data including the brightness obtained by the interpolation and the pattern is obtained for each image including the divided images.

[0015]

In the three-dimensional measuring method of the present invention, the position of the object is calculated by focusing on a specific pixel and specifying the image having the maximum brightness among the plurality of images. The position of the object can be measured with high accuracy regardless of the fineness of the pixels. Even if there is a color area on the surface of the object, since the focus is on a specific pixel, the condition does not change, and the position of the object is not affected by the color of the surface of the object. Accurately measured.

In the three-dimensional measuring apparatus of the present invention, when the pattern generating means generates a specific pattern and moves it in a predetermined direction, the pattern projected onto the surface of the object by the light projecting means moves in the same direction. Moving. The object on which the pattern is projected is imaged by the imaging unit at predetermined time intervals, and a plurality of images are generated. A particular pixel is focused on each image, and the time series data obtained by extracting the brightness of the pixel for each image is stored in the storage means. The calculating means obtains a correlation value between the time series data and the pattern for each image. In this case, in the three-dimensional measuring apparatus according to claim 10, the calculation unit temporally divides each image, interpolates the brightness of the pixel in each divided image, and then obtains the brightness obtained by the interpolation. The correlation value between the time-series data including the height and the pattern is obtained for each image including the divided images.
The position measuring means compares the respective correlation values to specify the image in which the brightness of the pixel is maximum, and then determines the object from the time when the specified image was obtained and the specific pixel position of interest. Calculate the position of.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows a schematic configuration of a three-dimensional measuring apparatus which is an embodiment of the present invention. The three-dimensional measuring device in the illustrated example is
A liquid crystal projector 2 that projects a specific pattern by projecting light onto the surface of the object 1, a workstation 3 that generates the pattern by an image and moves it in the horizontal direction, and a pattern that is generated by the workstation 3 on a video. A network video server 4 that converts the signal into a signal and outputs the signal to the liquid crystal projector 2 and a CCD camera 5 that captures an image of the object 1 onto which the pattern is projected at predetermined time intervals to generate a plurality of images are provided. A plurality of input images obtained by the CCD camera 5 are input to the workstation 3 via the video interface 6.

In the following description, the reference pattern generated by the workstation 3 is the "reference pattern" (shown in FIG. 2), and the pattern obtained by moving the reference pattern is the "movement pattern" (see FIG. 3 and FIG. 3). The pattern projected on the surface of the object 1 is referred to as a "projection pattern" for distinction. The liquid crystal projector 2 and the CCD camera 5 are located at the same height and apart from each other by a predetermined distance L, and are arranged toward the object 1.

The workstation 3 obtains time-series data on a pixel of interest, which will be described later, from a plurality of input images, performs correlation calculation with the reference pattern for each input image, and then calculates the position of the object. Performs processing for recognition of shapes and the like.

In this embodiment, the work station 3 and the network video server 4 and the work station 3 and the video interface 6 are connected via the Ethernet 7, but the present invention is not limited to this. There is no. A personal computer may be used instead of the workstation 3.

FIG. 2 shows an example of the reference pattern 10 generated by the workstation 3. The reference pattern 10 in the illustrated example is obtained by encoding a predetermined code into a striped pattern in which light and dark repeat in the horizontal direction, and black vertical stripes 11a to 11 are formed.
e and white vertical stripes 12 are alternately arranged. Each vertical stripe 1
The widths of 1a to 11e and 12 are set to correspond to the code strings.

FIG. 3 shows a specific example of the moving pattern 10 'obtained by moving the reference pattern 10 in the horizontal direction (from right to left). 3 (1) shows the moving pattern 10 'before the vertical stripes 11a to 11e of the reference pattern 10 appear, and FIG. 2 (2) shows the reference pattern 10'.
2 (3) shows a movement pattern 10 'in which the first vertical stripe 11a of the reference pattern 10 appears at the right end, and FIG. 2 (3) shows a movement pattern 1 in which all the vertical stripes 11a to 11e of the reference pattern 10 appear.
2'is the last vertical stripe 1 of the reference pattern 10.
Move pattern 10 'with only 1e left at the left end,
FIG. 2 (5) shows all the vertical stripes 10a to 1 of the reference pattern 10.
Movement patterns 10 'after the disappearance of 0e are respectively shown.

FIG. 4 shows vertical stripes 10a to 10a of the reference pattern 10.
The specific example of the movement pattern 10 'which makes 10e go around is shown. FIG. 4A shows the moving pattern 10 'in a state where all the vertical stripes 11a to 11e of the reference pattern 10 have appeared. In the moving pattern 10 ', the positions of the vertical patterns 11a to 11e are completely the same as those of the reference pattern 10. FIG. 4 (2) shows the movement pattern 10 'in which the reference pattern 10 is slightly moved to the left, and FIG. 4 (3) appears at the right end after the leading vertical stripe 11a of the reference pattern 10 disappears to the left end. The movement patterns 10 'of the state are shown respectively. FIG. 4 (4) shows a movement pattern 10 'when all the vertical stripes 11a to 11e of the reference pattern 10 make one round.
Indicates. This movement pattern 10 'is a state in which the reference pattern 10 is slightly displaced to the right. FIG.
(5) shows the movement pattern 10 'when the vertical stripes 11a to 11e of the reference pattern 10 make a complete cycle. The moving pattern 10 'includes a reference pattern 10 and vertical stripes 11a-1.
The positions of 1e are completely the same.

As the above-mentioned reference pattern 10, it is desirable to use a code having a strong autocorrelation coded in a striped pattern in which light and dark are repeated. In FIG. 5, the reference pattern 1 is formed by encoding the M-sequence code into a striped pattern in which light and dark repeat in the horizontal direction.
6 and FIG. 6 show the autocorrelation characteristics (autocorrelation value with respect to the number of pixel shifts) of the reference pattern 10 using this M-sequence code.

By using the reference pattern 10 having such a sharp autocorrelation characteristic, highly accurate position measurement becomes possible. If the code has a strong autocorrelation, not only the M-sequence code but also, for example, a chirp signal can be used.

FIG. 7 shows a specific example of the reference pattern 10 using the chirp signal, and FIG. 8 shows the autocorrelation characteristic of the reference pattern 10 using the chirp signal. This chirp signal is a signal in which the frequency of a sine wave is continuously changed, and according to this signal, a reference pattern having different density values for each column is generated.

FIG. 9 is a diagram for explaining the principle of the three-dimensional measuring method of the present invention. In the figure, M is a light source for projecting spot light, and the projection light 13 from this light source M produces a bright spot 15 on the surface of the object 14. The projection light 13 is shaken in the horizontal direction, whereby the bright spots 15 sequentially move to respective positions S _a , S _b , and S _{c in} the figure.

In the figure, C is a CCD camera, and CC
A plurality of input images 16 are obtained by imaging the target object 14 with the D camera C at every predetermined time. In the illustrated input image 16, pixels represented by rectangular cells mean pixels. When focusing on an arbitrary pixel X, the pixel X of each input image 16 obtained at a predetermined time
Fig. 1 shows the brightness distribution obtained by observing the brightness of
0 is shown.

In FIGS. 9 and 10, t _a , t
_b and t _c indicate the number of input images representing the number of input images of the input image 16 obtained when the bright point 15 is at each position of S _a , S _b and S _c . . Also θ (t _a ),
θ (t _b ) and θ (t _c ) are the directions of the positions S _a , S _b , and S _c with respect to the direction of the CCD camera C viewed from the light source M, that is, the scanning angles of the projection light 13.

Since the line-of-sight direction φ of the camera C with respect to the noted pixel X is fixed, when the bright spot 15 is at each position of S _a and S _c , the bright spot 15 cannot be observed through the pixel X, but When the point 15 is located at S _b , the bright point 15
Can be observed through pixel X. Therefore, in the input image 16 when the number of input images is t _b , the brightness of the pixel X becomes maximum (FIG. 10). If the number of input images t _b is known, the coordinates of the spatial position S _b can be obtained from the scanning angle θ (t _b ), the line-of-sight direction φ, and the distance L between the light source M and the camera C. In FIG. 9, in order to facilitate the explanation,
Although an example in which the projection light 13 from the light source M is spot light is shown, a vertically long slit light may be used instead of the spot light in order to speed up the processing.

FIG. 11 is an explanatory view of the principle of the present invention when three slit lights are used as the light projected onto the object. When three slit lights are projected onto the surface of the object and moved horizontally from left to right,
Three bright lines are generated by each slit light and move in the same direction. In FIG. 11 (1), the horizontal axis indicates the pixel position X in the horizontal direction of the input image, and the vertical axis indicates the number t of input images, that is, the time when the input image is obtained. The image areas 17, 18 and 19 of the respective bright lines shown in the figure are displaced to the right with the passage of time. This object is
Since the inclination of the surface facing the light source changes midway, the inclination of the image areas 17, 18, 19 of each bright line, that is, the moving speed also changes midway.

FIG. 11B shows the brightness distribution when the number of input images t is t _A , and FIG. 11C shows the brightness distribution when the number of input images t is t _B. Note that the horizontal axis is the pixel position X. As is clear from FIGS. 11 (2) and 11 (3), the respective brightness distributions are different from each other, and each brightness distribution is also different from the brightness distribution of the reference pattern shown in FIG. 11 (5). This is because the inclination of the surface facing the light source changes midway. FIG. 11 (4) shows
The change in brightness when the brightness is obtained in time series for the pixel at a certain pixel position X ₀ is shown. The time-series data of the brightness obtained for each input image for the pixel at the pixel position X ₀ obtains a high correlation with the brightness distribution of the reference pattern in FIG. 11 (5) at any time. As shown in (6), when the number of input images t is t ₀ , the correlation value between the two becomes maximum.

FIG. 12 shows a flow of the operation of the three-dimensional measuring apparatus based on the above principle in step 1 (indicated by "ST1" in the figure).
~ As shown in step 7. First, in step 1, the workstation 3 generates an image of a moving pattern to be projected on the object 1 based on the reference pattern, and the image is given to the liquid crystal projector 2 via the network video server 4. In the next step 2, when the liquid crystal projector 2 projects the movement pattern on the surface of the object 1, CC
The D camera 5 images the object 1 on which the movement pattern is projected to generate a first input image, and the input image is taken into the workstation 3 and stored in the image memory (steps 3 and 4).

In the next step 5, it is judged whether or not the scanning of the object 1 by the moving pattern is completed, that is, whether or not all the moving patterns are projected on the surface of the object 1, and the judgment is made. If “NO”, the process proceeds to step 6, the workstation 3 generates an image of the movement pattern moved by one pixel, returns to step 2, and the above-mentioned steps 2 to 4 are sequentially executed. The amount of movement of the movement pattern is not limited to one pixel and may be moved every several pixels.

When the same procedure is repeatedly executed and the judgment in step 5 is "YES", step 7
Then, the process for measuring the position of the object 1 is executed in the workstation 3. In the position measurement process of step 7, attention is paid to a specific pixel in all the input images stored in the image memory, and the brightness of the pixel is extracted for each input image to generate time series data. Then, the correlation value between the time-series data and the reference pattern is calculated for each input image by an arithmetic expression described below, and the input image having the maximum correlation value is specified as the input image having the maximum pixel brightness. After that, the coordinates of the spatial position of the object are calculated from the time when the specified input image was obtained and the noted pixel position. It is possible to measure the position and recognize the shape of each point of the object 1 by repeatedly performing the same procedure while sequentially changing the pixel of interest.

In this embodiment, in order to perform highly accurate measurement, each input image is temporally equally divided in the above-mentioned correlation calculation to set a virtual input image, and these virtual input images are set. With respect to the above, the brightness of the pixel is interpolated, and the correlation value between the time-series data including the brightness obtained by the interpolation and the reference pattern is obtained for all input images including the virtual input image. ing.

The number of input images (hereinafter referred to as “actual images”) actually obtained is f, and the number of input images (hereinafter referred to as “virtual image”) virtually set between the actual images, that is, division. The brightness of the pixel of interest x in the t-th actual image is B (t), the brightness of the t-th pixel in the reference pattern is M (t), and the brightness of the pixel of interest x in the f actual images is The average value of the heights is defined as AV. Note that B (t) is 0 ≦ B (t) ≦ 255 at the time of 256 gradations, and M (t)
Is "1" or "-1" in the case of the reference pattern using the M-sequence code.

Under this definition, of the f × b images including the virtual image, the brightness B ′ (t ′) of the target pixel x in the t′-th image and the t′-th image in the reference pattern. The brightness M '(t') is calculated by interpolation calculation. For example, in the case of the t'th virtual image set between the t'th and t + 1'th real images of the f real images, the brightness B '(t') of the pixel of interest x in that image is bright. B
(T) and B (t + 1), and the t'th brightness M '(t') in the reference pattern is the brightness M.
It is calculated using (t) and M (t + 1), respectively.

When the brightness is calculated by the above complementary calculation process, the correlation calculation is executed for each of the sth image (where 0 ≦ s ≦ f × b) by the following equation.

[0040]

[Equation 1]

The time-series data of the brightness at the time when the sth image for which the maximum correlation value is obtained by the above equation is obtained is the most similar to the reference pattern, and the s / bth image is It is specified as an image having the maximum brightness in the pixel of interest x. In the correlation calculation of the above equation, in order to improve the accuracy of the calculation result by using a code having a small DC component,
After the average value AV is removed as an offset for the brightness B ′ (t ′) of the pixel of interest x in each image, the obtained value is converted into a numerical value from −1 to 1 to perform the calculation.

FIG. 13 shows specific pixels in a plurality of input images obtained by projecting a movement pattern generated by a reference pattern using the M-sequence code of FIG.
The time-series data obtained by extracting the brightness of the pixel for each input image is shown. In this specific example, the peak values of the bright and dark portions are varied due to the influence of distortion of the optical system. In this case, if the offset of the average value is removed in the correlation calculation, the lightness / darkness width of the time-series pattern of the obtained brightness will not accurately reflect the lightness / darkness width in the reference pattern, and the measurement accuracy will deteriorate.

Therefore, as shown in FIG. 14, by projecting onto a target object a pattern (mark-to-mark modulation) that is modulated into a pattern having a peak value when the code (1 or -1) forming the reference pattern is switched. It is possible to improve the measurement accuracy. 14 (1) shows a change in brightness of the reference pattern, and FIG. 14 (2) shows a change in brightness of the time-series pattern when the reference pattern in FIG. 14 (1) is used. FIG. 14 (3) shows a change in the brightness of the pattern in which peak values are given to the rising part and the falling part of the brightness in FIG. 14 (1), and FIG.
Changes in the brightness of the time-series pattern when the pattern (3) is used are shown. As is clear from the illustrated examples, when using a reference pattern based on a digital signal such as an M-sequence code, an accurate measurement result can be obtained by using a pattern subjected to inter-mark modulation as a reference pattern.

On the other hand, in the case of the chirp signal described above, FIG.
As shown in Fig. 2, since a reference pattern that gradually changes for each column is generated, time-series data that accurately reflects the reference pattern is obtained without depending on the characteristics of the optical system, and an optical system with low resolution can be obtained. Even when used, highly accurate measurement results can be obtained.

FIG. 15 shows a first pattern in which the reference pattern is encoded as a vertical striped pattern in which a predetermined code repeats light and dark in the horizontal direction.
10A and a second pattern 10B in which a predetermined code is encoded in a horizontal striped pattern in which light and dark repeat in the vertical direction. In the case where the first pattern 10A and the second pattern 10B have the same color and show a vertical striped pattern and a horizontal striped pattern, first, the first pattern 10A is projected onto an object to complete scanning, and then the second pattern The pattern 10B is projected and scanned on the object. By using the reference pattern composed of such first and second patterns 10A and 10B, highly accurate measurement can be performed in each of the horizontal and vertical directions.

When the first pattern 10A and the second pattern 10B have vertical stripes and horizontal stripes in different colors, the first and second patterns 10A and 10B are used.
Are projected onto the object at the same time and scanned in the horizontal and vertical directions. By using the reference pattern composed of the first and second patterns 10A and 10B having different colors as described above, it is possible to perform high-accuracy and high-speed measurement in each of the horizontal and vertical directions.

FIG. 16 shows the configuration of the three-dimensional measuring apparatus when the above-mentioned first and second patterns 10A and 10B are used. The first and second patterns 10A and 10B are projected onto the object 1. A first CCD camera 5A for picking up an image of the object 1 onto which the first pattern 10A is projected, and a second CCD camera 5A in a direction orthogonal to each other.
Second CCD cameras 5B, 5B for picking up the object 1 on which the pattern 10B of FIG.

[0048]

As described above, according to the present invention, the position of the object is calculated by focusing on a specific pixel and identifying the image having the maximum brightness among the plurality of images. The position of the object can be measured with high accuracy without being affected by the fineness of the pixels of the input image. Even if there is a color area on the surface of the target object, since we focus on specific pixels,
The condition does not change, and the position of the object can be accurately measured without being affected by the color of the surface of the object.

According to the second aspect of the present invention, since a predetermined code having a strong autocorrelation is coded into a striped pattern in which light and dark are repeated, the measurement accuracy is improved.

According to the third aspect of the invention, since the pattern in which the M-sequence code having a strong autocorrelation is encoded is used, the measurement accuracy can be further improved.

According to the fourth aspect of the invention, since a pattern in which a chirp signal having a strong autocorrelation is encoded is used, a highly accurate measurement result can be obtained even when an optical system having a low resolution is used.

According to the fifth aspect of the present invention, the pattern is modulated into a pattern having a peak value at the boundary of the striped pattern and projected onto the object. Therefore, accurate measurement is possible even when a digital coded pattern is used. The result can be obtained.

According to the sixth aspect of the invention, as the pattern, the first pattern encoded in the vertical striped pattern and the second pattern encoded in the horizontal striped pattern are used. Is increased.

According to the eighth aspect of the invention, since the vertical striped pattern and the horizontal striped pattern of each of the first and second patterns are represented by different colors, the measurement accuracy in the vertical and horizontal directions can be improved and the measurement speed can be increased. Get rid of.

According to a ninth aspect of the present invention, a three-dimensional measuring device is constituted by the light projecting means, the pattern generating means, the imaging means, the storing means, the calculating means, and the position calculating means, and a specific pixel of interest among a plurality of images Since the position of the target object is calculated by specifying the image with the maximum brightness of, it does not depend on the fineness of the pixels of the input image and is not affected by the color of the surface of the target object. , The position of the object can be measured with high precision and accuracy.

According to the tenth aspect of the present invention, each image is temporally divided to interpolate the brightness of the pixel, and then the correlation between the time series data including the brightness obtained by the interpolation and the pattern. Since the value is obtained for each image including the divided image, there is an effect that the position of the object can be measured with high accuracy.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a three-dimensional measuring apparatus that is an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of a reference pattern.

FIG. 3 is an explanatory diagram showing an example of a movement pattern.

FIG. 4 is an explanatory diagram showing another example of a movement pattern.

FIG. 5 is an explanatory diagram showing a specific example of a reference pattern using an M-sequence code.

FIG. 6 is a characteristic diagram showing an autocorrelation characteristic of a reference pattern using an M-sequence code.

FIG. 7 is an explanatory diagram showing a specific example of a reference pattern using a chirp signal.

FIG. 8 is a characteristic diagram showing an autocorrelation characteristic of a reference pattern using a chirp signal.

FIG. 9 is a principle explanatory view showing the principle of the three-dimensional measuring method of the present invention.

FIG. 10 is an explanatory diagram showing a brightness distribution of time-series brightness data.

FIG. 11 is a principle explanatory view for explaining the principle of the present invention.

FIG. 12 is a flowchart showing a flow of operations of the three-dimensional measuring apparatus.

FIG. 13 is an explanatory diagram showing a specific example of brightness time-series data.

FIG. 14 is an explanatory diagram showing a specific example of inter-mark modulation.

FIG. 15 is an explanatory diagram showing another example of the reference pattern.

16 is an explanatory diagram showing an overall configuration of a three-dimensional measuring apparatus using the reference pattern of FIG.

FIG. 17 is a principle explanatory diagram showing a configuration of a conventional three-dimensional measuring apparatus.

[Explanation of symbols]

 1 Object 2 Liquid Crystal Projector 3 Workstation 5 CCD Camera 10 Reference Pattern 10 'Movement Pattern

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G06T 7/00

Claims (10)

[Claims] 1. An object is projected at a predetermined time while projecting a specific pattern by projecting light onto the surface of the object and moving the projected pattern in a predetermined direction, and a plurality of images are displayed. After generating and focusing on a specific pixel on each image, the brightness of the pixel is extracted for each image and the correlation value between the time-series data and the pattern is obtained for each image. A third order characterized by specifying an image in which the brightness of the pixel is maximum by comparing the correlation values, and calculating the position of the object from the time when the specified image is obtained and the pixel position. Original measurement method. 2. The three-dimensional measuring method according to claim 1, wherein the pattern is obtained by encoding a predetermined code having a strong autocorrelation into a striped pattern in which light and dark are repeated. 3. The pattern is an M-sequence code encoded in a striped pattern in which light and dark are repeated.
3D measurement method described in. 4. The three-dimensional measuring method according to claim 1, wherein the pattern is a chirp signal encoded into a striped pattern in which light and dark are repeated. 5. The pattern is modulated and projected into a pattern having a peak value at a boundary of a striped pattern.
The three-dimensional measurement method described in any one of. 6. The pattern comprises a first pattern in which a predetermined code is encoded in a vertical striped pattern in which light and dark are repeated, and a second pattern in which a predetermined code is encoded in a horizontal striped pattern in which light and dark are repeated. The three-dimensional measuring method according to claim 1. 7. The three-dimensional measuring method according to claim 1, wherein each of the first and second patterns has a vertical stripe pattern and a horizontal stripe pattern with the same color. 8. The vertical stripe pattern and the horizontal stripe pattern are expressed in different colors in each of the first and second patterns.
Or the three-dimensional measuring method described in 6. 9. A light projecting means for projecting a specific pattern by projecting light onto the surface of an object, and a pattern generating means for generating a pattern to be projected by the light projecting means and moving it in a predetermined direction. An image capturing unit that captures an image of the object onto which the pattern is projected at predetermined time intervals to generate a plurality of images, a storage unit that sequentially stores each image obtained by the image capturing unit, and a storage unit that stores the image. Focusing on a specific pixel on each image, a calculation unit that calculates, for each image, a correlation value between the time-series data obtained by extracting the brightness of the pixel for each image and the pattern; The position at which the position of the object is calculated from the time when the specified image was obtained and the pixel position while specifying the image in which the brightness of the pixel is maximum by comparing the correlation values obtained in Comprising calculation means Dimension measuring device. 10. The computing means temporally divides each image, interpolates the brightness of the pixel in each divided image, and time-series data including the brightness obtained by the interpolation and the time series data. The three-dimensional measuring apparatus according to claim 9, wherein the correlation value with the pattern is obtained for each image including the divided images.