JPS63126718A - Monitoring method for injection molder - Google Patents

Abstract

PURPOSE:To monitor always under stabilized conditions by stopping molding halfway automatically when brightness is varied much, and modifying binary coded level basing on the comparison of the darkest luminance when brightness is varied a little. CONSTITUTION:The darkest luminance level in the whole picture elements of a printed image of a mold, sensed by a CCD camera 10, is sensed by a control circuit 14, and when the same is out of the tolerance limits of the level preset by a level-upper limit setting device 20 for the darkest luminance and a level lower-limit setting device 22 for the darkest luminance, an instruction is issued to a molder controller 46 to stop injection molding. When the darkest luminance level is within the tolerance limits, the level difference value between said darkest luminance and the darkest luminance level in the hole picture elements of the normal printed image stored in a reference memory 18a beforehand is operated, the binary coded luminance level preset in a binary coded level setting device 36 is modified and the sensed printed image is binary coded by a binary coded difference sensing circuit 30.

Description

[Detailed description of the invention] (Industrial application field) The present invention relates to a method for monitoring an injection molding machine.

(Prior Art) During the injection molding process, if the mold is clamped with the molded product or part of it left in the mold, the mold may be damaged or the molded product may be defective.

In order to prevent this, conventional methods used a television camera to capture images of the condition of the mold, and by comparing the brightness level with a reference image stored in advance, the presence or absence of molded products inside the mold was determined. A monitoring method is in place to detect.

(Problems to be Solved by the Invention) However, the above conventional monitoring method has the following problems.

Abnormalities are detected by comparing the brightness between the detection image and the reference image, so if the brightness around the molding machine changes, for example if a light is turned on or off, or if it shines due to direct sunlight,
Alternatively, when a crane passes over the molding machine, the brightness level changes, so it may be judged as abnormal even though it is in a normal state, or it may be judged as normal even though it is in an abnormal state, and the brightness level may change. There is a problem in that it becomes impossible to make a judgment, and as a result, there is a problem in that a pile of defective molded products is created and the mold is damaged due to defective mold release.

(Means for Solving the Problems) In order to solve the above problems, the present invention has the following configuration.

That is, the image of the molding die is divided into pixel units, the brightness level of each pixel unit is detected, the image is binarized into white and black based on a preset binarization brightness level, and the image is stored in advance. In a method for monitoring an injection molding machine by comparing it with a binarized image of the normal state that has been stored, the darkest brightness level of all pixels in the detected image is detected, and the darkest brightness level is set as the preset allowable maximum level. If the dark brightness level is outside the range of the dark brightness level, the injection molding operation is stopped, and if the darkest brightness level of all pixels in the detected image is within the range of the preset allowable darkest brightness level, the injection molding operation is stored in advance. The darkest brightness level among all pixels of the image in the normal state that has been detected is compared with the darkest brightness level of the detected image, and if the level difference value is greater than or equal to the allowable value, the preset binary value is The detected image is binarized by vertically correcting the converted luminance level according to the level difference value.

(effect) The operation will be explained with reference to the drawings.

In FIG. 1, the control circuit 14 controlled by the CPU 34 detects the darkest brightness level of all pixels in the image detected by the CCD camera lO, and the darkest brightness level upper limit setter 2° and the darkest brightness level When the darkest luminance level of all the pixels of the detected image is not within the tolerance range of the level set in advance in the lower limit setter 22, C
The PU 34 sends a command to the molding machine controller 46 to stop the injection molding operation.

If the darkest brightness level among all pixels of the detected image is within the above-mentioned allowable range, the CPU 34 uses the darkest brightness level and the darkest brightness level among all pixels of the normal state image stored in advance in the reference memory 18a. A level difference value which is the difference from the darkest luminance level is calculated, and the binarization level setter 3 is set in advance using the level difference value.
The binarization brightness level set to 6 is corrected, and the detected image is binarized by the binarization/difference detection circuit 30.

(Embodiments) Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

First, the configuration of a monitoring device used in the monitoring method according to the present invention will be explained with reference to FIG.

10 is a CCD camera, which has a fixed image element (CCD
) is connected to a control circuit 14 via an A/D converter 12.

Further, the CCD camera 10 and the control circuit 14 are also connected to a synchronization circuit 16.

The control circuit 14 includes a reference memory 18a118b that stores each pixel of the reference image divided into 16 gradation brightness levels.
, 18c, 18d are connected to the darkest brightness level upper limit setter 20, the darkest brightness level lower limit setter 22, and the teaching instruction switch 24, and are also connected to the CRT 28 via the D/A converter 26, and are further binarized. - Connected to the CPU 34 via the difference detection circuit 30 and counter circuit 32.

The binarization/difference detection circuit 30 is a D/A converter 26.2
It is also connected to a value level setter 36.

The CPU 34 is connected to a shape/area switch 38 , a shape allowable width setter 40 , and an area allowable width setter 42 , and is also connected to a molding machine controller 46 via an interface 44 .

Next, the usage and operation of the above-mentioned monitoring device will be explained.

In this embodiment, the number of pixels is 48,960, and the brightness level is set to 16 gradations from 1 to 16.

First, set the molding conditions as usual, and once the conditions for molding a good product are set, open and stop the mold at the mold opening stop position with the molded product attached to the mold, so that you can reliably monitor the molded product. Then, adjust the position, focus, aperture, and illumination position of the CCD camera 10 while looking at the CRT 28 screen. At this time, it is efficient to focus on monitoring areas where molding defects are likely to occur in the molded product.

Next, the teaching instruction switch 24 is turned on.

Then, the image information from the CCD camera 10 is A/D converted by the A/D converter 12, a brightness level is assigned to each pixel and stored in the reference memory 18a, and the binarization difference detection circuit 30 sets the brightness level in advance. It is binarized at the brightness level "7" that was set as the boundary, and the CPU 3 uses it as a reference binarized image of the normal state.
4 is stored.

On the other hand, the number of black pixels is counted by the counter circuit 4, and the number is stored in the CPU 34 as a reference count number.

Next, the molded product is ejected, the molded product is released from the mold, and the image of the mold after the molded product is taken out is stored in the reference memory 1.
8b, and the image is similarly binarized and sent to the CPU.
34, the number of black pixels is also counted, and the CPU
34.

A case in which the shapes of molded products are compared will be explained with reference to FIG. 2.

For convenience of explanation, the number of pixels in the image is assumed to be 36 (6×6).

FIG. 2(a) shows an image when a molded product in a normal state is in the mold, and ■, ■, ■, ■ indicate the brightness level (the same applies to the other figures). This image is stored in the reference memory 18 as described above.
It is stored in a.

(b) is a diagram showing the case where the image in (a) is binarized with the luminance level r7J as the boundary, pixels with a luminance level of "7" or lower are treated as black, and the counter circuit 32 determines that the number of black pixels is "12".
” is counted and “12” is stored in the CPU 34.

Note that the number of black pixels in the image stored in the reference memory (18b) is, of course, "0" and is stored in the CPU 34.

After switching the shape area switching device 38 to the shape side, the mold is closed and injection filling work is performed, and the mold is opened after cooling and solidification is completed. When the mold opening reaches the mold opening position, the molding machine controller 46 switches the mold open. A completion signal is sent to CPU 34.

Then, a signal is sent from the CPU 34 to the control circuit 14, and the control circuit 14 detects the darkest luminance level "4" among all the pixels in the detection image sent from the CCD camera 10 (Fig. 2 (C)) at that point. and the brightness level “10” is set in advance in the darkest brightness level upper limit setter 20.
and the brightness level "1" set in advance in the darkest brightness level lower limit setter 22, that is, it is calculated whether it is within the range of the allowable darkest brightness level. As a result, if the darkest brightness level of the detected image is outside the range of the allowable darkest brightness level, an abnormal signal is sent from the CPU 34 to the molding machine controller 46.
The molding operation is interrupted.

In this case, the amount of light in the work environment is either too low or too strong, causing halation, so it is best to adjust the amount of light in the work environment.

If the darkest brightness level of the detected image is within the allowable darkest brightness level (in this example, the darkest brightness level is "4")
'', so it is within the range), and the difference from the darkest luminance level "5" among all pixels of the reference image stored in the reference memory 18a is calculated.

5-4=1 This obtained level difference value "1" is subtracted from the brightness level "7" set in advance in the binarization level setter 36? -1=6 The corrected binarization luminance level value "6" is determined, and the binarization/difference detection circuit 30 binarizes the detected image with luminance level [6] or lower as black and the others as white. A binarized image (FIG. 2(d)) of the detected image is obtained and stored in the CPU 34.

Then, a comparison with the reference binarized image (FIG. 2(b)) showing a normal state stored in advance in the CPU 34 is performed.
The difference detection circuit 30 generates a difference image (FIG. 2(e)).
is obtained. By switching the input to the Naori/A converter 26, this difference image can also be displayed on the CRT 28.

Next, the obtained difference image is sent to the counter circuit 32,
The difference bit number (difference in the number of black pixels) from the reference binarized image (Fig. 2(b)) is counted, and the count value "2" is
The data is sent to the PU 34, and the shape difference rate (number of difference bits/total number of pixels x 100 (%)) is calculated.

2/36X100 = 5.6 C%] and the allowable width set in advance in the shape allowable width setting device 40 (in this example, 4
%).

In this case, since the shape difference rate is greater than 4%, it is determined that there is a shape defect, and the CPU 34 sends a signal to the controller 46 of the molding machine to instruct it to interrupt the operation.

Further, if the shape difference rate is 4% or less, the molding machine advances to the next step and the detected image data is initialized.

When the ejection completion signal is sent from the molding machine controller 46, the binarized image of the image stored in the reference memory 18b is compared with the binarized image of the detected image after the ejection is completed, and the same as above is performed. The shape difference rate is calculated. If this shape difference rate deviates from the allowable range, it means that the molded product has not been completely ejected from the mold, and the CPU 34 sends an abnormality signal to the molding machine controller 46 to interrupt the operation. If the shape difference rate is within the allowable range, it is assumed that the molded product has been completely discharged, and one cycle of the process ends.

Although the above example was a method of monitoring by comparing shapes, it is also possible to adopt a method of counting the number of black pixels and comparing by area. At that time, switch the shape/area switch to the area side.

In this example, the two reference memories 18a and 18b were used to monitor the shape or area of the molded product when the mold was opened, the presence or absence of the molded product remaining in the mold, etc. good.

In addition, when the machine has multiple molds such as a rotary complete molding machine, it is possible to monitor and check the inserted product by adding a reference memory before and after each mold process.

Furthermore, if the reference image can be stored only when the brightness level range of the subject is within a predetermined brightness level range,
It is good that the correction width of the binarization level can be ensured.

(Effects of the Invention) According to the present invention, molding is automatically interrupted when the surrounding brightness changes significantly and normal monitoring cannot be performed, thereby preventing a series of molding defects due to poor monitoring and damage to the mold. Furthermore, small changes in brightness over time are handled by correcting the binarization level based on a comparison of the darkest brightness levels, so monitoring can always be performed under stable conditions, increasing reliability and It has remarkable effects such as greatly improving safety.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a monitoring device for an injection molding machine according to the present invention, and FIG. 2 is a diagram showing an example of an image when monitoring is performed using a shape comparison method. 10... CCD camera, 14... Control circuit, 18a, 18b, 18c, 18d... Reference memory, 20... Darkest brightness level upper limit setter, 22...
・Darkest brightness level lower limit setter, 28...CRT,
30... Binarization, difference detection circuit, 32... Counter circuit, 34... CPU, 36... Binarization level setter, 40... Shape allowable width setter, 46.
・Former controller.・Patent applicant: Yoshiharu Shima, representative of Nisseki Jushi Kogyo Co., Ltd.

Claims (1)

[Claims] 1. Divide the image of the mold into pixel units, detect the brightness level of each pixel unit, and divide the image into white and black based on a preset binarized brightness level. In a method for monitoring an injection molding machine, which is performed by comparing it with a binarized image of a normal state stored in advance, the darkest brightness level among all pixels of the detected image is detected, and the darkest brightness level is If the brightness level is outside the preset allowable darkest brightness level, the injection molding operation is stopped, and the darkest brightness level among all pixels of the detected image is within the preset allowable darkest brightness level. In this case, the darkest luminance level among all pixels of a normal state image stored in advance is compared with the darkest luminance level of the detected image, and if the level difference value is equal to or greater than the allowable value, the preset 2
A method for monitoring an injection molding machine, characterized in that the detected image is binarized by vertically correcting the digitized luminance level according to the level difference value. 2. A method for monitoring an injection molding machine according to claim 1, characterized in that a CCD camera is used as the image detection means.