JPS592484Y2 - Position dimension measuring device - Google Patents

Description

[Detailed Description of the Invention] The present invention is an apparatus for optically measuring the dimensions and position of an object to be measured using a charge transfer image sensor. The present invention relates to a measurement range selection circuit for eliminating the effects of disturbance objects and illumination on measurement that exist at both ends.

In recent years, charge transfer imaging devices such as CCDs (ChrgeCO-
Upled Device) or BBD (Bucke
t BrigadeDevice) is commercially available.

This has a signal coverage function and a scanning (transfer) function, and serially outputs signals photoelectrically converted by a plurality of elements as a series of pulse signals in synchronization with clock pulses.

This is converted into an on-off pulse train by level detection such as inputting it to a voltage comparator together with a reference voltage, and the number of on-pits or off-pits N is counted, and the bit interval d of the charge transfer image sensor and the magnification M of the optical system are calculated. The dimensions of the object to be measured can be measured by calculating L=N-d-M, or if the bit number on the charge transfer image sensor corresponding to this and the position of the reference point is No, then the reference point is It is used when measuring the distance from to the point to be measured, that is, the position P, by calculating P=(N-NO).d-M.

When performing such measurements in a place with many measurement constraints, such as a production site, there may be disturbances other than the object to be measured that are present or intruding into the measurement range, or if the object to be measured is If it is necessary to bring it to the center of the measurement range, measurement problems tend to occur at both ends of the measurement range due to illuminance.

For example, when measuring the distance between welded end faces immediately before welding an ERW steel pipe, a series of pulse signals converted into charge by a charge transfer imaging device is as shown in FIG.

In this case, the pulse signal a corresponding to the entire bit range of the charge transfer element is compared with the reference voltage using a voltage comparison circuit to determine whether it is on or off. Since the signal is small and at the same level as the signal between the welded end faces, the signals at both ends will be mistakenly counted as the signal between the welded end faces.

In other words, since the on-off pattern becomes unstable at both ends, the measurement accuracy not only decreases by the amount of bits that are incorrectly judged as on-off, but also means that you are measuring something completely different from what you intended to measure. Therefore, for use in production sites, it is highly desirable to add a function to the measuring instrument to eliminate the influence of such disturbances.

The present invention solves the problem caused by the disturbance in size and position measurement using a charge transfer element.The details of the present invention are shown in the drawings below. The measurement of dimensions between welded end faces will be explained as an example.

Figure 1 shows the distance (position) from the weld center line O to the weld end surface.
An example in which the noise removal circuit of the present invention is incorporated into a block circuit diagram for measuring dimensions l, l' and the dimensions between welded end faces is shown.

In this figure, 1 is an optical lens, 2 is a charge transfer image sensor in which a light receiving section is formed into multiple bits, and 3 is a charge transfer image sensor 2.
A preamplifier 4 amplifies the output signal of the preamplifier 3, and a sampling hold circuit 5 samples and holds each peak value of the output pulse signal of the amplifier 4 until the next peak arrives.

Reference numeral 6 denotes a reference voltage generation circuit that generates a threshold level voltage as a DC signal when converting the output signal of the sampling and holding circuit 5 into an on-off signal in the next level detector 7 according to its voltage level, and this voltage level By making the value variable from 0 to the saturation level of the charge transfer image sensor 2, the threshold level for converting into an on-off signal can be arbitrarily selected.

7 is a voltage comparator that compares the level of the output signal of the sampling hold circuit 5 and the DC signal of the reference voltage generation circuit 6 and converts it into an on/off signal depending on the magnitude thereof; Differential circuit to extract, 9 is a digital switch to set the bit number of welding center line O,
A preset counter 10 presets the set value of the digital switch 9 with a start pulse and then counts up clock pulses until the gate is closed by the output signal of the differentiating circuit 8.

11 is a buffer register that buffers the digital signal after the count of the preset counter 10 is completed; 12 is a D/A converter that converts the digital signal of the buffer register 11 into an analog signal; and 13 is a circuit that changes the period of the clock pulse to adjust the sensitivity. It is a clock pulse oscillator that can be adjusted.

A pulse generator 14 generates a start pulse every clock pulse number corresponding to the number of bits of the element 2, and is composed of a preset counter.

Reference numeral 15 denotes a signal processing device that receives and receives the digital signal from the buffer register 11 and processes the data as necessary, and 16 denotes a measurement range selection circuit of the present invention.

Next, the distance (position) from the welding center line to the welded end faces and the measurement of the dimensions between the welded end faces will be explained.

When the clock pulse g from the clock pulse generator 13 is applied to the charge transfer image sensor 2 and also input to the start pulse generator 14, the generator generates a start pulse h, which is input to the charge transfer element 2. The charge transfer element 2, on which the optical image of the steel pipe P is projected by the optical lens 1, outputs an output pulse signal a (Fig. 2) at a level proportional to the product of its illuminance and the scanning period at intervals of clock pulses g. .

This is amplified through a preamplifier 3 and an amplifier 4 to produce a signal a.
', a'', and the signal a'' (same as Fig. 2 a) is sampled and held in the sampling and hold circuit 5, timing the output pulse signal a'' with the clock pulse g, until the next clock pulse arrives. .

This output pulse signal is passed through the voltage comparator circuit 7 to the reference voltage C (
When the on-off signal d (Fig. 2 d) is compared with THL), both ends of the steel pipe become OFF signals as shown in the figure, and this changes the weld end face dimension or the weld end face position l, l from the weld center line O. ′, measurement errors occur.

The present invention eliminates measurement errors caused by disturbances that appear at both ends of the measurement range.Using the fact that the disturbances are located at both ends of the measurement range, the output pulse signal train from the charge transfer image sensor is At the stage where the level detector (voltage comparator 7) converts the raw on-off signal train into an on-off signal train, the measurement range from the lower limit bit number to the upper limit bit number on the charge transfer image sensor is determined by converting the raw on-off signal train. Leave it to me,
For other areas outside the measurement range from the first bit to the lower limit bit number and from the upper limit number to the final bit, it is possible to determine and select in each measurement whether the signal should be an on signal or an off signal. It removes the influence of objects.

This will be explained with reference to FIG. 3. This figure shows the details of the measurement range selection circuit 16, 21 is a lower limit preset counter,
22 is a preset counter for upper limit, 23 is a digital switch for lower limit, 24 is a digital switch for upper limit, 25
．． 27 is a knot circuit, 26, 28, 29 are AND gates, 30 is an OR gate, and 31 is an out-of-measurement-range on/off switch.

The input terminal to which the on/off signal d of one of the anime games 1 and 29 is applied and the output terminal of the anime game (or game) 30 are the main circuit 16.
The output terminal becomes the output terminal, and the clock pulse g and start pulse h are applied to the counters 21 and 22.

In this example, the number of bits of the charge transfer image sensor 2 is 100 to 99.
It's 9th.

At the start of scanning (transfer), the lower limit bit number and upper limit bit number selected by each 3-digit digital switch 23.23 are loaded into each preset counter 21.24 in synchronization with the start pulse h, and then the clock pulse g is Each time an input is made to both counters 21 and 22, the contents of the counters are decremented by l, and first, when the counter 21 becomes 0, it is understood that the lower limit bit number has been exceeded and it is within the measurement range. The range until the counter 22 reaches O is within the measurement range.

It can be seen that the counter 22 has exceeded the upper limit bit number of O and has returned to the outside of the measurement range, and this state continues until the final bit is scanned.

Therefore, the measurement range is within the measurement range while the preset counter 21 is in a count completion state and the preset counter 22 is counting.

Therefore, the signal m indicating that the measurement is within the measurement range is obtained by inverting the count completion signal of the preset counter 21 and the count completion signal of the counter 22 by a NOT gate 25, and inputting the signal to the AND gate 26, and outputs the signal m. formed by m.

A signal n indicating that it is outside the measurement range is formed as the output of the signal m inverted by a not gate 27.

Next, this signal m and the on-off signal d are combined into an AND gate 2.
9, a signal q is formed which has the original on-off pattern within the measurement range and is off outside the measurement range.

Then, when a signal in which either the on signal U or the off signal ■ is selected by the changeover switch 31 is inputted to the AND gate 28 together with the signal n, the selected signal is either on or off outside the measurement range. A signal r, which is an off signal, is formed within.

Next, by inputting the signal g and the signal r to the OR game 30, the required total on-off pattern e is obtained.

Next, the operation of the position and dimension measuring device incorporating this measurement range selection circuit 16 will be explained with reference to FIG.

First, regarding the position measurement system, when the output signal of the sampling hold circuit 5 and the reference voltage signal C generated by the variable reference voltage generation circuit 6 are input to the voltage comparator 7, the signal will either be higher than or lower than the reference voltage signal C. A signal d, which becomes an on signal or an off signal, is output.

This signal d is no longer a pulse signal, but a step voltage that changes on and off only when there is a change from a state higher than the reference voltage to a state lower or vice versa.

When this signal d is input to the differentiating circuits 8 and 8', when there is a change from an on signal to an off signal or vice versa, that is, the scanning on the charge transfer element is detected at the measurement point (welding material end face d, d').
A pulse is generated from the circuit 8, 8' only when a certain bit of the image on the charge transfer image sensor corresponding to the position is reached.

By incorporating the measurement range selection circuit 16 between the voltage comparator 7 and the differentiating circuits 8 and 8', there will be no change in the usage at both ends of the steel pipe due to the above-mentioned reason, and the welding material end faces d, d ' can be obtained accurately.

These pulses f and f' are sent to the preset counter 10.10'
, the preset counter counts the number of clock pulses from the start pulse h to the output signals f and f' of the differentiating circuits 8 and 8', and this result is recorded as the bit number N1. Indicates N2.

Next, from digital switch 9 to preset counter 1
When the start pulse h is input at 0.10', the complement of the bit number No. on the charge transfer image sensor at the reference point measured in advance is preset, and (N, -N
O), (N, -NO) are counted output signal i
, i'.

As soon as the preset counter 10.10' finishes counting, it must start counting for the next scan, so immediately after the count ends, the signals i and i' are transferred to the register 11.11' before starting the next scan. Get ready.

When the signals i, i' are buffered in the register 11, the output signals j, j' from this register 11.11', i.e. (N1
-No) and (N2-No) are digitally input to the signal processing device 15, 15', where the position of the measurement point is determined based on the bit interval d of the charge transfer image pickup device stored in advance and the optical system magnification M. In other words, the distance l of the measurement point from the reference point
, l' is l = (N, -N,) -d -M, l'
= (N2-N,) -d -H.

Note that if the magnification M of the optical system is constant within the measurement range, l is approximately proportional to (N1 - No) and l' is proportional to (N2 - N,), so the signals j and j' are 12. Since 12' is input to 12' and its output is an analog quantity proportional to l and l', the position of the measurement point can be recorded by adjusting the gain and recording it on recording paper. and can be used effectively.

Next, to explain the dimension measurement system, measurement range selection circuit 1
6 is shared with the position measurement system, and the on-off signal e, the start pulse h, and the clock pulse g, which are the output signals of this measurement range selection circuit 16, are input to the counter 10'', On-pit number or off-pit number N
Count.

Immediately after the counter 10'' finishes counting, it must start counting for the next scan, so immediately after the count ends, the signal i// is transferred to the register 11'' and preparations for the next scan begin.

When the signal i// is buffered in the register 11'', the output signal j//, that is, N, from the register 11'' is digitally input to the signal processing device 15'', where it is pre-stored in the charge transfer image sensor. From the bit interval d and the optical magnification M, the measurement dimension is calculated as L=N-d-M.

In addition, input the signal j// to the D/A converter 12", and the output "" is an analog signal proportional to L, so if you adjust the gale and record it on recording paper, you can record the measured dimensions. It can be taken.

Since the apparent sensitivity of the charge transfer image sensor can be increased by changing the scanning period of the lock pulse generator or the start pulse generator, it is convenient to make the period variable.

In addition, the lower limit bit number and upper limit bit number must be set appropriately depending on where on the charge transfer image sensor the image of the disturbance object and the image of the object to be measured are formed, and can be arbitrarily selected using a digital switch. I took it as a thing.

In addition, the reason for selecting an on signal or an off signal for complementation outside the measurement range is that the interface with the subsequent signal processing always has the same interface specifications, which requires special design of the subsequent signal processing. This is because there is no problem and it becomes easier.

As is clear from the above explanation, this measurement range selection circuit can eliminate the effects of poor illumination at both ends of the measurement range and the effects of disturbing objects, and can perform dimension and position measurements using a charge transfer image sensor. It has become possible to achieve high precision and stability.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a waveform diagram for explaining operation, and FIG. 3 is a block diagram of a measurement range selection circuit. In the drawing, 2 is a charge transfer image sensor, 7 is a voltage comparator for level detection, and 16 is a measurement range selection circuit.

Claims (1)

[Scope of utility model registration request] A charge transfer image sensor that projects an image of the measurement object, a comparison circuit that divides the output of the image sensor into a threshold and outputs a signal that turns on and off at the measurement point of the measurement object, and two preset counters. The bit values of the upper and lower limits of the measurement range are read in, and by counting down with a clock pulse applied to the charge transfer image sensor, it is determined whether it is within the measurement range or outside the measurement range.If it is within the measurement range, the output of the comparison circuit is A position and dimension measuring device comprising a measurement range selection circuit that selects a signal and, if it is outside the measurement range, selects and complements the selected fixed signal, either an ON fixed signal or an OFF fixed signal. .