JPH0210988B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a figure reading device, and more particularly to an automatic figure reading device used for making tape for a numerically controlled hole punching machine for drilling holes for component mounting parts and through holes in printed circuit boards. be.

Currently, very few holes for component mounting or through holes are quantified in advance when designing printed patterns, and most printed patterns are created on a film sheet using a hole reading device. The print pattern is being read manually.

As mentioned above, conventional hole reading devices are operated manually, so the worker visually observes each hole with a magnifying glass, etc., and the center of the hole is aligned with the center of the crosshair mark on the magnifying glass. By operating the handle, the movable stage or optical system on which the print pattern is placed is moved, and when the centers are aligned, the coordinates of the hole are read and the hole diameter is commanded to a control device such as a computer.

The above-mentioned conventional apparatus has the disadvantage that it causes a great deal of fatigue to the worker as he or she has to match each hole one by one, and that this leads to errors in reading the holes or failures in reading the holes. In particular, with the improvement of print pattern technology and the miniaturization of mounted components, the mounting density of printed circuit boards tends to increase more and more, and manual operation is therefore reaching its limits both physically and mentally. In addition, if the worker aligns the center of the hole with the center of the crosshair mark on a magnifying glass, there is no ability to read the hole diameter. It is necessary to attach it to a sheet-like printed pattern, and pre-work is required before the hole reading operation, and the time required for this work is also a non-negligible drawback.

In order to eliminate the above-mentioned drawbacks, the present invention provides a figure reading device that automates the figure reading operation by introducing simple pattern recognition.

The present invention can be widely used for reading various figures, but when limited to reading printed patterns, the present invention scans a figure including a target hole with an image sensor, etc.
The present invention is characterized in that the area where the hole exists is detected from the position of the tip and the rear end of the hole from the image signal, and the center coordinates and diameter of the hole are calculated within the area.

An embodiment in which the present invention is utilized for reading printed patterns will be described in detail below with reference to the drawings.

FIG. 1 is a schematic diagram showing the configuration of a figure reading device according to the present invention. A sheet 1 on which a target figure is formed is placed on a movable table 2 that can be moved one-dimensionally or two-dimensionally. As shown in FIG. 2, the target figure has a pattern portion 1a that is black, and the other portions 1b, including the hole portions, are transparent. Mobile platform 2
is driven by a drive unit 3, and when moving two-dimensionally, two drive units 3 are provided at right angles to each other. An imaging unit 4 that optically reads target figures formed on the sheet is suspended opposite the sheet 1 . The imaging unit 4 is composed of a photoelectric conversion element such as a one-dimensional photodiode array, a two-dimensional photodiode array, or an image pickup tube, and the optical unit uses a lens system that determines the magnification when reading the target figure. Furthermore, a light source section 5 is provided to provide an image pickup section 4.
Provides uniform illumination within the field of view. The light source section 5 is
In FIG. 1, the configuration is such that light is reflected from the target figure, but it may be configured to detect transmitted light from the target figure. The imaging unit 4 includes a processing device 6.
is connected, processes the light source conversion signal of the target figure obtained by the imaging unit 4, and recognizes the feature amount of the target figure, and is the main component of the present invention, and is composed of the functional circuit shown in FIG. There is.

In FIG. 3, the processing device includes a leading end position detecting circuit 10 and a trailing end position detecting circuit 11 of the target figure. As shown in FIG. 4, the tip position detection circuit 10 electrically scans the imaging unit 4 in the direction of ○a and moves the movable table 2 in the opposite direction to the direction of ○a at a constant speed. By scanning in the vertical direction, the position of the tip of the target figure is detected as a point. The rear end position detection circuit 11 detects the rear end position at the point shown in FIG. Details of the leading and trailing end position detection circuits 10 and 11 are shown in FIG.

FIG. 5 is an example of determining the leading and trailing positions of a target graphic in the automatic graphic reading device according to the present invention. FIG. 6 is a time chart for explaining the above embodiment. In FIG. 5, the output of the imaging section 4 is amplified to a predetermined level by an amplifier 30, and converted into a digital signal by a slice circuit 31. The output signal of this slice circuit 31 is delayed by a predetermined horizontal scan amount in a shift register 32, and is input to an AND circuit 33 together with the next scan output of the slice circuit 31. The signal is input to the first counting circuit 38 together with the horizontal scanning direction clock signal, and the width of the portion of the target figure other than the pattern portion is counted. Second counting circuit 41
counts the width at which the output signal of the AND circuit 33 becomes "1". The contents of the first and second counting circuits are read out by the output signal of the first trigger generation circuit 34 generated at the falling edge of the signal. The counting contents of the first counting circuit 38 are later
It is compared in a first comparator 39 with the value set by the first setter 40 in order to eliminate detection errors as explained in connection with the figure. The output of the second counting circuit 41 is input to the first zero detection circuit 42, and zero detection is performed. The first and second counting circuits 38, 41 are cleared by the output signal of the second trigger generation circuit 35 generated at the end of the first trigger signal.

The conditions for detecting the tip position of the target figure are as shown in FIG.
This is a case where a transparent part corresponding to a hole is scanned.
That is, in FIG. 5, when the output signal of the first counting circuit 38, which contains the contents of the current horizontal scan, is equal to or greater than a predetermined value, and the second counting circuit 41, which contains the contents of the previous horizontal scan, When is zero, the position of the tip of the hole is detected by the second AND circuit 48. As shown in FIG. 7, even if the target figure is not a hole, there is a possibility that an output signal will be output from the second AND circuit 48. If the determination is made taking into consideration the scanning speed of , no problem will occur.

Detection of the rear end position of the target figure is performed as follows. The output signal of the shift register 32 is input to a third counting circuit 43 together with a clock signal in the horizontal scanning direction, and the width of a portion of the target figure other than the pattern portion is counted. The fourth counting circuit 46
The width at which the output signal of the first AND circuit 33 similar to the counting circuit 41 becomes "1" is counted. The content of the third and fourth counting circuits is the output signal of the third trigger generation circuit 37 generated at the falling edge of the signal.
d' and the third counting circuit 43
is compared with the value set by the second setter 45 by the second comparator 44. Fourth counting circuit 4
The output g' of 6 is input to a second zero detection circuit 47, where zero detection is performed. The third and fourth counting circuits 43, 46 are cleared by the output signal e' of the fourth trigger generating circuit 36 generated at the end of the third trigger generating circuit output signal d'. The condition for detecting the rear end position of the target figure is that, as shown in Figure 6b, when the pattern part is being scanned in the current horizontal scan, the transparent part corresponding to the hole part was scanned in the previous horizontal scan. This is the case. That is, in FIG. 5, the output signal f' of the third counting circuit 43, which contains the contents of the previous horizontal scan, is greater than or equal to a predetermined value, and the fourth counting circuit, which contains the contents of the current horizontal scan, When the circuit 46 is zero, the rear end position of the hole is detected by the third AND circuit 49.

Tip position detection circuit 10 and rear edge position detection circuit 11
The output of is given to the area detection circuit 12 as shown in FIG. FIG. 8 shows details of the target figure area detection circuit.

In FIG. 8, the flip-flop 20 is set by the output of the target figure tip position detection circuit 10,
It is reset by the output of the target figure trailing edge position detection circuit 11, and the output of the flip-flop 20 defines the vertical area of the target figure.

The counting circuit 22 that counts clock signals in the horizontal scanning direction is reset when the number of clocks determined by the imaging unit 4 and its resolution is counted, and one horizontal scan in the imaging unit 4 is completed, and the next horizontal scan is started. will be started. When the tip position of the target figure is detected at the point shown in FIG.
The output of 2 is stored in the storage circuit 23. That is, the storage circuit 23 stores a count value corresponding to the coordinate X in the horizontal scanning direction of the tip position of the target figure shown in FIG. The comparison circuit 24 calculates the horizontal coordinates of the tip position of the target figure and the counting circuit 22.
After detecting the tip position of the target figure, it is detected that the target figure has been scanned to a position equal to the horizontal coordinate X of the tip position in each subsequent horizontal scan. The flip-flop 25 is set by the output of the comparison circuit 24 and reset by the fall of the image signal, and defines the rear end of the horizontal area of the target figure. That is, the area detection circuit 12 defines the horizontal and vertical areas of the target figure, and the feature extraction circuit 13 extracts the feature amount of the target figure from the image signal within the area. Find the center coordinates, size, etc.

FIG. 9 shows an embodiment of a circuit for extracting features using the automatic figure reading device according to the present invention.

In each horizontal scan from when the leading edge position of the target figure is detected to when the trailing edge position is detected, the logical product of the output signal of the flip-flop 25 shown in FIG. 8 and the output signal of the first trigger generation circuit shown in FIG. is determined by the circuit 50, and using the output signal as a clock, the output of the counting circuit 22 in FIG.
to be memorized. The output signal of the register 51 is used by a first maximum value detection circuit 52 to determine the maximum horizontal rear end position of the target figure. Furthermore, using the output signal of the AND circuit 50 as a clock,
The output of the counting circuit 38 is stored in the second register 53, and the maximum horizontal width of the target figure is determined by the second maximum value detection circuit 54. Maximum rear end position (X max in Figure 10) and maximum width (1st
The information of Wx max) of figure 0 is input to the arithmetic processing unit 55, and the horizontal center coordinate (Xc) of the target figure is input to the processing unit 55.
is obtained using equation (1). At this time, let the origin be X ₀ ,
If the amount of movement Xi is the i-th intermittent movement of the moving table or optical system in the X direction from the origin X ₀ , the horizontal center coordinate Xc of the target figure after the i-th intermittent movement in the X direction is Xc = Xmax - 1/2Wx max + ΣXi ...(1) On the other hand, for vertical information, a counting circuit 56 counts the pulse inputs of the drive unit 3 of the movable table 2, and the output signal is converted to the output signal of the trigger generation circuit 21 in FIG. is stored in the third register circuit 57 as a clock, and is input to the arithmetic unit 55. Counting circuit 58
Then, the output signal of the flip-flop 20 in FIG. 8 is counted as the pulse input to the interval moving section 3 of "1", and an output signal (Wy in FIG. 10) is obtained. 8. The output signal of the trigger generation circuit 59, which is activated when the output of the flip-flop 20 changes from "1" to "0", is stored as a clock in the fourth register circuit 60, and this is input to the arithmetic processing unit 55. Then, the center coordinate Yc in the vertical direction of the target figure is determined by equation (2).

Yc=Y+1/2Wy (2) That is, the center coordinates and size of the target figure can be found.

Furthermore, from the output Wx max of the second maximum value detection circuit 54 and the output Wy of the counting circuit 58 in FIG.
Calculate Wx max−Wy|.

|Wx max−Wy|≦a ...(3) where a is a preset value When this formula (3) is satisfied, a circuit that determines that the target figure is a circle can be included in the feature extraction circuit. It is possible.

Furthermore, the flip-flop circuit 2 in FIG.
0, trigger generation circuit 21, memory circuit 23, comparison circuit 24, flip-flop 25, as shown in FIG.
AND circuit 50, first register circuit 51, first maximum detection circuit 52, second register circuit 5
3. Second maximum detection circuit 54, third register circuit 57, counting circuit 58, trigger generation circuit 5
9. By providing a plurality of fourth register circuits 60, it is possible to detect a plurality of target figures simultaneously in the horizontal scanning direction.

FIG. 11 shows an embodiment of means for determining the amount of overlap of the effective fields of view of the imaging sections using the automatic figure reading device according to the present invention. When using an automatic reading device to scan the entire surface of a printed pattern and read the center coordinates and hole diameter of the target hole, the printed pattern is generally formed in a size that is smaller than the effective field of view of the imaging unit. The size of the film sheet used is larger, and therefore it is necessary to add mechanical scanning in addition to the electrical scanning provided by the image pickup device. When mechanical scanning is performed intermittently, it is impossible to accurately read a target figure unless scanning is performed such that the target figure is completely within the effective field of view of the imaging section. That is, it is necessary that the effective field of view of the imaging section partially overlaps between one mechanical scan and the next mechanical scan. Generally, scanning is carried out with the amount of overlap determined at a fixed amount by taking into account the size of the target figure, but there are various sizes of target figures, and the amount of overlap is determined by the largest amount. Since it is uniquely determined by the target figure, the actual effective field of view of the imaging section becomes small, and the time required to scan the entire surface of the film sheet on which the film pattern is formed is disadvantageous. In this embodiment, the overlapping amount of the effective field of view is not fixed to a constant value but is made variable, thereby making it possible to shorten the figure reading time.

In FIG. 11, an AND circuit 61 calculates the logical product of the output signal of the flip-flop 25 in FIG. 8, the horizontal scanning termination signal, and the output signal of the first trigger generation circuit 34 in FIG. Counting circuit 38 in FIG. 5 to circuit 62
The output signal of is stored. The maximum value detection circuit 63 is
The maximum value of the contents of the register circuit 62 is determined. That is, the maximum value detection circuit 63 stores the largest value among the target figures at the end in the horizontal scanning direction X. In FIG. 12, We max is the largest among We. If the effective field of view length in the horizontal scanning direction X of the imaging unit 4 is D, then the arithmetic processing unit 55 in FIG. .
This enables efficient scanning and reduces the figure reading time.

Furthermore, according to the present invention, when optically measuring a target figure existing at an arbitrary position, the presence and size of an incomplete figure at the edge of the field of view is automatically detected, and the amount of movement of the field of view for the next scan is set to the optimum value. Therefore, the optimum pattern reading sequence can be determined and the reading time can be shortened.

[Brief explanation of drawings]

Fig. 1 is a schematic diagram of a figure reading device according to the present invention, Fig. 2 is a diagram showing an example of a target figure pattern formed on a sheet, and Fig. 3 is a block diagram of a processing device which is the main part of the present invention. Figure 4 is a diagram for explaining detection of the leading edge position and trailing edge position, Figure 5 is a block diagram of the leading edge position detection circuit and trailing edge position detection circuit, and Figure 6 is a diagram showing each part of the circuit in Figure 5. a waveform diagram showing the waveform appearing in relation to the hole pattern of FIG. 4;
Figure 7 is a waveform diagram that appears in the circuit of Figure 5 when the target figure detects a pattern with no holes, Figure 8 is a block diagram of the area detection circuit of the target figure, and Figure 9 is a block diagram of the feature extraction circuit. , FIG. 10 is a diagram for explaining the process of determining the center coordinates and size of a target figure by the circuit of FIG. 9, FIG. 11 is a block diagram of a circuit for determining the amount of overlap of the effective fields of view of the imaging section, FIG. 12 is a diagram illustrating a method for determining the amount of overlap using the circuit shown in FIG. 11. [Explanation of symbols of main parts], light source...5, imaging unit...4, drive unit...3, moving table...2, tip position detection means...10, rear end position detection means...1
1. Area detection means...12.

Claims (1)

[Scope of Claims] 1. A light source that uniformly illuminates a target figure, and an imaging unit having a predetermined field of view width arranged one-dimensionally or two-dimensionally and receiving transmitted light or reflected light of the illumination light by the target figure. , a figure reading device having a drive unit that intermittently moves the target figure or an imaging unit one-dimensionally or two-dimensionally, detecting a leading end position and a trailing end position of the target figure from an image output signal of the imaging unit. and means for detecting the position of the tip and determining the amount of intermittent movement of the drive unit based on the maximum width of the incomplete target figure appearing at the edge of the field of view. A figure reading device. 2. In the device according to claim 1,
The overlap amount detecting means is configured to electrically scan a horizontal width determined from the effective width of the imaging section when the movable table or the imaging section to which the target figure is fixed is located at a certain horizontal position. means for detecting the largest width Wemax in the horizontal scanning direction of a target figure detected at a position at the end of horizontal scanning from an image signal obtained when moving the movable table or the imaging section in the vertical direction; effective width D in the direction
The present invention is characterized by comprising a means for calculating D-Wemax and a means for commanding an amount based on the calculation result to the drive unit of the movable table or the imaging unit as an intermittent movement amount in the horizontal direction. Automatic figure reading device.