JPS6125081B2 - - Google Patents

Description

Detailed Description of the Invention The present invention captures an image of an object to be measured using a television camera, displays the object image on a screen such as a video monitor using the video signal, and displays a plurality of cursor lines on the screen. This invention relates to a tilt imaging correction method for an apparatus that simultaneously displays a cursor line with a predetermined part of a subject image, electrically measures the interval between each cursor line, and measures dimensions between the predetermined parts of the subject. It is.

FIG. 1 is a block diagram of a conventional object dimension measuring device, in which a television camera (hereinafter referred to as camera) 1
A subject 4 on a plane orthogonal to the imaging optical axis 3 of the lens 2 is imaged by the camera 1, and the video signal output is processed by the camera controller 5 and then sent to the dimension measuring device 6, where the cursor signal is converted into an image. The signal is synthesized and sent to a display 7 such as a video monitor, and cursor lines 10A and 10B parallel to the subject image 9 are displayed on the screen 8, and these cursor lines 10A and 10B are By the operation, the cursor lines 10A and 10B are moved in the direction of the interval between them, that is, in the example shown in the figure, in the left and right direction, and the cursor line 1
0A, 10B to match a predetermined part of the subject image 9, and then cursor lines 10A, 1 at this time.
The distance between 0B is determined electrically using the dimension measuring device 6,
Dimensions between predetermined parts of the subject 4 are being measured.

Incidentally, in many cases, the outline of the subject image 9 is used as the predetermined part of the subject image 9 where the cursor lines 10A, 10B are made to match. In order to control the camera, a rotation mechanism 11 that is driven by a motor and rotates in forward and reverse directions is added to the camera 1 to maintain a constant relationship between the outline of the subject 4 and the rotation angle of the camera 1 with respect to the imaging optical axis 3. The rotation controller 12, which generates drive output based on various signals from the device 5, rotates the motor of the rotation mechanism 11, and images are taken so that the cursor lines 10A, 10B and the outline of the subject image 9 are in a parallel relationship. The camera 1 is rotated around an optical axis 3.

That is, as shown in FIGS. 2 to 4A, the relationship between the subject 4 and the imaging perpendicular 21 of the camera 1 is shown, and FIG. For example, as shown in FIG. 3B, the subject image 9 is displayed tilted to the right with respect to the screen perpendicular 22, and as shown in FIG.
In relation to imaging, the display is tilted to the left as shown in Figure B, so in order to display the images in Figure 4 A and B,
It is necessary to rotate the camera 1 in the forward direction F in FIG. 3A, rotate the camera 1 in the reverse direction R in FIG. 3A, and stop the camera 1 after reaching the state shown in FIG. 4A.

Therefore, the rotation controller 12 shown in the block diagram in FIG. 5 is used, which drives the motor of the rotation mechanism 11 and controls the camera 1 to achieve the imaging relationship shown in FIG. 4A. It is automatically rotated.

In FIG. 5, a video signal VIDEO, a vertical drive signal VD, and a horizontal drive signal HD are given from the camera controller 5, and based on these signals, FIG. 6 shows the waveforms of each part in FIG. It is designed to operate according to the time chart.

That is, in FIGS. 2 to 4B, blocks are provided in order to perform sampling at multiple locations in the direction intersecting the cursor lines 10A and 10B, and the blocks allow sampling locations 2
3A, and the block performs sampling corresponding to the sampling point 23B, and the block's pulse generator PG _1A , which uses a monostable multivibrator, etc., is connected to the vertical position 101 shown in FIG. It is driven by the leading edge of the drive signal VD and generates a pulse 102A with a pulse width t _1A determined by a resistor R _1A and a capacitor C _1A , and its trailing edge sets a flip-flop circuit (FFC) FF _1A . are doing.

Since the FFC/FF _1A is reset by the leading edge _of the horizontal drive signal HD shown as ₁₀₃ in FIG. The pulse generator PG _2A of is driven, and the resistor R _2A ,
A pulse 105A of pulse width t _2A determined by capacitor C _2A is generated, and FFC.FF _3A is set by the leading edge of pulse 104A.

Pulse 105A is FFC/FF _2A due to its trailing edge
However, FFC FF _2A is reset by the leading edge of the horizontal drive signal HD, shown at 103, producing a pulse 106A which resets FFC FF _3A , which was previously set by the trailing edge.

Then, the output 107A of FFC・FF _3A is the 6th
As shown in the figure, the pulse width is 1H (1H is the period from the start of one horizontal scanning line to the next horizontal scanning line), and this is applied to the input of gate _G1A using a NAND gate or the like.

On the other hand, the video signal VIDEO is shown in Figures 2 to 4B.
When the subject image 9 is shown in FIG. In addition to extracting the level higher than the video component V as the comparison output 109,
It is given to _G2A using NAND gate etc.

Further, the other input of the gate G _1A is given the horizontal drive signal 110 via the inverter INV ₁ , and the output 107 of the FFC/FF _3A and the horizontal drive signal 110 are both at a high level (hereinafter referred to as “ When the output becomes "H"), its output 111A becomes a low level (hereinafter referred to as "L"), and this is applied to the other input of the gate _G2A , so that the output 112A of the gate _G2A also has both inputs 109 and 111A " It becomes "L" only when it becomes "H", and the output becomes "1" under other conditions.
12A becomes "H" and is applied to gate _G3A such as a NAND gate.

Gate G _3A has FFC/FF _3A output 107A.
is also given, and when both inputs are "H", its output 113A becomes "L", and as shown in FIG. The subsequent video component V is extracted.

This output 113A is given to the FFC/FF _4A which repeats setting and resetting according to the falling edge of the input, and is set by the leading edge of the horizontal drive signal HD and reset by the leading edge of the video component V.
The output 114A of the FFC/FF _4A has a pulse width corresponding to the distance d ₁ between the horizontal scanning start point of the sampling point 23A and the outline of the subject image 9 in FIG. 2B, and is sent to the delay circuit DL.

Further, a similar operation is performed in the block, and the output 114B of the FFC/FF _4B is a pulse width corresponding to the distance d ₂ between the horizontal scanning start point of the sampling point 23B and the outline of the subject image 9 in FIG. 2B. I get the output. However, the position of the sampling point 23A on the screen 8 is determined by the pulse width t _1A of the pulse _102A generated by the block's pulse generator PG 1A, whereas the pulse generated by the block's pulse generator PG _1B determines the position of the sampling point 23A on the screen 8. 1
The position of the sampling point 23B is determined by the pulse width t _1B of 02B, and there is an interval of n horizontal scanning lines, that is, nH, between the sampling points 23A and 23B.

The delay circuit DL has a delay time equivalent to the aforementioned nH, and by passing through this delay circuit DL, the output 114A of the FFC/FF _4A becomes a pulse 1.
16, and as shown in Fig. 6, the leading edge of the output 114B of FFC/FF _4B and the pulse 116 match, making it easy to detect the difference in pulse width between the two, and the difference between the pulse widths of the two and the cursor line 10A. The tilt condition of the subject image 9 is determined.

That is, if the pulse widths d ₁ and d ₂ of both pulses are the same, the cursor line 10A and the outline of the subject image 9 will be in a parallel state as shown in FIG. 4B, whereas the pulse width d ₁ of the pulse 116 is the pulse width of output 114B
If d is larger than ₂ , the state shown in FIG. 2B is shown, and if it is the opposite, the state shown in FIG. 3B is shown.

Therefore, the pulse 116 and the pulse 115 obtained by inverting the output 114B by the inverter INV ₂ are
If given to gate _G4 such as a NAND gate, pulse 1
When the pulse width d ₁ of the pulse 16 is smaller than the pulse width d ₂ of the pulse 115, that is, in the state shown in FIG. 3B, the output 117 of the gate G ₄ becomes "H" and maintains this state.

A pulse generator to which this output 117 is given
PG ₃ uses a monostable multivibrator or the like that is driven by the change from "H" to "L", and its output 118 is always "H", so this output 118 causes relay RL to be activated. is in working condition.

For this reason, the white make contacts of the relay contacts RLa to RLd are on, and the gate _G5 of the NAND gate etc. receives the pulse 119 obtained by inverting the pulse 116 by the inverter _1NV3 , and the output 114B of the FFC/FF _4B . is given, and its output 120 becomes "L" when both inputs are "H", and the pulse width of both
The difference between d ₁ and d ₂ is detected.

This pulse width difference is detected by the detector DET and becomes a voltage in a specific direction, and is amplified by the amplifier A and then applied to the motor M. However, since the relay contacts RLc and RLd are operating, the polarity of the arrow R is This causes the rotating mechanism 11 to move to the third position.
Rotate in the opposite direction R in figure A.

In contrast to the above, the state shown in FIG. 2B, that is,
Pulse 116 than output 114B of FFC/FF _4B
If the pulse width d ₁ of is large, 114 shown in FIG.
BR, 115R, and 117R, the output 117R of the gate _G4 changes to "L", thereby driving the pulse generator _PG3 , and the resistor _R3 ,
An "L" pulse 118R is generated having a pulse width of approximately 1V (1V is the period from the beginning of a field to the next field) determined by the capacitor _C3 .

Then, relay RL is restored and this time pulse 11 is sent to gate _G5 via the break contact indicated by the black mark.
Since 5R and 116R are given, the difference in pulse width between the two is detected as the output 120R and this is given to the motor M in the same way as described above. However, at this time, relay contacts RLc and RLd have been restored.
Even if the output polarity of the amplifier A does not change, the output of the amplifier A is applied to the motor M with the polarity indicated by the arrow F, thereby causing the rotating mechanism 11 to rotate in the forward direction F in FIG. 2A.

If the imaging perpendicular 21 becomes the state shown in FIG. 4A due to the rotation in the forward direction F in FIG. 2A and the rotation in the reverse direction R in FIG. The outline of the subject image 9 is in a parallel relationship, and the output of FFC/FF _4B is 114B.
Since the pulse widths d ₂ and d ₁ of and pulse 116 are equal, the output 120 or 120R of gate G ₅
When "L" disappears, the motor M does not rotate, and the rotating mechanism 11 rotates to the state shown in FIG. 4A and then stops, and the relationship shown in FIG. 4B is automatically set, and the cursor line It becomes possible to measure dimensions using 10A and 10B.

As described above, the rotation controller 12 operates and automatically controls the rotation of the camera 1, but the sampling points 23A and 23B have a pulse width t generated by the pulse generators PG _1A and PG _1B as described above. _1A ,
Since it is fixedly determined by t _1B , if
When the outline of the subject image 9 that coincides with the sampling points 23A and 23B is protruding or recessed as shown in FIGS. 7A and 7B, the rotation of the camera 1 is controlled based on this, and the cursor line 10A is The camera 1 stops while the object image 9 is not parallel to the outline of the object image 9, resulting in inaccurate dimension measurement.

The present invention aims to fundamentally eliminate such drawbacks of the conventional art, and in the object dimension measuring device equipped with the above-mentioned tilt imaging correction function, each sampling location is displayed as a bright line on the screen, and
The inclination of a highly rational object dimension measurement device that allows the interval between each bright line to be set arbitrarily and controls the rotation of the camera based on the distance from the horizontal scanning start point to the intersection of each bright line and the outline of the object image. An imaging correction method is provided.

The details of the present invention will be explained below with reference to FIG. 8 and subsequent figures showing embodiments.

FIG. 8 shows the configuration, in which bright lines 31A and 31B representing sampling points are displayed on the screen 8 of the display 7, and a bright line signal for displaying the bright lines 31A and 31B is generated and sent to the camera controller 5. A rotation controller 32 is used which has the function of giving a signal and a switch SW for setting the interval between the bright lines 31A and 31B, and other than that, the configuration is the same as that of FIG. 1.

FIG. 9 is a block diagram of the rotation controller 32 in FIG. 8, which operates in accordance with the time chart in FIG. 10 showing the waveforms of each part in FIG.

That is, the pulse generator PG ₁₁ is driven by the leading edge of the vertical drive signal VD shown at 101 in FIG.
A pulse 102 is generated to determine the display position of the bright line 31A, and its trailing edge sets FFC/FF ₁₁ , while the leading edge of the horizontal drive signal HD shown as 103 in FIG. 10 resets FFC/FF ₁₁ . , an output 104 is obtained, and its trailing edge drives the pulse generator PG ₁₂ , producing a pulse 1 of pulse width t ₁₂ .
05 and output 1 of FFC/FF ₁₁ .
The trailing edge of pulse _{105 sets FFC/FF 13 ,} which is set by the trailing edge of pulse 105 and reset by the leading edge of horizontal drive signal HD _. Output 106
It is reset by the trailing edge at , and an output 107 with a pulse width of 1H is obtained.

However, as pulse generators PG ₁₁ and PG ₁₂ ,
_Similar to _the pulse generators PG _1A and PG _2A in FIG _. uses a variable resistor VR ₁₁ , whereby the pulse width t ₁₁ can be varied and the display position of the bright line 31A can be freely changed.

In addition, a video signal shown as 108 in FIG.
VIDEO, as in Fig. 5, only the video component V is extracted by the comparator CP, and the output becomes 109.
The fifth input is applied to gate G _12A , such as a NAND gate, and is applied to the other input from similar gate G _11A .
By interacting with the output 111 obtained by the same operation as shown in the figure, it becomes an output 112 and is applied to a gate _G13A such as a NAND gate.

The output 107 of FFC/FF ₁₃ is given to the other input of gate G _13A , thereby obtaining the horizontal drive signal HD and the following video component V as output 113 of gate G _13A . Yes, in Figure 5
FFC/FC _14A , which is similar to FFC/FF _4A , is set by the leading edge of the horizontal drive signal HD and reset by the leading edge of the video component V, and the interval d ₁ in FIG. 3B is set.
generates an output 114 with a pulse width corresponding to .

On the other hand, the frequency of the horizontal drive signal HD shown as 103 in _FIG.
~DV ₇ divides the frequency in steps of 1/2 to 1/128, and each of these divided outputs is selected by the selector.
Select and take out with SEL, shift pulse SP
to shift register SR ₁ as
The output 107 of FFC/FF ₁₃ applied to the data input DI of SR ₁ is shifted by the shift pulse SP and then sent out as shift output 115.

Note that the selector SEL is controlled by the switch SW to select the input, and assuming that the selected frequency division output is 1/m and that the shift output 115 is generated by one shift pulse SP, the horizontal drive signal Shift output 1 when only m HD occurs
15 occurs, and as shown in Figure 10, for the output 107
A shift output 115 delayed by mH is obtained, and this delay amount mH can be freely set by operating the switch SW.

This shift output 115 is applied together with the output 110 of the inverter INV ₁ to a gate G _11B such as a NAND gate, and its output 116 is a horizontal drive signal.
HAA is extracted, and this, together with the output 109 of the comparator CP, is connected to the gate G ₁₂ using a NAND gate, etc.
given to _B.

Then, a signal obtained by inverting the horizontal drive signal HD and the following video component V is obtained at the output 117 of the gate _G12B.If this is applied to the similar gate _G13B together with the shift output 115, the output 117 is inverted. If FFC/FF _14B , which is similar to FFC/FF _14A , is controlled using this, the interval shown in Figure 3B is obtained.
A pulse with a pulse width corresponding to d ₂ is obtained as output 119.

That is, the sampling point 23A in FIG. 2B is determined by the pulse width t ₁₁ of the pulse 102 generated by the pulse generator PG ₁₁ , and the second
The sampling point 23B in Figure B has been determined, and since the sampling point 23B can be freely selected by operating the switch SW, both sampling points 23B
The interval between A and 23B can be set arbitrarily.

Furthermore, the output 107 of the FFC/FF ₁₃ and the shift output 115 of the shift register SR ₁ are combined by a mixer MIX ₁ to become a bright line signal 120, and then combined with the video signal VIDEO in a mixer MIX ₂ to produce a combined output. 121 and video signal output
It is sent to the camera controller 5 as VIDEO OUT.

Therefore, this video signal output VIDEO OUT
is given to the display 7 via the dimension measuring device 6, and the sampling points 23A and 23B are displayed by bright lines 31A and 31B as shown in FIG.
It is shown whether it is A or 23B.

On the other hand, the output 114 of FFC/FF _14A is given to the data input DI of shift register SR ₂ , and this register SR ₂ also performs the same operation with the same shift pulse SP as shift register SR ₁ , so it is output. 114 is delayed by a period of mH and becomes the shifted output 122, thereby causing the FFC
The leading edge matches the output 119 of FF _14B , and both outputs 11
If the pulse width difference between pulse widths d ₂ and d ₁ between pulse widths 9 and 122 is detected, the motor M is controlled in the same manner as in FIG. If the rotation mechanism 11 is in the state shown in FIG. 2A, the rotation mechanism 11 can be rotated in the forward direction F.

In addition, the inverter that performs these controls
INV ₂ , INV ₃ , and subsequent operations are the same as those shown in FIGS. 5 and 6, but the shift outputs 112 and after in the state of FIG. 2 A are shown in FIG. 10 as 122R to 127R.
It is shown as.

Therefore, the position of the bright line 31A in FIG. 8 is determined by the variable resistor VR ₁₁ , and the switch
If the position of the bright line 31B is set by SW and the positions of both bright lines 31A and 31B are selected while avoiding the protrusion or recess of the subject image 9 shown in FIGS. 7A and 7B,
The rotation of the camera 1 is controlled with reference to the normal outline of the object image 9 regardless of the recess, so that accurate dimension measurements can be performed at all times.

Note that the illustrated configuration can be modified in various ways, such as programmable counters or the like may be used for _the frequency dividers DV ₁ to DV ₇ in FIG. 9, and multi-stage shift registers may be used for the shift registers SR 1 and SR ₂ . The display of the bright lines 31A and 31B can also be arbitrarily selected depending on the situation of the screen 8, such as a dark line, an inverted bright line, or a dotted line.

As is clear from the above description, according to the present invention, the sampling location of the subject image is displayed by brightness, and the location can be freely selected, so the accurate imaging state is automatically determined, making it extremely easy and reliable. This makes it possible to measure the dimensions of various objects, and provides remarkable effects in measuring the dimensions of various objects.

[Brief explanation of the drawing]

Fig. 1 is a configuration diagram showing a conventional example of a subject dimension measuring device, Figs. 2 to 4 show the relationship between the imaging situation and the subject image, A is a diagram of the relationship between the subject and the imaging perpendicular of the television camera, and B 5 is a block diagram of the rotation controller in FIG. 1, FIG. 6 is a time chart showing the waveforms of each part in FIG. 5, and FIG. 7 is a diagram showing the subject image on the screen. 8 is a block diagram showing an embodiment of the present invention, FIG. 9 is a block diagram of the rotation controller in FIG. 8, and FIG. 10 is a diagram showing each part in FIG. 9. This is a time chart showing waveforms. 1... Camera (TV camera), 3... Imaging optical axis, 4... Subject, 5... Camera controller, 6...
Dimension measuring device, 7...Display device, 8...Screen, 9...
Subject image, 10A, 10B...cursor source, 11
... Rotation mechanism, 23A, 23B ... Sampling point, 31A, 31B ... Bright line, 32 ... Rotation controller, SW ... Switch, M ... Motor.

Claims (1)

[Claims] 1. An image of a subject is captured by a television camera, the image of the subject is displayed on the screen using the video signal, and a plurality of cursor lines parallel to each other and movable in mutually spaced directions are displayed on the screen, A cursor line is aligned with a predetermined part of the subject image, and the dimensions of the subject are electrically measured based on the interval between the cursor lines at this time, and the subject image is marked at a plurality of locations in a direction intersecting the cursor line. Rotating the television camera about the imaging optical axis of the television camera based on each interval between the outline of the subject image obtained by sampling and a point at which horizontal scanning for displaying the subject image is started; In a subject dimension measuring device equipped with an oblique imaging correction function that makes the cursor line and the outline of the subject image parallel to each other, each of the sampling locations is displayed as a bright line on the screen, and
The distance between each of the bright lines can be arbitrarily set, and the television camera is rotated based on the distance from the horizontal scanning start point to the intersection of each of the bright lines and the outline of the subject image. A method for tilted imaging of an object dimension measuring device.