JPS6320996A - Focusing method for visual sensor - Google Patents

Abstract

PURPOSE:To simplify a circuit constitution and to enable a high speed processing by converting first and second patterns of a sample label positioned at a position to be focused by a video camera into a picture image signal, and making the area ratio of the patterns based on a binary data a predetermined value. CONSTITUTION:An object to be focused is stuck with the sample label SL, and is image-picked-up by the video camera 1. Thus, a composite video signal CV is outputted, and from a synchronizing separation circuit 3, the picture image signal GV corresponding to the pattern of the sample label SL, e.g. a circle A and an annulus ring B, is outputted to a binarization circuit 11. The binarization circuit 11 converts said picture image signal GV into a binarization data, and outputs it. This output is written in a picture image memory 15. A CPU 6 reads out this binary coded data, and the area Sa and the area Sb of the circle A and the annulus ring B are obtained. Next, a focusing motor built in the video camera 1 is driven according to an area difference DELTA S= Sa-Sb. Next, the motor is again driven according to the area difference DELTAS, and so forth, the above-mentioned process is repeated until the area difference comes to zero, and the focusing is completed.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention This invention relates to a focusing method in a visual sensor using a video camera.

"Prior Art" FIG. 6 is a block diagram showing the configuration of a conventional visual sensor. In this figure, 1 is a video camera, which produces a composite video signal CV that combines an image signal and a synchronization signal.
Output. 2 is the visual sensor body. In the main body 2, numeral 3 denotes a synchronous separation circuit which removes the synchronous signal from the composite video signal CV and outputs only the image signal GV to an A/D (analog/digital) converter 4. A
The /D conversion circuit 4 sequentially samples the supplied image signal GV using clock pulses of a constant period, converts the sampled data into digital data, and outputs the digital data. The data output from this A/D conversion circuit 4 is written into the image memory 5.

6 is a CPU (central processing unit), which reads data in the image memory 5 and performs various processing. 7 is a motor control circuit, which controls the focusing motor (motor that drives the focusing ring) built into the video camera l by the CPU.
It is driven according to the output data of No. 6. A display control circuit 8 displays images on the display device 9 based on display data output from the CPU 6.

Next, a method of focusing in the conventional visual sensor described above will be explained. First, when the video camera l is completely focused, the image signal GV included in the composite video signal CV reaches its maximum level, as shown by the solid line L1 in FIG. As the level decreases. Therefore, in the conventional visual sensor described above, images of the subject H are sequentially written into the image memory 5 while moving the focusing motor little by little.
Based on the data in the image memory 5, the position where the level of the image signal GV is the highest is detected and the rotation of the focusing motor is stopped.

"Problems to be Solved by the Invention" However, in the conventional focusing method described above, it is necessary to convert the image signal GV into multi-gradation digital data and write it into the image memory 5. A high-speed, multi-bit A/D conversion circuit 4 and a large-capacity image memory 5 are required, making the configuration complicated and expensive. Another problem is that it takes a long time to process, making it unsuitable for real-time control.

This invention was made in view of the above-mentioned circumstances, and makes the focusing circuit simple and inexpensive! It is an object of the present invention to provide a focusing method for a visual sensor that can perform a focusing operation faster than conventional methods.

"One-shot means for solving the problem" This invention places a sample label on which a first and second figure are drawn at the position where the focus should be adjusted, and the video camera detects the first and second figures on the sample label. 1. The second figure is converted into an image signal, the image signal is sampled and converted into binary data, and based on this binary data, the first...
Find the area of each of the second figures, and calculate the area of each of the first and second figures. The present invention is characterized in that the video camera is focused so that the area ratio of the second figure becomes a predetermined value.

"Operation" According to the present invention, there is no need to convert the image signal into multi-gradation digital data, and it is only necessary to convert it into binary data, so the configuration of the A/D conversion circuit becomes extremely simple and inexpensive. , the processing speed is also faster.

"Embodiment" Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a visual sensor to which the focusing method according to the present invention is applied. In this figure, parts corresponding to those in FIG. 6 are given the same reference numerals. In this figure, the symbol SL is a sample label. This sample label SL is affixed to the subject to be focused on, and as shown in Figure 2, there is a black circle A on a white mount, and a black circle drawn on the outside of this circle A with the same center. The circular rings B and B are respectively displayed.

In this case, radius rl of circle A + radius rt of inner circumference circle of ring B
, the radius r of the outer circle of the ring B, there is a relationship rl"+rt"=r3"""' (1).This means that the area Sa of the inner circle A and the radius r of the outer circle of the ring B. This means that the area sb is equal.In other words, Sa=πr, 1...(2), and by substituting the above equation (1) into this equation (2), S
a= πr, "= W (rs" rts=Sb・
...(3). In other words, the sample label SL includes a circle A and a circle ff1B that have exactly the same area.
are each depicted.

Next, in FIG. 1, 11 is a binarization circuit.

The binarization circuit 11 samples the cape image signal GV output from the synchronization separation circuit 3, and converts the sampled value into a binary number of "1"/"0" based on a predetermined threshold level sh. This binary data is then written into the image memory 5.

Next, the focusing operation of the visual sensor shown in FIG. 1 will be explained.

When performing focusing, first, a sample label SL is attached to the subject to be out of focus, then the video camera l is directed to the sample label SL, and an image is taken. As a result, the composite video signal CV is output from the video camera l, and therefore the image signal GV corresponding to the figure of the sample label SL is output from the synchronization separation circuit 3 to the binarization circuit! ■Output to. The binarization circuit 11 converts the image signal GV into binary data and outputs it. This output binary data is written into the image memory 5. CPU
6 reads the binarized data written in the image memory 5, and calculates the surfaces fasa and Sb of the circle A and the circle B from the read data. There are various ways to obtain this, but for example, first find the radii r1 to r3, and then calculate these radii "rl"
The areas Sa and Sb may be calculated from '-r3, or the areas 5aySb may be calculated from the number of dots of "l". Next, the area difference between circle A and circle B is Δ5=Sa−
9b is calculated, and the focusing motor built into the video camera l is driven via the motor control circuit 7 in accordance with this calculation result. Next, the area difference ΔS between the circle A and the ring B is calculated again, and the focusing motor is driven according to this calculation result until the area difference becomes O, that is, until the focusing is completed. Repeat the above process. The above is the focusing operation of the visual sensor shown in FIG.

Here, the reason why focusing can be performed based on the area difference is as follows. Now, when the black-on-white image shown in FIG. 3 is captured by the video camera 1, the image signal GV corresponding to the scanning line a becomes like the solid line Ll shown in the same figure when in focus, Moreover, when the object is out of focus, the image becomes as shown by a broken line L2. In other words, when the focus is on, the level becomes large and the rising/
The fall will also be steep, while when the focus is out of focus, the level will be small and the rise/fall will be gradual. Therefore, if the threshold level sh in the binarization circuit 11 is set to the position shown in the figure, when the focus is on, the width of the black image and the pulse width of the image signal G■ will be approximately equal. , if the image signal GV is out of focus, the pulse width of the image signal GV is the length h of the diagram.
It becomes l. In this way, when the image is out of focus, the width of the black image is detected to be different from the actual length.

Therefore, when imaging the sample label SL in FIG. 2, if the image is out of focus, the radius r1 to r
, is detected to be different from the actual length, and as a result, the area Sa of the circle A and the area sb of the ring B are also calculated as different values. Therefore, the area Sa of circle A and the area s of circle B
b is calculated by the CPU 6, and then the difference ΔS between the two is calculated.
By calculating ΔS and checking whether or not this value ΔS is 0, it is possible to detect whether the image is in focus or not.

Next, the relationship between the value of ΔS and the driving of the focusing motor will be described.

Now, assume that the radii r1 to r3 shown in FIG. 2 each change by Δr due to defocus. In this case, the area difference ΔS is ΔS=π(r+〒Δr)”−(π(r−+Δr)”−
π(rt −Δr)2) (4) Expanding this equation (4) and substituting the above equation (1), ΔS=πΔr(2(rt −r3− rt)+Δr)
--Equation (5) is obtained. Here, if k=rt-rs-rt--(6), the above equation (5) becomes ΔS=πΔr(2+Δr)--(7). In addition,
k<O. When this equation (7) is illustrated, it becomes a quadratic curve as shown in FIG. From this curve, we can see the following: That is, when ΔS is positive, the Iha point matching Yamakawa motor is rotated in the direction in which ΔS decreases, but
In this case, it is necessary for Δr to increase, or in other words, for the area Sa of circle A to increase. Otherwise, even if ΔS=0 (see point P in Figure 4), Δ
r is not 0, so it is out of focus. That is, if ΔS is positive, the area Sa
Focusing is completed when ΔS h(O) increases.Similarly, when ΔS is negative, focusing is completed when the area Sa decreases and ΔS h<o.

In the above embodiment, as is clear from FIG. 4, the point where ΔS=0 is not an inflection point, but has a steep slope, so the focal position can be detected at high speed.

Note that the Threnojord level Sh is determined by the error Δ shown in FIG.
It is desirable to select a value close to the upper limit level or lower limit level of the image signal GV so that r is as large as possible.

Further, in the above embodiment, the video camera l was provided with a focusing motor, but in the case of a video camera in which focusing is performed manually, the above-mentioned USA,
sb and surface kl difference ΔS may be displayed on the display device (for example, a liquid crystal panel) 9 every time a certain period of time elapses. FIG. 5 shows a display example in this case. in this case,
While looking at this display, the operator uses the focus ring to
Alternatively, focusing can be done by changing the distance between the camera and the subject.

Further, in the above embodiment, the circle A and the ring B are drawn on the sample label, but other shapes may be drawn. Furthermore, the area ratio between the circle A and the ring B does not have to be 1:1.

``Effects of the Invention'' As explained above, according to the present invention, it is not necessary to convert the image signal into multiple values, it is only necessary to convert the image signal into a binary signal, and after the image signal is converted into a binary signal, the area and area difference can be calculated. As a result, the circuit configuration can be made simple and inexpensive, and the advantages of high-speed processing can be obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.
Figure 2 shows the sample label S used in the same example.
FIG. 3 is a diagram showing changes in image signal GV according to the focal position, FIG. 4 is a diagram showing the relationship between area difference ΔS and radius error Δr, and FIG. 5 is a representation in FIG. 1. FIG. 6 is a block diagram showing an example of the configuration of a conventional visual sensor, and FIG. 7 is a diagram for explaining a focusing method in the same visual sensor. l...Video camera, 5...Image memory, 6...CPU, 7...Motor control circuit, 11...Binarization circuit , SL... Sample label, GV... Image signal.

Claims (1)

[Claims] In a method for focusing a visual sensor using a video camera, a sample label on which a first and a second figure are drawn is arranged at a position to be focused, and the first and second shapes of the sample label are set by the video camera. Convert the figure into an image signal, sample the image signal and convert it into binary data, calculate the areas of the first and second figures, respectively, based on this binary data, and calculate the areas of the first and second figures. A focusing method for a visual sensor, characterized in that the video camera is focused so that the area ratio of the figure becomes a predetermined value.