JP2010091517A - Position measuring device - Google Patents

Abstract

A position measurement apparatus that uses a landmark that is not affected by an imaging distance and that does not change its shape in an image and that performs position detection processing that is not easily affected by ambient brightness.
A plurality of landmarks 9 having a radial pattern imaged in a similar shape regardless of an imaging distance and attached separately to a measurement object whose position is to be measured are the same as the plurality of landmarks 9 A camera 8 installed and fixed so as to capture an image, and the positions of the plurality of landmarks 9 are detected from the captured image of the camera 8, and the positions of the plurality of landmarks 9 in the image are used to detect the measurement object. And an image processing unit 7 that obtains the position of the spreader 5 that is an object by calculation.
[Selection] Figure 1

Description

  The present invention relates to a position measuring device that measures the position of a spreader provided in a crane, for example.

  When a container is transported using a transfer crane or the like, the position of a spreader that holds the container is measured to detect the lifting height or shaking of the container. As a technique for measuring the position of the spreader, for example, Non-Patent Document 1 discloses a position in which laser light is emitted from a laser light source installed on a spreader to two CCD cameras installed at two locations on a trolley. A measuring device is described. This apparatus measures the position of the spreader by obtaining the center position of the laser beam picked up by each CCD camera using a gravity center position detection circuit.

  There is also a position measuring device in which a landmark is attached to the upper surface of a spreader of a transfer crane, the landmark is imaged by a camera provided in the trolley, and the position of the spreader is obtained from the captured image. This position measuring device includes, for example, a single camera 108 at the bottom of a trolley 104 as shown in FIG. 13A, and a landmark 109 with a plurality of concentric circles as shown in FIG. 13B. Is attached to the upper surface of the spreader 105. Such a landmark 109 is imaged using the camera 108. The image data output from the camera 108 is processed using a personal computer or the like, for example, and the center position of the landmark 109 in the image is detected to measure the position of the spreader 105, that is, the shake or height of the container 102. .

Mitsubishi Heavy Industries Technical Review, Vol. 5 (2000-11)

However, when the spreader is provided with a laser light source, there is a possibility that the driver or the like may be irradiated with laser light when the operator or the like gets on the cockpit provided in the trolley.
In addition, when the landmarks of concentric patterns provided on the spreader are imaged, when the lifting height of the spreader is changed, one of the concentric circles becomes the same size and appears in the image at a predetermined height. Include. However, when the spreader is not positioned at the above-mentioned predetermined height, the concentric circles appearing in the image have various sizes, causing a problem in the image processing, and the detection result of the spreader position is largely deviated. There was a problem that could occur.

  The present invention has been made in view of such circumstances, and an object thereof is to use landmarks that are not affected by the imaging distance and in which the shape in the image does not change and are not easily affected by ambient brightness. The object is to obtain a position measuring device that performs position detection processing.

  The position measuring device according to the first aspect of the present invention has a pattern that is imaged in a similar manner regardless of the imaging distance, and a plurality of landmarks that are attached separately to a measurement object for measuring the position; A camera that is installed and fixed so as to capture the plurality of landmarks in the same image, and a position of the plurality of landmarks is detected from a captured image of the camera, and the positions of the plurality of landmarks in the image are used. An image processing unit that obtains the position of the measurement object by calculation.

  Preferably, the landmark has a radial pattern in which a plurality of colors having different densities are alternately arranged.

  Preferably, the image processing unit obtains the position of the landmark by detecting an edge of the landmark pattern.

  Preferably, the image processing unit obtains a correlation between a template set in advance using a vector code correlation method and a landmark in the image, and detects the position of the landmark.

  Preferably, the template corresponds to a central portion of a radial pattern of the landmark.

  According to the present invention, it is possible to accurately measure the position of a measurement object while suppressing the influence of ambient brightness.

An embodiment of the present invention will be described below.
<First Embodiment 1>
FIG. 1 is an explanatory diagram showing the configuration of the position measuring apparatus according to the first embodiment of the present invention.
FIG. 1A shows a schematic configuration of, for example, a transfer crane 1 including the position measuring device according to the first embodiment, and shows a state in which the transfer crane 1 lifts the container 2.

The transfer crane 1 includes a horizontally disposed guide 3, a trolley 4 supported by the guide 3, and a spreader 5 that holds the container 2.
The trolley 4 is configured to move along the guide 3. For example, the spreader 5 is lifted by using four wire ropes 6. Each wire rope 6 is provided so that it can be wound up by a winch or the like, and the trolley 4 is configured to freely adjust the ground height of the spreader 5, that is, the length of the wire rope 6.

On the upper surface of the spreader 5, a landmark 9 to be imaged by the image processing unit 7 is attached. In the spreader 5 illustrated in FIG. 1A, two landmarks 9 are separated from each other along the longitudinal direction of the spreader 5.
The image processing unit 7 is provided at an appropriate location of the transfer crane 1, and is disposed, for example, at a driver's seat of the trolley 4.
The camera 8 is installed and fixed to the lower portion of the trolley 4 so that two landmarks 9 attached to the upper surface of the spreader 5 can be imaged.

  The landmark 9 has a pattern that is imaged in a similar shape regardless of the imaging distance. For example, the landmark 9 has a radial pattern as shown in FIG. This radial pattern is configured such that, for example, a pattern in which white and black are alternately arranged spreads radially from the center point of the landmark 9. Specifically, the pattern is configured so that the boundary between the two colors is 16 places. The landmark 9 preferably has a radial pattern divided into 16 to 32.

  In addition, said radial pattern is not limited to white and black, You may comprise by two colors from which density differs. Further, the radial pattern may be configured by combining two or more colors. That is, the radial pattern is configured by combining colors that make the edges of the pattern clear in the image captured by the camera 8. 1B illustrates the landmark 9 having a quadrilateral outer peripheral shape, the shape of the edge portion of the landmark 9 may be formed in any manner.

FIG. 2 is a block diagram showing the configuration of the position measuring apparatus according to the first embodiment. This figure mainly shows the configuration of the image processing unit 7 shown in FIG.
The position measuring device includes a plurality of landmarks 9 attached to the spreader 5 as described above, for example, a camera 8 such as a CCD camera having an industrial lens, and an image processing unit that processes image data output from the camera 8. 7.
The image processing unit 7 includes image input processing means 11 that converts image data output from the camera 8 into a predetermined data format, specifically, data that can be used by an arithmetic processing unit 12 described later. Yes.

Further, the image processing unit 7 obtains the position of the landmark 9 imaged by the camera 8 from the output data of the image input processing means 11 by calculation, and the height and shaking of the spreader 5 based on the position of the landmark 9. Is provided with arithmetic processing means 12 for calculating
Further, the image processing unit 7 includes an output processing unit 13 that generates a predetermined output signal according to a calculation result of the calculation processing unit 12, and a storage unit 14 that stores a template and the like to be described later.
The output processing means 13 is connected and configured to output position information or the like, which is the aforementioned calculation result, to, for example, a control mechanism unit (not shown) that prevents the spreader 5 from shaking. The control mechanism unit is mounted on the spreader 5 and is suitable for suppressing the spreader 5 swinging in accordance with the swing of the spreader 5 and the length of the wire rope 6 obtained from the position information. It has a function to generate inertial force.
Further, the output processing means 13 may be connected and configured to display the above-described position information and the like on a display device provided in a cockpit (not shown) of the transfer crane 1.

  The image processing unit 7 may be configured using, for example, a personal computer. Specifically, each unit of the personal computer or the like is set / configured to include each means having the processing functions described above using a device such as a processor and a software program installed in the personal computer or the like.

  The transfer crane 1 of FIG. 1 suspends the spreader 5 with the length of the wire rope 6 being about 20 [m], for example. When measuring the position of the spreader 5 suspended by the wire rope 6 having such a length, for example, the camera 8 and the landmark 9 shown in the following requirement (i) or requirement (ii) are used. In addition, the landmark 9 illustrated here has an outer periphery formed in a quadrilateral shape.

(I) The camera 8 uses a CCD element having an image size of 640 × 480 [pixel], for example, 1/3 size, a pixel size of 7.4 [μm], and a focal length of 75 [mm]. ] Imaging performance.
The landmark 9 used at this time has a size of four sides of 70 [mm] or more.

(Ii) The camera 8 uses a CCD element having an image size of 1024 × 1024 [pixel], for example, a 1/2 size, a pixel size of 5.5 [μm], and a focal length of 50 [mm]. ] Imaging performance.
The landmark 9 used at this time has a square size of 80 [mm] or more.
When the camera 8 and the landmark 9 exemplified in (ii) are used, the resolution of the image is higher than that of (i), and the measurement range of the position of the landmark 9 or the spreader 5 is widened. Is possible.

Next, the operation will be described.
[Operation to measure the position of the spreader]
3 and 4 are explanatory diagrams illustrating a measurement operation of the position measurement apparatus according to the first embodiment.
FIG. 3A shows a state in which the spreader 5 suspended from the trolley 4 is viewed from the installation position of the camera 8 in the transfer crane 1 shown in FIG. FIG. 3B shows an image 20 in which the camera 8 captures the spreader 5 shown in FIG.

The spreader 5 shown in FIG. 3A is suspended from the trolley 4 as described above, and represents a state of shaking as indicated by an arrow a from a stationary position 5a indicated by a broken line in the drawing. . In the spreader 5 of FIG. 3, landmarks 9 a and 9 b, which are the aforementioned landmarks 9, are attached with a distance D.
Here, an intermediate point between the landmark 9a and the landmark 9b is defined as a position P. Also, orthogonal coordinates are set with the center point of the stationary position 5a as the origin Os, the longitudinal direction of the spreader 5 as the Y axis, and the short direction as the X axis. In this coordinate system, the coordinate value of the position P of the spread spreader 5 is (Xs, Ys).

The position P of the spreader 5 is the same point as the origin Os when the spreader 5 is stationary.
In FIG. 3, the landmarks 9a and 9b are shown in a circular shape for easy understanding of the distance D. However, the shapes of the landmarks 9a and 9b are not limited to those illustrated.
As shown in FIG. 1B, the landmarks 9a and 9b have a radial pattern, and the center points of the radial patterns of the landmarks 9a and 9b are separated by a distance D.

Also, in the image 20 shown in FIG. 3B, the center point of the stationary position 5a indicated by the broken line is the origin Oc, the longitudinal direction of the spreader 5 in the image 20 is the y axis, and the short direction is the x axis. Set the orthogonal coordinates. In this coordinate system, the coordinates of the position P of the spreader 5 in the image 20 are (x _s , y _s ). The coordinates of the landmark 9a in the image are (x ₁ , y ₁ ), and the coordinates of the landmark 9b are (x ₂ , y ₂ ). Further, the distance between the landmark 9a and the landmark 9b in the image is d.

The origin Os is set in advance in a state where the spreader 5 of the transfer crane 1 is stationary. Further, since the camera 8 is fixed to the trolley 4 as described above, the origin Oc set in the stationary spreader 5 in the image 20 is also set in advance in a predetermined position in the image 20. is there.
For example, the image processing unit 7 stores the position of the origin Oc in the image 20 in advance in the storage unit 14, and is stored in the storage unit 14 when the arithmetic processing unit 12 performs image processing to be described later. The origin Oc is used.

FIG. 4A shows each value used when the image processing unit 7 measures the position of the spreader 5. The orthogonal coordinates shown in FIG. 4A are set in the same manner as the coordinate system shown in FIG. 3A with respect to the X axis and the Y axis.
In FIG. 4A, the origin Os is set, for example, at the lower end surface of the trolley 4, and the Z axis orthogonal to the XY plane formed by the X axis and the Y axis is set. Further, FIG. 4B shows the distance between the respective parts and the like, by extracting the positional relationship of the respective parts shown in FIG. 4A in the Z-axis direction.

  Here, an image plane 20 a indicated by a broken line in FIG. 4A represents the position of the light receiving element surface of the camera 8, and the focal point Fo represents the position of the imaging focal point of the camera 8. Let f be the distance from the image plane 20a to the focal point Fo. The distance from the origin Os, that is, the trolley 4 to the focal point Fo is Zc. Further, the distance from the focal point Fo to the spreader 5, that is, the position P is assumed to be h. Further, the distance from the origin Os to the spreader 5, that is, the position P is set to Zs. In other words, the coordinate value of the position P in the Z-axis direction is Zs.

The arithmetic processing means 12 of the image processing unit 7 calculates the displacement due to the shaking of the spreader 5 shown in FIG.
An image captured by the camera 8 is converted into image data of a predetermined format by the image input means 11. The arithmetic processing means 12 to which the image data is input obtains the coordinate values of the landmarks 9a and 9b included in the image 20 shown in FIG. 3B by template matching processing described later. The coordinate values of the landmarks 9a and 9b are the coordinate values of the center point of the radial pattern of each landmark. Specifically, the coordinate values of the landmark 9a shown in FIG. ₁ , y ₁ ) and the coordinate value (x ₂ , y ₂ ) of the landmark 9b.
After obtaining the coordinate value (x ₁ , y ₁ ) and the coordinate value (x ₂ , y ₂ ) in the image 20, the distance d in the image 20 is obtained using the following equation (1).

After obtaining the distance d, the arithmetic processing means 12 obtains the position of the spreader 5 in the image 20 using the following equation (2). Specifically, the coordinate value (x _s , y _s ) of the position P included in the image 20, that is, the center point of the spreader 5 that is shaken and displaced, that is, the image 20 is obtained.

  The distance D shown in FIG. 3A is a distance between the landmark 9a and the landmark 9b attached to the upper surface of the actual spreader 5, and is a predetermined numerical value. Therefore, in the position measurement process of the spreader 5, the value of the distance D is stored in advance in the storage unit 14 or the like. The arithmetic processing unit 12 appropriately reads the value of the distance D from the storage unit 14 and uses it for the calculation.

When the distance d in the image 20 is obtained by the above equation (2), since the actual distance D is known in advance, the coordinate values (x _s , y _s ) of the spreader 5 shown in the image 20 are obtained. ), Coordinate values (Xs, Ys) representing the actual position of the spreader 5 can be obtained.
The above description relates to the coordinate values in the X direction and the Y direction in the coordinate systems shown in FIGS. 3A and 4A, but the coordinate values in the Z direction, that is, the spreader 5 is actually positioned. The height to be obtained can also be obtained by calculation.

The focal point Fo of the camera 8 and the position of the light receiving element, that is, the image plane 20a are known in advance. That is, the distance f and the coordinate value Zc shown in FIG. 4 are predetermined values. Each value of the distance f and the coordinate value Zc is stored in the storage unit 14 in advance.
The arithmetic processing means 12 uses the distance D, distance d, distance f, and coordinate value Zc stored in the storage means 14 to calculate the following expression (3), and to indicate the coordinates indicating the actual position of the spreader 5. The values (Xs, Ys, Zs) are obtained. Note that the distance h shown in FIG. 4 corresponds to (D / d) × f included in Equation (3).

  Thus, the image processing unit 7 obtains the position P of the spreader 5 using triangulation. The coordinate value of the position P described here is position information of the spreader 5 that is shaking. Data representing the coordinate values (Xs, Ys, Zs), etc. of the position P is converted into a predetermined data format by the output processing means 13 and output to, for example, the control mechanism section described above.

[Processing to obtain landmark position in image]
A processing operation for obtaining the positions of the landmarks 9a and 9b included in the image 20 by the image processing means 12 will be described.
The image processing means 12 detects a change in lightness / darkness or lightness / darkness of an image using a vector code correlation method.
The vector code correlation method encodes the gradient of the luminance change for eight pixels around the pixel of interest. Such encoding is performed on a captured image, for example, and an encoded image is prepared in advance. The correlation with the template is calculated.
The template is data obtained by vector-encoding a predetermined image pattern.

FIG. 5 is an explanatory diagram illustrating a processing operation of the position measurement apparatus according to the first embodiment. This figure shows a process of encoding a part of an image input to the image processing unit 7.
For example, when the pixel b shown on the left side in FIG. 5 is the target pixel, 8 pixels around the pixel b are extracted. These pixels have values of 0 to 255 representing luminance, for example. The pixel b and 9 pixels of the peripheral pixels are processed by the Prewitt filter.

  The Prewitt filter is composed of the horizontal filter of the image shown in the upper center of FIG. 5 and the vertical filter shown in the lower central of FIG. Specifically, the above nine pixel values are multiplied by a coefficient matrix that is a horizontal filter, and the value, that is, the horizontal inclination of the target pixel b is obtained. Further, the nine pixel values are multiplied by a coefficient matrix that is a vertical filter, and the value, that is, the vertical inclination of the target pixel b is obtained.

  The horizontal inclination (value indicated as “−68” in the figure) and the vertical inclination (value indicated as “68” in the figure) thus obtained are respectively compared with predetermined threshold values. This threshold value is an appropriate two value that classifies each of the above-described inclinations into, for example, “positive inclination”, “no inclination”, and “negative inclination”. These threshold values are compared with the above-described slopes, and encoded into three values according to the comparison result. Specifically, “01” is set for “positive slope”, “00” for “no slope”, and “10” for “negative slope”, and the codes are converted into codes represented by 2 bits. In this way, a vector code of a predetermined part of the image is obtained.

FIG. 6 is an explanatory diagram illustrating a processing operation of the position measurement apparatus according to the first embodiment. This figure shows a process of obtaining the correlation between the input image from the camera 8 and the template by vector encoding the inclination of each pixel.
The image captured by the camera 8 is vector-encoded by the arithmetic processing unit 12 as described above. Further, the template stored in the storage unit 14 is also vector-encoded as described above.
Note that the template is data corresponding to a portion T indicated by a broken line in FIG. 1B, for example, and represents a central portion of the radial pattern of the landmark 9.

The arithmetic processing means 12 obtains XOR (exclusive or) between the vector-encoded partial image shown in FIG. 6 and the template, and detects matching for each pixel.
An exclusive OR (XOR result) of each vector code of the partial image obtained by processing the image captured by the camera 8 as described above and each vector code of the template is obtained, and the same value illustrated in the lower right side of FIG. Using the relationship table, the degree of difference between the aforementioned XOR results is obtained.
The above equivalence relation table is stored in the storage unit 14, for example, and is provided inside the image processing unit 7 so that the arithmetic processing unit 12 can refer to it as appropriate.

The arithmetic processing unit 12 obtains a value representing the degree of difference in each pixel as described above, and further obtains the sum of these values. The sum of the differences is a value representing the difference between the partial image and the template.
The arithmetic processing unit 12 obtains the degree of difference between the sequentially different partial images and the template described above, and searches for a partial image in which the degree of difference is smaller than a predetermined value. When there is a partial image with a small difference, the coordinate value of the partial image is obtained.

The template used here represents the center portion of the landmark 9 as described above. Therefore, there is a high possibility that the center portion of the landmark 9a or the landmark 9b is shown in the partial image with a small difference. That is, the coordinate value of the partial image is considered to be the coordinate value (x ₁ , y ₁ ) of the landmark 9a or the coordinate value (x ₂ , y ₂ ) of the landmark 9b.
As described above, the arithmetic processing unit 12 searches the center portion of the radial pattern of the landmarks 9a and 9b from the image 20 captured by the camera 8 in units of pixels of the image 20, and the coordinate value of the center portion. Ask for.

  When the position measurement is performed by obtaining the correlation as described above using the concentric pattern landmark 109 shown in FIG. 13B, the position measurement is performed using the radial pattern template 9 as described above. Compare with the case.

7-10 is explanatory drawing which shows the position of the landmark which the position detection apparatus by 1st Embodiment detected from the image.
FIGS. 7 and 8 show the case where the height of the spreader 5, that is, the position of the landmark is changed along the Z-axis direction shown in FIG. 4, and the spreader 5 or landmark in a stationary state is imaged at each height. The position of the landmark in the captured image is shown.
Specifically, FIGS. 7 and 8 are graphs showing the relationship between the actual height of the spreader 5 and the position of the landmark in the image captured at this height. The position of the landmark is shown in FIG. The position in the x-axis direction and the position in the y-axis direction of the coordinate system shown are shown separately.

In the graphs of FIGS. 7 and 8, the vertical axis represents the pixel [pixel] at the position where the landmark is detected in the image captured by the camera 8, and the horizontal axis represents the height of the spreader 5, that is, the camera 8 is installed. Represents the distance [mm] between the trolley 4 and the landmark.
The height illustrated in the graph here is not necessarily the actual height of the spreader 5 in the transfer crane 1 shown in FIG.

In FIG. 7, the landmark 109 shown in FIG. 13B is attached to the spreader 5, and this is picked up by the camera 8, and detection when the image processing unit 7 detects the position of the spreader 5 in the image. It is a graph which shows a result.
FIG. 7A shows the detected position in the x-axis direction in the image when the height of the spreader 5 is changed, that is, when the position of the landmark 109 is changed in the Z-axis direction shown in FIG. It is the shown graph. The horizontal axis of FIG. 7A is the height of the spreader 5, that is, the actual position of the landmark in the Z-axis direction, and the vertical axis is the detected position of the landmark 109 in the captured image in the x-axis direction. is there.
FIG. 7B is a graph showing the detected position in the y-axis direction in the image when the height of the spreader 5 is similarly changed. The horizontal axis of FIG. 7B is the height of the spreader 5 as in FIG. 7A, and the vertical axis is the detection position of the landmark 109 in the captured image in the y-axis direction.

  When the imaging performance of the camera 8 is affected and the distance between the camera 8 and the landmark 109 is a certain distance, it is difficult to capture the concentric pattern in the image so as to have an appropriate size. In other words, an image in which a concentric pattern is preferably included in the portion T indicated by the broken line in FIG. 13B cannot be obtained. The portion T is a portion for obtaining a correlation with a template described later.

As described above, when the image of which the concentric pattern is not captured in an appropriate size is processed by the vector code correlation method and the correlation with the template corresponding to the concentric pattern not shown is obtained, It becomes difficult to accurately detect the central portion of the mark 109;
Therefore, as shown in FIGS. 7A and 7B, even if the actual landmark 109 is stationary, the position of the landmark 109 detected in the image is not constant, which is May become stable.

FIG. 8 shows the detection result when the landmark 9 shown in FIG. 1B is attached to the spreader 5, the image is captured by the camera 8, and the image processing unit 7 detects the position of the spreader 5. It is a graph.
FIG. 8A shows the detected position in the x-axis direction in the image when the height of the spreader 5 is changed, that is, when the position of the landmark 9 is changed in the Z-axis direction shown in FIG. It is the shown graph. The horizontal axis of FIG. 8A is the height of the spreader 5, that is, the actual position of the landmark 9 in the Z-axis direction, and the vertical axis is the detected position of the landmark 9 in the captured image in the x-axis direction. It is.
FIG. 8B is a graph showing the detected position in the y-axis direction when the height of the spreader 5 is similarly changed. The horizontal axis in FIG. 8B is the height of the spreader 5 as in FIG. 8A, and the vertical axis is the detection position in the y-axis direction of the landmark 9 in the captured image.

  When the radial pattern landmark 9 is image-processed by the vector code correlation method, it is possible to specify the position of each pixel of the captured image and detect the central portion of the radial pattern as described above. As shown in FIGS. 8A and 8B, when the landmark 9 is actually stationary, the position of the landmark 9 detected from the image is also constant.

  9 and 10, when the height of the spreader 5, that is, the position of the landmark is changed along the Z-axis direction shown in FIG. 4, the spreader 5 or the landmark swaying at each height is imaged. The positions of the landmarks in the x-axis direction or the y-axis direction in the captured image are shown.

FIG. 9 is a graph showing the shake of the spreader 5 detected by the image processing unit 7 when the landmark 109 having a concentric pattern is attached to the spreader 5, picked up by the camera 8.
FIG. 9A is a graph showing the position of the landmark 109 in the x-axis direction detected from the image according to the shake of the spreader 5.
FIG. 9B is a graph showing the position of the landmark 109 in the y-axis direction detected from the image according to the shake of the spreader 5.
FIG. 10A is a graph showing the position of the landmark 9 in the x-axis direction detected from the image according to the shake of the spreader 5.
FIG. 10B is a graph showing the position of the landmark 9 in the y-axis direction detected from the image according to the shake of the spreader 5.

  From the measurement results shown in FIG. 9 and FIG. 10, the landmark detection process by the image processing unit 7 is performed when the radial pattern landmark 9 is used and when the concentric pattern landmark 109 is used. It can be seen that there is no big difference. In other words, with respect to shaking of the spreader 5 parallel to the imaging plane 20a of the camera 8 shown in FIG. 4 or movement of the landmark, position measurement using the landmark 9 and position measurement using the landmark 109 are performed. Can be performed with the same accuracy.

  As described above, when the position measurement using the vector code correlation method is performed, the height of the spreader 5 is greater when the landmark 9 having the radial pattern is used than when the landmark 109 having the concentric pattern is used. It can be seen that stable measurement results can be obtained. That is, when the landmark 9 is attached to the measurement object such as the spreader 5, the image processing unit 7 more accurately measures the distance between the imaging position of the trolley 4 where the camera 8 is installed and the measurement object. can do. In addition, the position of the spreader 5 can be measured more accurately from the captured image.

FIG. 11 is an explanatory diagram illustrating the operation of the position measurement apparatus according to the first embodiment. This figure shows the landmark 9 in the image captured by the camera 8.
FIG. 11A shows the landmark 9 in the image when the periphery of the landmark 9 is imaged with normal brightness using, for example, an illumination device. In other words, it is an image captured in a surrounding environment with brightness suitable for the imaging performance of the camera 8.
FIG. 11B shows a landmark 9 in an image when there is no illumination light by the above-described illumination device or the like, for example, an image taken indoors without illumination light.
FIG. 11C shows, for example, an image in a case where an additional illumination device is added to the illumination device having normal brightness that is imaged as shown in FIG. The landmark 9 is slightly halated at the center of the radial pattern.

As described above, the image processing unit 7 detects the position of the landmark 9 by using the vector code correlation method. Similarly, in each image shown in FIG. Can detect the coordinate value of the center portion of the landmark 9 with sufficient accuracy.
Specifically, for example, when the landmark 9 attached to the spreader 5 in FIG. 1 is imaged, even when the spreader 5 is illuminated by using an illumination device (not shown) provided in the transfer crane 1. It is possible to perform stable position measurement.

When the transfer crane 1 operates, the irradiation light from the above-described lighting device is blocked by some structure, and the brightness around the landmark 9 on the upper surface of the spreader 5 may change.
The arithmetic processing means 12 of the image processing unit 7 processes the image data from the camera 8 using the vector code correlation method that is robust against fluctuations in the brightness of the image. The position of the landmark 9 can be accurately detected even when the brightness that illuminates the landmark 9 changes.

  FIG. 12 is an explanatory diagram illustrating a detection result of the position detection device according to the first embodiment. This graph is a graph showing the frequency of deviation between the position obtained from each image illustrated in FIGS. 11A to 11C and the actual position. FIG. 12 shows “error when detecting a landmark in normal illumination” corresponding to the image of FIG. 11A, and “error when detecting a landmark in the absence of illumination” corresponding to the image of FIG. , And “error at the time of landmark detection in additional illumination” corresponding to the image of FIG. 11C.

  The error distribution shown in FIG. 12 shows an error generated in the position detection in the x-axis direction shown in FIG. The vertical axis in FIG. 12 indicates the frequency [%] in which an error is included when image processing for detecting the position of the landmark 9 in the captured image is performed, and the horizontal axis detects the above position. It represents the magnitude of an error that has occurred in image processing. The magnitude of the error is expressed in units of pixels [pixel] in the image. When no error is included, “0” [pixel] is used, and the number of pixels deviating from the position of this pixel is used for the error. Represents the distribution.

In the example shown in FIG. 12, the error distribution of the measurement position is within the range of ± 4 [pixel], and the frequency of occurrence of the measurement error is less than 25 [%] when the frequency of occurrence of the measurement error is 1 [pixel]. It is.
When the arithmetic processing unit 12 performs image processing using the vector code correlation method, the error occurrence frequency of the detection position does not change significantly depending on the brightness of the landmark 9, that is, the presence or absence of illumination light from an illumination device or the like, The position detection process can be performed while suppressing the influence of the brightness change around the landmark 9.
Further, by performing image processing using the vector code correlation method as described above, the processing time can be shortened compared to image processing using a general normal correlation method.

  As described above, according to the first embodiment, the radial pattern landmark 9 is attached to the measurement target spreader 5, and the image processing unit 7 uses the vector code from the image including the landmark 9 captured by the camera 8. The coordinate value in the image of the landmark 9 is detected using the correlation method. Further, the height of the spreader 5 of the measurement object or the distance between the installation position of the camera 8 and the position of the spreader 5 is obtained from the detected coordinate values. By doing in this way, the position of the landmark 9 can be detected in pixel units of the captured image, and highly accurate measurement can be performed.

  Further, when the image processing unit 7 detects the landmark 9 in the captured image, the edge of the radial pattern of the landmark 9 is detected, and at this time, processing by the vector code correlation method is performed. In this way, since the inclination of each pixel value of the captured image is handled, the position of the landmark 9 can be accurately detected from the captured image even if the brightness around the landmark 9 to be captured changes. Can do.

  In the present embodiment, the position measuring device that images the spreader 5 positioned below with the camera 8 disposed on the upper side of the transfer crane 1 and measures the position of the spreader 5 has been described. The position measuring device of the present invention is not limited to measuring the position of the spreader suspended on the crane as described above, and the imaging direction of the camera 8 is horizontal, and the measurement object arranged in the imaging direction is It is also possible to measure the position three-dimensionally.

It is explanatory drawing which shows the structure of the position measuring device by the 1st Embodiment of this invention. It is a block diagram which shows the structure of the position measuring device by 1st Embodiment. It is explanatory drawing which shows the measurement operation | movement of the position measuring device by 1st Embodiment. It is explanatory drawing which shows the measurement operation | movement of the position measuring device by 1st Embodiment. It is explanatory drawing which shows the processing operation of the position measuring device by 1st Embodiment. It is explanatory drawing which shows the processing operation of the position measuring device by 1st Embodiment. It is explanatory drawing which shows the position of the landmark which the position detection apparatus by 1st Embodiment detected from the image. It is explanatory drawing which shows the position of the landmark which the position detection apparatus by 1st Embodiment detected from the image. It is explanatory drawing which shows the position of the landmark which the position detection apparatus by 1st Embodiment detected from the image. It is explanatory drawing which shows the position of the landmark which the position detection apparatus by 1st Embodiment detected from the image. It is explanatory drawing which shows operation | movement of the position measuring device by 1st Embodiment. It is explanatory drawing which shows the detection result of the position detection apparatus by 1st Embodiment. It is explanatory drawing which shows the structure of the conventional position measuring apparatus.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Transfer crane, 2 ... Container, 3 ... Guide, 4 ... Trolley, 5 ... Spreader, 6 ... Wire rope, 7 ... Image processing part, 8 ... Camera, 9, 9a, 9b ... Landmark, 11 ... Image input means , 12 ... arithmetic processing means, 13 ... output processing means, 14 ... storage means, 20 ... image, 20a ... image plane.

Claims (5)

A plurality of landmarks that have a pattern that is imaged in a similar manner regardless of the imaging distance, and are provided separately from the measurement object for measuring the position;
A camera installed and fixed so as to capture the plurality of landmarks in the same image;
An image processing unit that detects the positions of the plurality of landmarks from the captured image of the camera, and calculates the positions of the measurement objects by using the positions of the plurality of landmarks in the image;
A position measuring device comprising:   The position measurement apparatus according to claim 1, wherein the landmark has a radial pattern in which a plurality of colors having different densities are alternately arranged.   The position measuring apparatus according to claim 1, wherein the image processing unit detects an edge of the landmark pattern to obtain a position of the landmark.   4. The image processing unit detects a position of the landmark by obtaining a correlation between a template set in advance using a vector code correlation method and a landmark in the image. The position measuring device according to any one of the above.   The position measurement apparatus according to claim 4, wherein the template corresponds to a central portion of a radial pattern of the landmark.

Images