JPH0611316A - Picture processor for measuring object - Google Patents

Abstract

PURPOSE:To obtain required measurement information improved in measurement accuracy about a plurality of objects existing at intervals without largely increasing the cost by using one set of picture processor to perform required picture processing in real time without installing picture processors to each of a plurality of image pickup devices. CONSTITUTION:A right-side CCD camera 11R picks up the image of a light cutting line produced on a right-side rail RR by planar light S from a light source 2 and a left-side CCD camera 11L picks up the image of a light cutting line produced on a left-side rail RL. Both cameras 11R and 11L perform simultaneous scanning based on a synchronizing signal from an external synchronizing signal generator 12. A video signal synthesizer 13 synthesizes video signals from the cameras 11R and 11L into video signals corresponding to one picture at every field by switching the signals in a required section in the same field. A picture processor 14 analyzes the picture composed of the synthesized video signals and finds the two-dimensional coordinates of inside upper-surface corners of the rails RR and RL.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a shape of a plurality of target objects existing at intervals such as detecting the positions of specific parts of the left and right rails of a railroad by utilizing a light cutting method. The present invention relates to an object measurement image processing device used for optical measurement of dimensions, positions, and the like.

[0002]

2. Description of the Related Art As one of the maintenance and inspection works in railways, in order to accurately grasp the maintenance time by checking the secular change of the structures around the railway such as tunnels, bridges, and station buildings, and the construction of the structures around the railway. In order to perform the inspection for the limit, the three-dimensional shape of these railroad peripheral structures along the rail (track) is measured.

The applicant of the present invention has previously proposed a three-dimensional shape measuring device for a track peripheral structure which measures the three-dimensional shape of the track peripheral structure on the side along the rail ( Japanese Patent Application No. 3-332101). FIG. 7 is a diagram showing an overall configuration of a three-dimensional shape measuring apparatus for a line peripheral structure according to this conventional technique, and FIG. 8 is a perspective view showing an outline of the three-dimensional shape measuring apparatus shown in FIG. 9 is a diagram for explaining a three-dimensional shape measuring method by the three-dimensional shape measuring apparatus shown in FIG. 7, and FIG. 10 is a method for obtaining a midpoint between rails by the three-dimensional shape measuring apparatus shown in FIG. FIG. 11 is an explanatory view of an image of a light cutting line generated on the tunnel wall surface obtained by the structure CCD camera shown in FIG. 7, and FIG. 12 is obtained by the rail CCD camera shown in FIG. It is an explanatory view of an image of a light section line which is generated on a rail.

As shown in FIGS. 7 and 8, this three-dimensional shape measuring apparatus for a track peripheral structure includes a moving carriage 1 which travels on rails (tracks) RR and RL, and is provided on the moving carriage 1. Radially toward the wall surface of the tunnel T and the rails RR, RL in the plane substantially orthogonal to the rails RR, RL, that is, in the plane substantially orthogonal to the traveling direction of the moving carriage 1 (indicated by a white arrow in the figure). A light source 2 for projecting a spread flat light S, and a structure CCD camera 3 for sequentially capturing light cut lines generated on the wall surface of the tunnel T by the flat light S from the light source 2 as the moving carriage 1 travels. A rail CCD camera 4 that scans in synchronization with the camera 3 and images the light cutting lines generated on the rails RR and RL by the flat light S, and an image processing device 5.
It is composed of a computer 6.

The above structure CCD camera 3 is a field storage type CCD in which an image of a light cutting line (see FIG. 11) generated on a wall surface of a tunnel is obtained for each field (1/60 seconds).
The camera is fixed to the movable carriage 1 so that its optical axis CP ₁ intersects the plane of the flat light S at an angle θ ₁ . Further, the rail CCD camera 4 provided for the purpose of eliminating the measurement error due to the shaking of the moving carriage 1 is
A field-storage CCD camera in which an image of the light-section line generated on the RR and RL (see FIG. 12) is obtained for each field (1/60 seconds), and its optical axis is on the plane of the flat light S. CP
₂ is fixed to the moving carriage 1 so as to intersect at an angle theta _2. The structure CCD camera 3 is a field storage type CCD in which an image of a light cutting line generated on the tunnel wall surface (see FIG. 11) is obtained for each field (1/60 seconds).
The camera is fixed to the movable carriage 1 so that its optical axis CP ₁ intersects the plane of the flat light S at an angle θ ₁ . Further, the rail CCD camera 4 provided for the purpose of eliminating the measurement error due to the shaking of the moving carriage 1 is
A field-storage CCD camera in which an image of the light-section line generated on the RR and RL (see FIG. 12) is obtained for each field (1/60 seconds), and its optical axis is on the plane of the flat light S. CP
₂ is fixed to the moving carriage 1 so as to intersect at an angle theta _2. Although the rail CCD camera 4 is not shown in FIG. 7, u and v define an orthogonal coordinate system on the optical cutting line image.
Indicates a horizontal scanning line direction, and v indicates a direction perpendicular to u (see FIGS. 11 and 12).

The image processing device 5 obtains the two-dimensional coordinates (two-dimensional position coordinates) of the light cutting line showing the wall surface of the tunnel T on each image for each light cutting image obtained by the structure CCD camera 3. , Each rail R on the image for each light section image obtained by the rail CCD camera 4
This is for obtaining the two-dimensional coordinates (two-dimensional position coordinates) of the light cutting line showing the inner corner of the upper surface of R and RL. The computer 6 uses the coordinates obtained by the image processing device 5 as a reference in the structure orthogonal coordinate system O-XYZ based on the midpoint O _i between the rails RR and RL (gauge) in accordance with the traveling of the moving carriage 1. The point is for calculating the three-dimensional coordinates of the light cutting line indicating the wall surface shape on the wall surface of the tunnel T as a point.

The Cartesian coordinate system will be described below. In the structure orthogonal coordinate system O-XYZ, as shown in FIG. 8, the midpoint between the rails RR and RL at predetermined positions of the rails RR and RL is set as the origin O, and the direction in which the rails RR and RL extend (the moving carriage 1 The traveling direction) is set as the X axis, and its Y
The axis indicates the width direction of the tunnel T, and its Z axis is the tunnel T.
Shows the height direction of. Further, as shown in FIG. 9, the structure camera orthogonal coordinate system O _1i -XYZ having the origin O _{1i as} the principal point of the imaging lens of the structure CCD camera 3 and the main imaging lens of the rail CCD camera 4. the rail camera Cartesian coordinate system O _2i -xyz to the point as the origin O _2i, the axes are set to be parallel therewith of the structure orthogonal coordinate system O-XYZ. Where the suffix i (i =
1 to n) are flat plate-shaped lights S as the movable carriage 1 travels.
Represents the numbers of the tunnel cross-sections sequentially formed by the projection of the above-mentioned coordinate systems O _1i -XYZ, O _2i -XYZ.
Indicates a coordinate system set in a tunnel section i corresponding to one section of the tunnel T.

The three-dimensional shape measuring apparatus for a line peripheral structure in this example configured as described above sequentially images the optical cutting lines generated on the wall surface of the tunnel T and the rails RR and RL, and the tunnel T The three-dimensional shape of the wall surface of the tunnel T along the rails RR and RL is continuously measured while the moving carriage 1 is traveling, based on the position coordinates of the light cutting line for each cross section i. Here, in order to facilitate understanding, the operation of the tunnel T in one cross section i will be described below.

When the flat light S is projected by the light source 2,
The structure CCD camera 3 images the light cutting lines generated on the wall surface of the tunnel T along the circumferential direction thereof, and the rail CCD camera 4 images the light cutting lines generated on the rails RR and RL. The video signal of is supplied to the image processing device 5.

As shown in FIG. 11, the image processing device 5 includes a CCD camera 3 for a structure in one of the image processing devices.
For the video signal of one image given in every field (1/60 seconds) from, the analysis process is performed to detect the position where the brightness level on each horizontal scanning line changes more than a predetermined value. The two-dimensional coordinates C _k (u, v), k = 1 to N of the light section line indicating the tunnel wall surface of No. 1 are obtained at the video rate (1/60 seconds). Here, k represents the number of horizontal scanning lines, and the number thereof is about 240.

Further, the image processing device 5 is arranged such that, in the other image processing device, as shown in FIG.
For the video signal of one image given from the D camera 4 for each field (1/60 second), perform an analysis process to extract the characteristics of the pattern at the position where the brightness level on each horizontal scanning line changes by a predetermined value or more. Thus, the two-dimensional coordinate C of the light cutting line indicating the position of the inner corner of the upper surface of the right rail RR
_{k = kR} (u, v) and the two-dimensional coordinates C _{k = kL} (u, v) of the light cutting line indicating the position of the inner corner of the upper surface of the left rail RL are calculated at the video rate (1/60 seconds). . Here, kR represents the number of the horizontal scanning line corresponding to the light cutting line position indicating the inner corner of the upper surface of the right rail RR, and kL corresponds to the light cutting line position indicating the inner edge of the upper surface of the left rail RL. Indicates the number of horizontal scan lines. The data of these two-dimensional coordinates obtained at the video rate (1/60 seconds) is used for the blanking time (about 1.2 ms).
ec) is given to the computer 6.

Given the above two-dimensional coordinate data,
Calculator 6 has the following field time (about 16 msec)
First, the two-dimensional coordinate C _k (u, from the image processing device 5 of the light cutting line showing the wall surface of the tunnel T on the light cutting image is shown.
v), k = 1 to N, the three-dimensional coordinates P _i ^k (X _i ^k , Y _i ^k , Z _i ) of the light cutting line on the tunnel wall surface in the structure camera orthogonal coordinate system O _1i -XYZ. ^k ), k =
1 to N are calculated by a predetermined formula. Then, as shown in FIG. 9, the origin O, which is the principal point of the CCD camera 3 for structure,
Starting from _1i , the calculated point P _i ^k (X _i ^k , Y _i
A vector P _i ^k _meas whose end point is ^k , Z _i ^k ) is obtained.

Further, the computer 6 sets the two-dimensional coordinates C _{k = kR} (u, v) of the light cutting line from the image processing device 5 which indicates the inner edge of the upper surface of the right rail RR on the light cutting image. Based on this, the three-dimensional coordinates r _Ri of the light cutting line indicating the inner corner of the upper surface of the right rail RR in the rail camera orthogonal coordinate system O _2i -XYZ with the principal point of the rail CCD camera 4 as the origin O _2i.
(X _Ri , Y _Ri , Z _Ri ) is calculated by a predetermined formula. Similarly, the two-dimensional coordinates C _{k = kL} (u,
Based on v), the three-dimensional coordinates r _Li (X _Li , Y _Li , Z _Li ) of the light cutting line showing the inner edge of the upper surface of the left rail RL in the Cartesian coordinate system O _2i -XYZ can be calculated by a predetermined formula. Calculate by Then, as shown in FIG. 10, a vector R _Ri whose origin is the origin O _2i of the rail CCD camera 4 and whose end point is the point r _Ri (X _Ri , Y _Ri , Z _Ri ) calculated above.
And an origin O _2i as a starting point, and the point r calculated above
A vector R _Li whose end point is _Li (X _Li , Y _Li , Z _Li ) is obtained. From these vectors R _Ri and R _Li , R _Oi = (R
_The vector R _Oi having the origin O _2i as the starting point and the midpoint O _i between the rails RR and RL as the ending point is obtained by the calculation of _Ri + R _Li ) / 2.

On the other hand, since the structure CCD camera 3 and the rail CCD camera 4 are rigidly fixed and their positional relationship is predetermined, the main point (origin O _2i ) of the rail CCD camera 4 is the starting point. The vector K having the principal point (origin O _1i ) of the structure CCD camera 3 as the end point is constant.
Therefore, the vector P _i ^k _true whose starting point is the middle point O _i between the rails RR and RL and whose end point is the above point P _i ^k (X _i ^k , Y _i ^k , Z _i ^k ) is P _i ^k _true = P _i ^k _meas + K-R
Given by _Oi . This vector P _i ^k _true is calculated by the computer 6.

In this way, the tunnel T is set with the midpoint O _i between the rails RR and RL as a reference point according to the traveling of the mobile carriage 1.
The three-dimensional coordinates of the light cutting line on the wall surface of the tunnel wall surface of the tunnel wall surface along the rails RR and RL are set so that even if the moving carriage 1 is swaying, the measurement error due to this is extremely small. The three-dimensional shape is measured without being affected by the shaking of the moving carriage 1.

[0016]

In the three-dimensional shape measuring apparatus according to the above-mentioned conventional technique, the total measurement accuracy (measurement accuracy) is obtained by correcting the measurement error (measurement error) due to the movement of the moving carriage 1. _Is determined by the sum of the measurement accuracy of the vector R _{Oi and} the measurement accuracy of the vector P _i ^k _meas . Therefore, the rail CC that images the light cutting lines generated on the left and right rails RR and RL by the flat light S
The detection resolution of the D camera 4 should be as high as possible.

Therefore, the detection resolution of the rail CCD camera 4 as the image pickup device in the above-mentioned conventional example will be described. FIG. 13 is a diagram for explaining the field of view of the imaging device when imaging the railway rail that is mainly used in Japan in the three-dimensional shape measuring device shown in FIG. 7. Rail R
The gauge between R and RL is set to 1067 mm, and when one rail CCD camera 4 is used to capture the optical cutting lines on the left and right rails RR and RL, the range shown in FIG. It is necessary to take a view of.

In other words, the field of view in the v direction (direction perpendicular to the scanning line) has both rails RR and RL even if the moving carriage 1 is shaken.
Is set to 1300 mm with a margin of ± 20 mm so that it always fits within the field of view. In the u-direction (scanning line direction) field of view, the optical axis of the CCD camera for rails 4 makes an angle of 45 degrees with the plane of the flat light S. From the calculation based on the fact that they are arranged, it becomes 2675 mm. The dimensions of the CCD image pickup elements arranged two-dimensionally are 15.75 μm × 512 pixels to 8.064 mm in the u direction and 26 μm × 240 pixels to 6.24 mm in the v direction. Therefore, when it is attempted to image the optical cutting lines generated on the left and right rails RR and RL using one rail CCD camera 4, the detection resolution Δu in the u direction is Δu = 2675 /
512 = 5.2 mm, and the detection resolution Δv in the v direction is Δv
= 1300/240 = 5.4 mm.

In order to increase the detection resolution to a value higher than the above value and perform highly accurate measurement, the number of image pickup devices must be increased as long as the image pickup device of the NTSC system is used. In the three-dimensional shape measuring apparatus according to the above-mentioned conventional technique, the optical cutting lines generated on the left and right rails RR and RL as an example of the optical images of a plurality of target objects existing at intervals are defined by the left and right rails RR and RR. In the case where two image pickup devices are picked up corresponding to the RL, the three-dimensional shape of the tunnel wall surface is measured while the moving carriage 1 is running. If a total of two image processors are provided for each image pickup device in order to analyze the image obtained by the image pickup device in real time, there is a drawback that the device cost increases significantly.

The present invention has been made in view of the above circumstances, in which optical images of a plurality of target objects existing at intervals are synchronously picked up by each image pickup device for each target object, Each of the video signals from the plurality of image pickup devices is switched to synthesize a video signal corresponding to one image for each field, and an image by the synthesized video signal is analyzed by an image processor in real time. Thus, for example, one image processor can perform required image processing in real time without providing an image processor for each of a plurality of image pickup devices, and the intervals can be increased without causing a large increase in the cost of the device. It is an object of the present invention to provide an image processing apparatus for object measurement, which can obtain required measurement information with improved measurement accuracy for a plurality of existing target objects.

[0021]

In order to achieve the above object, an object measuring image processing apparatus according to the present invention comprises a light source for projecting light on a plurality of target objects existing at intervals, and an external synchronization signal. A generator, a plurality of image pickup devices that are synchronously scanned by a synchronizing signal from the external synchronizing signal generator, and pick up optical images of the plurality of target objects for each target object, and a plurality of image pickup devices from the plurality of image pickup devices. A video signal is switched based on a sync signal from the external sync signal generator in a required section in the same field, and a video signal corresponding to one image is synthesized for each field and output as a synthesized video signal. A signal synthesizer and an image based on a synthesized video signal from the video signal synthesizer are analyzed for each field based on a synchronization signal from the external synchronization signal generator, and the plurality of images are analyzed. Characterized by comprising an image processing device for obtaining the required measurement information about the elephants object.

[0022]

In the image processing apparatus for object measurement according to the present invention, the light from the light source is projected onto a plurality of target objects existing at intervals, and the optical images of the plurality of target objects are provided for each target object. The image is taken by the imaging device. Each video signal from these image pickup devices, which is synchronously scanned by the sync signal from the external sync signal generator, is based on the sync signal from the external sync signal generator in a required section in the same field by the video signal synthesizer. Switched to 1
It is combined into a video signal corresponding to one image for each field and given to the image processor as a combined video signal. The image processor sets the image based on the composite video signal from the video signal combiner to 1 based on the sync signal from the external sync signal generator.
Analysis processing is performed for each field to obtain required measurement information for a plurality of target objects.

[0023]

[Examples] The present invention will be described below in the three-dimensional shape measurement of a tunnel wall surface in a railway, which is performed by using the optical cutting method.
An example in which the present invention is applied to position detection of specific parts of left and right rails will be described. FIG. 1 is a perspective view showing an overall configuration of a three-dimensional shape measuring apparatus for a track peripheral structure to which an object measuring image processing apparatus according to an embodiment of the present invention is applied, and FIG. 2 is an object measuring object shown in FIG. It is a figure for explaining the composition of an image processing device. It should be noted that parts that are substantially the same as the configuration of the apparatus shown in FIG. 7 above are given the same reference numerals, and descriptions thereof will be omitted.

1 and 2, the object measuring image processing apparatus of this embodiment has a flat plate shape that spreads radially toward the wall surface of the tunnel T and the rails RR and RL in a plane substantially orthogonal to the rails RR and RL. A light source 2 for projecting the light S and a right rail C for capturing an image of a light cutting line generated on the right rail RR by the flat light S from the light source 2 as an optical image of the target object.
As with the CD camera 11R, as an optical image of the target object,
Left rail CCD camera 11L for picking up an image of a light cutting line generated on the left rail RL, and a rail image in which an external synchronization signal generator 12, a video signal synthesizer 13 and an image processor 14 described later are built in. And the processing device main body 15. The left and right rail CCD cameras 11R,
11L is fixed to the movable carriage 1 such that its optical axis intersects with the surface of the flat light S at an angle θ ₂ = 45 degrees.

Reference numeral 6 is a computer, and 7 is a structure image processing apparatus. The structure image processing device 7 captures an image of the light cutting line generated on the wall surface of the tunnel T by the flat light S from the light source 2, and the light cutting line image obtained for each field by the structure CCD camera 3 Is analyzed and the two-dimensional coordinates (two-dimensional position coordinates) of the light cutting line indicating the wall shape of the tunnel T on the image are obtained. Further, the computer 6 uses the two-dimensional coordinates obtained by the structure image processing device 7 and the insides of the upper surfaces of the left and right rails RR, RL obtained by the image processing device 14 of the object measurement image processing device described later. Based on the two-dimensional coordinates (two-dimensional position coordinates) indicating the position of the corner, as described above, the structure orthogonal coordinate system O-XYZ
The three-dimensional coordinates of the light cutting line indicating the wall shape of the tunnel T are calculated with the midpoint between the rails RR and RL (gauge) as a reference point in accordance with the traveling of the moving carriage 1.

The external synchronizing signal generator 12 outputs the synchronizing signal Ssysc.
The synchronizing signal Ssysc is given to the left and right rail CCD cameras 11R and 11L for synchronous scanning, and the video signal synthesizer 13 and the image processor 14 also synchronously control them. Given for. The structure CCD camera 3 is also designed to scan in synchronization with the rail CCD cameras 11R and 11L.

The video signal synthesizer 13 is a CC for the left and right rails.
The video signals from the D cameras 11R and 11L are switched based on the sync signal Ssysc from the external sync signal generator 12 in a required section in the same field to synthesize a video signal corresponding to one image per field. The composite video signal VSsynth corresponding to one image is supplied to the image processor 14. The video signal synthesizer 13 can be easily configured by using, for example, an analog switch.

The image processor 14 analyzes the image of the composite video signal VSsynth from the video signal synthesizer 13 for each field (for each image) based on the synchronization signal Ssysc from the external synchronization signal generator 12. , Two-dimensional coordinates of the corners inside the upper surfaces of the left and right rails RR, RL as required measurement information for a plurality of target objects.

The operation of the object measuring image processing apparatus configured as described above will be described below with reference to FIGS. 2, 3 and 4. FIG. 3 shows an example in which the images captured by the two rail CCD cameras shown in FIG. 2 are divided in the direction (v direction) perpendicular to the horizontal scanning line, and these are combined into one image. Figure for illustration, Figure 4
Is 2 in the video signal synthesizer in the example shown in FIG.
It is a figure for demonstrating each video signal from the CCD camera for rails of a stand, and its synthetic | combination video signal.

The rail CCD cameras 11R and 11L are horizontally and vertically synchronously scanned by a synchronizing signal Ssysc from an external synchronizing signal generator 12, of which the right rail CCD camera 11R is driven by a flat light S from a light source 2. Right rail RR
The image of the light cutting line generated above is taken, and the CCD for the left rail
The camera 11L captures an image of the light cutting line generated on the left rail RL. The video signal synthesizer 13 counts the number of horizontal synchronizing signals of the synchronizing signals Ssysc from the external synchronizing signal generator 12 in the same field, and activates an analog switch (not shown) at the 120th scanning line to synthesize them. The connection between the video signal output terminal and the video signal input terminal from the right rail CCD camera 11R is changed from the on state to the off state, and the combined video signal output terminal and the left rail CCD camera are connected.
Connect to the video signal input terminal from 11L. as a result,
The video signal synthesizer 13, as shown in FIG. 4, VS _R, VS _L is taken as a video signal from the CCD camera 11L for the left rail as a video signal from the CCD camera 11R for the right rail, these video signals 1 when VS _R and VS _L are combined
The composite video signal VSsynth corresponding to the image is provided to the image processor 14. Such an operation is performed for each field.

As described above, in the image processor 14, as shown in FIG. 3, the video signals (video information) of the rail CCD cameras 11R and 11L in the non-effective field of view NF shown by the hatched portions are discarded. Then, a combined video signal VSsynth corresponding to one image obtained by synthesizing the video signals of the remaining effective visual fields EF, which captures the image portions of the optical cutting lines generated on the left and right rails RR and RL.
Will be given. 3 and 4, D indicates an image dividing line.

Composite video signal VS from the video signal synthesizer 13
When synth is input, the image processor 14 performs pre-processing such as noise removal on the image of the composite video signal VSsynth to detect the position of the brightness level on each horizontal scanning line that changes to a predetermined value or more. By performing an analysis process for extracting the characteristics of the pattern, the two-dimensional coordinates of the light cutting line indicating the position of the upper inside corner PR of the right rail RR and the light cutting indicating the position of the upper inside corner PL of the left rail RL. The two-dimensional coordinates of the line are calculated at the video rate (1/60 seconds). Such an operation is performed by the synchronization signal S from the external synchronization signal generator 12.
It is performed for each field (for each image) based on sysc.

In this case, the field of view in the v direction (direction perpendicular to the horizontal scanning line) is not limited to the rails RR and RL even if the moving carriage 1 is shaken.
If a margin of ± 20 mm is set so that the lens always fits within the field of view, the rail width is 127 mm, so [127 + (2
X20)] x2 = 334 mm. Also, the u-direction (horizontal scanning line direction) field of view is 628 mm from the calculation based on the fact that the optical axes of the rail CCD cameras 11R and 11L are arranged at an angle of 45 degrees to the plane of the flat light S. Becomes Therefore, u
Since the number of pixels in the direction is 512 and the number of pixels in the v direction is 240, u
The detection resolution Δu in the direction is Δu = 628/512 = 1.2 mm, and the detection resolution Δv in the v direction is Δv = 334/240 =
It will be 1.4 mm.

As a result, it is possible to obtain the positional information of the corners inside the upper surfaces of the left and right rails RR and RL with the measurement accuracy which is improved about four times as compared with the conventional one, and one image processor 14 can be used. Since the required image processing can be performed in real time, it is not necessary to provide an image processor for each of the two rail CCD cameras 11R and 11L, and a large increase in the cost of the apparatus can be avoided. The data of the two-dimensional coordinates obtained by the image processor 14 and the data of the two-dimensional coordinates obtained by the image processing device for structure 7 are taken into the computer 6, and the computer 6 causes the traveling of the mobile carriage 1 to travel. According to the rail
The three-dimensional coordinates of the light cutting line showing the wall shape of the tunnel T are calculated with the midpoint between RR and RL (trajectory) as a reference point.

In the above embodiment, the example of the image in which the video signals from the two rail CCD cameras are switched in the v direction vertical to the horizontal scanning line and the divided ones in the v direction are combined is described. It is also possible to combine the images divided in the horizontal scanning line direction as one image. Figure 5
2 is a diagram for explaining an example in which an image captured by each of the two rail CCD cameras shown in FIG. 2 is divided in the horizontal scanning line direction (u direction), and these are combined into one image. FIG. 6 is a diagram for explaining each video signal from the two rail CCD cameras in the video signal synthesizer and the synthesized video signal in the example shown in FIG.

In this embodiment, the video signal synthesizer 13 detects the horizontal synchronizing signal of the synchronizing signals Ssysc from the external synchronizing signal generator 12 in the same field, and detects one horizontal scanning line for each horizontal scanning line. At a pixel address of 256, an analog switch (not shown) is operated to change the connection between the composite video signal output terminal and the video signal input terminal from the left rail CCD camera 11L from the ON state to the OFF state, and the composite video signal output terminal. And the video signal input terminal from the CCD camera 11R for the right rail are connected. As a result, as shown in FIG. 6, the video signal synthesizer 13 outputs VS _L as a video signal from the left rail CCD camera 11L and VS _R as a video signal from the right rail CCD camera 11R.
Are taken, these video signals VS _L, VS _R is supplied to the image processor 14 as a composite video signal VSsynth corresponding to one image are combined. Such an operation is performed for each field.

When the above-mentioned synthesized video signal VSsynth from the video signal synthesizer 13 is input, the image processor 14 performs the luminance on each horizontal scanning line on the image by the synthesized video signal VSsynth in the same manner as described above. By performing the analysis process to extract the characteristics of the pattern at the position where the level changes to a predetermined value or more, the two-dimensional coordinates of the optical cutting line indicating the positions of the inner corners PR and PL of the left and right rails RR and RL are obtained. , Video rate (1/60 seconds). Such an operation is performed for each field (for each image) based on the synchronization signal Ssysc from the external synchronization signal generator 12.

In this case, in the u-direction (horizontal scanning line direction) visual field, if a margin of ± 20 mm is set so that the rails RR and RL are always within the visual field even if the moving carriage 1 is shaken, the rail width is 127. Since it is mm, [127 + (2 x 20)]
× 2 = 334 mm. In addition, the view in the v direction is calculated based on the fact that the optical axes of the rail CCD cameras 11R and 11L are arranged at an angle of 45 degrees with respect to the plane of the plate-like light S.
It will be 1 mm. Therefore, since the number of pixels in the u direction is 512 and the number of pixels in the v direction is 240, the detection resolution Δu in the u direction is
Δu = 334/512 = 0.7 mm, and the detection resolution Δv in the v direction is Δv = 371/240 = 1.5 mm.

As a result, similar to the previous embodiment, it is possible to obtain the positional information of the corner portions inside the upper surfaces of the left and right rails RR and RL with higher measurement accuracy than in the prior art.
Since the required image processing can be performed in real time with the single image processing device 14, it is not necessary to provide an image processing device for each of the two rail CCD cameras 11R and 11L, and a large increase in the cost of the device is avoided. You can

The video signal synthesizer 13 shown in the above embodiments
Can also be configured by a digital circuit. In that case, an AD converter for converting the video signals from the rail CCD cameras 11R, 11L into digital signals and a demultiplexer instead of the analog switch are provided. It can be configured by using it. Although the object measuring image processing apparatus according to the present invention has been shown in the above-described embodiment as an example applied to the position detection of the specific portions of the left and right rails in the railroad by utilizing the optical cutting method, Also, various applications are possible.

[0041]

As described above, according to the object measuring image processing apparatus of the present invention, the light source for projecting light to a plurality of target objects existing at intervals, the external synchronization signal generator, and the external synchronizing signal generator are provided. A plurality of image pickup devices, which are synchronously scanned by a sync signal from a sync signal generator, respectively picking up optical images of the plurality of target objects for each target object, and video signals from the plurality of image pickup devices, in the same field. A video signal synthesizer that switches in a required section of the above based on the sync signal from the external sync signal generator, synthesizes a video signal corresponding to one image for each field, and outputs as a synthesized video signal; An image obtained by the composite video signal from the video signal synthesizer is analyzed for each field based on the synchronization signal from the external synchronization signal generator, and Since it is equipped with an image processor that obtains necessary measurement information, one image processor can perform the required image processing in real time without providing each image processor for each of a plurality of imaging devices. Therefore, it is possible to obtain required measurement information with improved measurement accuracy for a plurality of target objects existing at intervals without causing a large increase in the cost of the device.

[Brief description of drawings]

FIG. 1 is a perspective view showing an overall configuration of a three-dimensional shape measuring apparatus for a track peripheral structure, to which an object measuring image processing apparatus according to an embodiment of the present invention is applied.

FIG. 2 is a diagram for explaining the configuration of the image processing apparatus for object measurement shown in FIG.

FIG. 3 shows images captured by the two rail CCD cameras shown in FIG. 2 in a direction (v
It is a figure for explaining an example in the case of dividing by (direction) and combining them into one image.

FIG. 4 is a diagram for explaining each video signal from two rail CCD cameras in the video signal synthesizer and the synthesized video signal in the example shown in FIG. 3;

FIG. 5 is a view for explaining an example of a case where the images respectively captured by the two rail CCD cameras shown in FIG. 2 are divided in the horizontal scanning line direction (u direction) and they are combined into one image. FIG.

FIG. 6 is a diagram for explaining each video signal from the two rail CCD cameras in the video signal synthesizer and the synthesized video signal in the example shown in FIG. 5;

FIG. 7 is a diagram showing an overall configuration of a three-dimensional shape measuring apparatus for a line peripheral structure according to a conventional technique.

8 is a perspective view showing an outline of the three-dimensional shape measuring apparatus shown in FIG.

9 is a diagram for explaining a three-dimensional shape measuring method by the three-dimensional shape measuring apparatus shown in FIG.

10 is a diagram for explaining a method for obtaining a midpoint between rails by the three-dimensional shape measuring apparatus shown in FIG.

11 is an explanatory diagram of an image of a light cutting line generated on a tunnel wall surface, which is obtained by the structure CCD camera shown in FIG. 7. FIG.

12 is an explanatory diagram of an image of a light cutting line generated on a rail, which is obtained by the rail CCD camera shown in FIG. 7. FIG.

FIG. 13 is a diagram for explaining the field of view of the imaging device in the case of imaging a railway rail that is mainly used in Japan in the three-dimensional shape measuring device shown in FIG. 7.

[Explanation of symbols]

1 ... Movable carriage 2 ... Light source 3 ... CCD camera for structure
6 ... Computer 7 ... Image processing device for structure 11R ... CCD camera for right rail 11L ... CCD camera for left rail
12 ... External sync signal generator 13 ... Video signal synthesizer 14 ...
Image processor 15 ... Image processor for rails RR, RL ...
Rail S ... Flat light T ... Tunnel

Front page continuation (72) Inventor Yuichiro Goto 1-5-5 Takatsukadai, Nishi-ku, Kobe-shi, Hyogo Inside Kobe Steel Co., Ltd., Kobe Steel Co., Ltd. (72) Inspector Yamaguchi 1 Takatsukadai, Nishi-ku, Kobe-shi, Hyogo Prefecture 5-5-5 Kobe Steel Research Institute, Kobe Steel Co., Ltd. (72) Inventor Yasuharu Kami 1-5-5 Takatsukadai, Nishi-ku, Kobe-shi, Hyogo Prefecture Kobe Steel Research Institute, Kobe Steel Co., Ltd. (72) Inventor Yoshimoto Nishimoto 1-5-5 Takatsukadai, Nishi-ku, Kobe-shi, Hyogo Prefecture Kobe Steel Works, Ltd. Kobe Research Institute

Claims (1)

[Claims] 1. A light source for projecting light on a plurality of target objects existing at intervals, an external synchronization signal generator, and synchronous scanning by a synchronization signal from the external synchronization signal generator,
A plurality of image pickup devices for picking up the optical images of the plurality of target objects for each target object, and video signals from the plurality of image pickup devices from the external synchronization signal generator in a required section in the same field. The video signal synthesizer for switching and synthesizing the video signal corresponding to one image for each field for output as a synthesized video signal, and the image by the synthesized video signal from the video signal synthesizer 1 based on the sync signal from the external sync signal generator
An image processing device for object measurement, comprising: an image processor that performs analysis processing for each field and obtains required measurement information about the plurality of target objects.