JPH01219603A - Method and apparatus for measuring fine dimension - Google Patents

Abstract

PURPOSE:To measure a predetermined fine dimension within a specified range with high accuracy, by taking the time of a fine dimension region by an imaging camera to apply image processing thereto and counting the number of clock pulses in the time width of the dimension to be measured. CONSTITUTION:The fine dimension of an object 8 to be measured is taken out as a synthetic image signal by an imaging camera 2 and only the image signal of one horizontal scanning of the position corresponding to the fine dimension is taken out from said synthetic image signal to be stored in a memory 17 as an image signal A'. Subsequently, a CPU 7 reads the image signal A' to perform threshold value processing and binarizes the background and fine dimension region of the object 8 to be measured and counts the number of high speed clock pulses within the time width of the fine dimension to convert the same to an actual dimension. As a result, when a head gap width is measured, an arbitrary fine dimension can be measured with high accuracy while a range being wide to a specified degree is confirmed.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a method and apparatus for measuring minute dimensions such as gap width and azimuth angle of various heads of video tape recorders, floppy disks, etc. in the field of image processing.

Small dimensions such as the gap width of prior art heads are typically measured visually under magnification with an optical microscope, or with an electronic marker using an industrial television camera or the like.

When using an industrial TV camera, the gap width of the object to be measured is extremely narrow, so accurate measurements cannot be made unless the magnification is very high. However, when the magnification is increased, the field of view becomes narrower and It becomes difficult to search for the gap itself.

For example, if the optical magnification is 100x, the field of view is
ITV camera 273 inch size 60 x 60 [μm]
Since the resolution of the monitor is about 500 in terms of the number of scanning lines per frame, the resolution that can be expected is at most about 0.12 (μm).

OBJECT OF THE INVENTION Accordingly, an object of the present invention is to enable measurement of arbitrary minute dimensions with high precision in the field of image processing while confirming a relatively wide range.

Solution to the Invention In view of the above object, the present invention extracts minute dimensions of a measurement object as a composite image signal using an ↑most image camera, and extracts only an image signal of one horizontal scan at a position corresponding to the minute dimension from this composite image signal. is extracted, threshold processing is performed on the image signal, the background of the measurement target and the minute dimension region of the measurement target are binarized, and the final High speed black 7
The actual dimensions are calculated from the number of pulses.

With such technology, only specific minute dimensions can be measured with high precision while confirming the entire area of the measurement target, and the necessary processing can be performed on this measurement using a small storage capacity, and other processes can be performed as necessary. It is also possible to measure parts at the same time.

Structure of the Measuring Apparatus for Implementing the Method The measuring apparatus 1 of the embodiment shown in FIG. 1 is an example of implementing the method of the present invention in the field of a program.

Imaging camera 2, A-D converter 3, frame memory 4, D
-A converter 5 and monitor 6 are sequentially connected in series. The CPU 7 is connected to the frame memory 4, ROM 8, RAM 9, and synchronization separation section 10 in order to implement the measurement method of the present invention.

In addition, the imaging camera 2 passes through a vertical synchronization signal separation circuit 11 and a horizontal synchronization signal separation circuit 12 of the synchronization separation unit 10.
It is connected to a counter 13 and a count comparator 14. Further, the imaging camera 2 is connected to an A-D conversion circuit 15. This A/D conversion circuit 15 is connected on its other input side to a count comparator 14 and to a high-speed clock pulse generator 16, and on its output side to the CPU 7 via a memory 17.

Contents of the measurement method FIG. 2 shows each step of the measurement method of the present invention in chronological order.

First, the imaging camera 2 is configured to measure an object 18 such as a magnetic head.
The direction of the dimension to be measured is set as the horizontal scanning direction, and the image is converted into a composite image signal A (imaging process). Of course, as shown in FIG. 3, this composite image signal A includes a vertical synchronizing signal B for each frame, and also includes a number of horizontal synchronizing signals C corresponding to one frame. Then, this composite image signal A
is converted into a digital quantity by the A-reconverter 3,
The data is stored in the frame memory 4, converted into an analog quantity by the DA converter 5 as required, and displayed as an overall image on the monitor 6.

The vertical synchronization signal B of this composite image signal A is generated by the synchronization separation unit 1
It is taken out by the zero separation circuit 11 at the beginning of one frame, and becomes a reset signal to reset the counter 13 at the beginning of one frame. Furthermore, since the separation circuit 12 extracts the horizontal synchronization signal C from the composite image signal A of one frame and sends it to the counter 13,
The counter 13 extracts the horizontal synchronization signal C from the composite image signal A of one frame and counts the number thereof. - way, C
The PU 7 reads the composite image signal A from the frame memory 4, measures the position of the dimension to be measured from one frame as the number of horizontal scans through image processing, and calculates the number, for example, the signal E of r125J, by the counting comparator 1.
I am sending it to 4. Therefore, when the value of the counter 13 matches the horizontal scanning number signal E, the count comparator 14 generates a synchronizing signal F of a predetermined level, for example, "H" level, for one horizontal scanning period, and this @
The A-D conversion circuit 15 is set to an operable state during the high level period. For this reason, the A-D conversion circuit 1
5 extracts only the image signal of one horizontal scan in the dimension direction to be measured from the synthesized image signal A of one frame, and converts the image signal A' into, for example, in synchronization with the clock pulse G from the high speed clock pulse generator 16. It is converted into a 256-gradation signal and stored in the memory 17 (separate storage process).

After that, the CPU 7 executes the program based on the present invention, first reads out the image signal of one horizontal scanning period from the memory 17, and sets the density threshold value H in order to extract the area of the measured size from the background area. Then, in the time width of the measurement dimension area, dots (memory) corresponding to the number of clock pulses G are
(counting process), and finally converting the number into actual dimensions (calculation process).

Here, for example, if the clock pulse G is sampled at 60 (MH2), the effective portion of one horizontal scanning period can be divided into about 4000, and the resolution at that time can be expected to be about 1/4.000 of the visual field range. Now, if the optical magnification is 50 times, the field of view is 120 x 120 [μm] for a 2/3 inch ITV camera, and the resolution is 120
/4000=0.03Cμm]. Also, the capacity of the memory 17 required at this time is 4 (KB), which is 1
Compared to the case where all of the composite image signal A of a frame is stored, it is sufficiently small.

Configuration of Measuring Apparatus Next, FIG. 5 shows an example in which the method of the present invention is realized by a circuit. This measuring device 1 is similar to A-
In place of the D conversion circuit 15 and memory 17, it is configured by a comparison circuit 19, a gate circuit 20, a counter 21, a gate circuit 22, a counter 23, and the like.

The D-A converter 24 inside the comparator circuit 19 has a C
It is connected to PU7, and is connected to one input end of comparator 25 on the output side. In addition, this comparator 25
The other input end of is connected to the output side of the imaging camera 2,
Further, it is connected to one input terminal of the AND gate 26 of the gate circuit 20 on the output side. This AND gate 26 is
The other input terminal is connected to the synchronous separator 10, and the output side is connected to one input terminal of the AND gate 27 and the gate circuit 2.
It is connected to the input terminal of the second flip-flop 30.

The other input end of the AND gate 27 is connected to the high speed clock pulse generator 16, and the output side is connected to the CPU 7 via the counter 21. Furthermore, the AND gate 28 of the gate circuit 22 is connected at one input end to the high-speed clock pulse generator 16 and at the other input end to the synchronous separator lO, and at the output side to one input end of the AND gate 29. It is connected. This AND gate 29 is connected to the output side of the flip-flop 30 at the other input terminal, and is connected to the output side of the flip-flop 30 via the counter 23 at the output side.
Connected to U7.

In the case of this embodiment, the CPU 7 is a calculation unit, and in addition to sharing the function of the conversion process in the case of the above method, it also performs a control function for sequentially operating other circuit elements.

Operation of the Measuring Device Similar to the above embodiment, the imaging camera 2 converts the optical image of the measurement object 18 into an electrical composite image signal A, and the synchronization separation unit 10 converts the optical image of the measurement object 18 into the electrical composite image signal A. During a period corresponding to the scanning of the exponent portion, an "H" level synchronizing signal F is generated and sent to one input terminal of the AND gates 26 and 28. Further, since the CPU 7 generates a threshold value H at a predetermined level in advance and sends it to the D-A converter 24, the D-Ai converter 24 inputs an analog amount to one input terminal of the comparator 25. The signal of the threshold value H is inputted as a reference value.

Therefore, the comparator 25 compares the composite image signal A and the threshold value H, and generates an "H" level comparison signal I in a period corresponding to the dimension to be measured, as shown in FIG. . Therefore, the AND gate 26 uses the synchronization signal F and the comparison signal ■ as an AND condition, and during one horizontal scanning period,
The clock 7 from the high speed clock pulse generator 16 is used to open the AND gate 27 of the next stage during a period corresponding to the measured dimension.
The pulses G are sent to the counter 21 over a time width corresponding to the dimension to be measured during one scanning period. Therefore, the count value of this counter 21 corresponds to the dimension to be measured.

Therefore, the CPU 7 sets the count value of the counter 21 to RA.
After storing the value in M9, the value is read, the actual size is converted from the number of clock pulses, and the result is output by a printer or the like or displayed on the monitor 6.

In this embodiment, the output signal J of the flip-flop 30 changes from the rising edge of the synchronizing signal F to the comparison signal I.
Since it is set to “H” level until the rise of
The counter 23 counts the clock pulses G from one edge of one frame image to the time width of the dimension to be measured. Therefore, this dimension can be measured as well.

Effect of the invention The present invention provides the following effects.

First of all, a fairly wide range can be confirmed during the imaging process, and arbitrary minute dimensions within that range can be measured with high precision.

Second, the minute dimensions of the object to be measured can be measured digitally with high precision using high-speed clock pulses. Also,
In such a measurement process, it is sufficient to store only one horizontal scanning signal, so the memory capacity is small when implementing the measurement method.

Furthermore, thirdly, when the measurement method is implemented by a measurement device, the computer (CPU) program can be simplified and can be implemented by simple comparison circuits, gate circuits, counters, etc., so that the measurement method can be implemented easily. It's relatively easy.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of a measuring device when implementing the method of the present invention, Fig. 2 is an explanatory diagram of the procedure of the method of the present invention, Fig. 3 is a waveform diagram of a composite image signal, and Fig. 4 is during operation. FIG. 5 is a block diagram of the measuring device of the present invention, and FIG. 6 is a time chart during operation. 1... Measuring device, 2... Imaging camera, 7... CPU, 1
0...Synchronization separation unit, 15...A-D conversion circuit, 16...
High-speed clock pulse generator, 17...Memory, 18...
Measurement target, 19... Comparison circuit, 20... Gate circuit, 2
1...Counter, 22...Gate circuit, 23...Counter. Figure 1 Figure 2 Figure 3 Synthetic plane a signal A Figure 4

Claims (2)

[Claims] (1) An imaging process in which the measurement target is converted into a composite image signal with the dimensional direction of the measurement target in the horizontal scanning direction; a separation storage process in which an image signal corresponding to one horizontal scan is extracted from the composite image signal of one frame and stored; 1 A binarization process that separates the measured dimension area and the background area from the horizontal scanning signal by density threshold processing, and the number of dots (memory) corresponding to the number of high-speed clock pulses in the time width of the measured dimension area. A method for measuring minute dimensions, comprising a counting process and an arithmetic process of converting the number of memories corresponding to the number of high-speed clock pulses into actual dimensions. (2) An imaging camera (2) that converts the measurement object into a composite image signal with the dimension direction of the measurement object as the horizontal scanning direction, and a synchronization signal at a predetermined level over one horizontal scanning period of the predetermined number from the composite image signal of one frame. a comparison circuit (19) that compares one frame of composite image signal with a threshold level and generates a comparison signal of a predetermined level at a portion corresponding to the minute dimension of the measurement object; A gate circuit (20) generates a high-speed clock pulse over the time width of the measurement dimension using this comparison signal and the synchronization signal of the synchronization separation section as input conditions;
); and a calculation unit (7) that converts the number of high-speed clock pulses into actual dimensions.