JPH04352286A - Image processor - Google Patents

Abstract

PURPOSE:To easily and speedily adjust a threshold level by providing a means which properly varying the value of the threshold level, a signal processing means and a display control means. CONSTITUTION:A display control part 5 converts difference data into switching data according to a specific level by using a color table 52 and transfers the data to a switching part 54 in sequence and in a switching part 5, analog switches SW1-SW3 are selected and switched into a conducting state. The value of the threshold level is properly varied by adjustment, etc., and the difference between image data to be decided and reference image data is displayed on a display means by making plural different colors correspond to the values of the differences. Consequently, the difference can be confirmed visually and instinctively. Therefore, the threshold level for deciding whether or not there is abnormality and the difference point from the reference image data in an optimum state can easily be adjusted over a look at the colored display.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus used, for example, to discover the presence or absence of a change in an object to be determined.

0002

2. Description of the Related Art Conventionally, in an image processing apparatus that determines whether or not a change such as an abnormality has occurred in an object to be determined, first,
The object to be determined is imaged with a video camera, etc., and image data corresponding to one frame image is stored in a storage device such as a video RAM as pixel data for each pixel, and the data of this one frame image is determined as reference data. Furthermore, a threshold level (threshold value level) serving as a reference for making change judgments, etc. is set in advance.

[0003] When performing actual judgment processing,
The object to be determined is imaged with a video camera, each pixel data of one frame fraction obtained by this imaging is compared with each pixel data of one frame of reference data in the storage device, and the level of each pixel data is determined. Calculate the difference. That is, the absolute value of the difference between the image data obtained by imaging the object to be determined and the reference data is determined for each pixel, and then the absolute value of the difference for each pixel is compared with the above threshold level to determine the threshold value. It is determined that an abnormality has occurred in the portion of the object to be determined that corresponds to the pixel in which the difference is greater than the level.

As a more specific example, regarding a monitoring device used in a plastic molding machine, when constantly monitoring whether or not an abnormality has occurred in a mold during molding operation, the first step is to detect an abnormality. By storing an image of the mold without any mold as reference data, you can prevent foreign matter from adhering to the mold during the molding operation, or from parts of the molded product still adhering to the mold when the mold is opened. If so, by comparing the difference for each pixel between the image data taken during the molding operation and the reference data with a predetermined threshold level, only the abnormal area exceeds the threshold level, so it is possible to identify the area where the abnormality has occurred. It became possible to detect.

[0005]

[Problems to be Solved by the Invention] However, in such conventional image processing devices, the above-mentioned threshold level is set in advance to be the difference between the reference data and the image data obtained by imaging the object to be determined. Differences between the standard data and the image data of the object to be judged are detected by comparing them as a standard, so the judgment accuracy changes depending on the threshold level setting, so it is difficult to determine the optimal threshold level. There is a problem that setting is extremely complicated and difficult.

[0006] That is, if the threshold level is set to a high value, the so-called sensitivity decreases, so abnormalities etc. cannot be detected, and conversely, if the threshold level is set to a low value, Since the so-called sensitivity increases, problems arise such as noise being judged as abnormal or the judgment result changing due to a slight change in the light intensity of the illumination for illuminating the object to be judged.

[0007] As described above, in order to set the optimum threshold level, delicate adjustments are required.In the case where an image processing device is installed in the above-mentioned plastic molding machine to monitor the occurrence of abnormality, it is as follows. First, it was necessary to perform a trial run of the molding machine with the image processing device installed, and to perform optimization by repeatedly adjusting the threshold level through trial and error. Furthermore, if the object to be judged differs, it is necessary to make the above-described complicated adjustment by trial and error each time, which is inefficient.

The present invention has been made in view of the above problems, and it is an object of the present invention to provide an image processing device that can quickly and easily adjust the threshold level.

[0009]

[Means for Solving the Problems] In order to achieve such an object, the present invention stores reference image data obtained by taking an image of an object to be determined in advance, and uses the data to store the reference image data obtained by taking an image of the object to be determined in advance. An image processing device that compares the image data to be determined obtained by imaging an object and determines that a change such as an abnormality has occurred in a portion where the difference between the reference image data and the image data to be determined exceeds a predetermined threshold level. The target is

[0010]The value of the threshold level can be changed as appropriate by adjustment or the like, and
Regarding the difference between the above image data to be determined and the reference image data,
By displaying a plurality of different colors on the display means in correspondence with the size of the difference based on the threshold level, the size of the difference can be confirmed visually and intuitively. I made it.

[0011] Note that the present invention does not limit the resolution of the image data to be determined and the reference image data to be subjected to image processing. That is, for example, when capturing an image of a target object with an imaging device such as a video camera, the target image data described above is composed of pixel data for one frame of the highest resolution corresponding to the period of point sequential scanning. Coloring processing using multiple colors as described above is performed for each pixel for the difference between each pixel data and the reference image data, which is composed of each pixel data for one frame of the highest resolution corresponding to the cycle of point sequential scanning. For the difference between the image data to be determined whose resolution has been lowered by averaging each of multiple pixel groups and the reference image data which has been processed in the same way, the above image data can be calculated for each part whose resolution has been lowered. It may be colored with multiple colors such as . Alternatively, the image may be processed with an appropriate resolution using other well-known image processing means.

0012

[Operation] With this configuration, different colored displays are created depending on the size of the difference between the image data to be determined obtained by imaging during the determination process and the reference image data set in advance, and the threshold level is The coloring state changes as the value is changed appropriately. Therefore, it is possible to easily adjust the threshold level for determining the presence or absence of an abnormality and the difference from the reference image data in an optimal state while looking at the colored display.

[0013] This is more visually appealing than the conventional case where adjustments are made by trial and error, and the tendency of the threshold level to obtain the optimum judgment state can be easily recognized. To facilitate adjustment operations and enable highly accurate adjustment.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. First, the configuration of the image processing apparatus will be explained based on FIGS. 1 to 3. This embodiment relates to an image processing device for monitoring whether or not abnormality occurs in a mold in a plastic molding machine. In FIG. 1, 1 is a movable mold provided in a molding machine, and 2 is a fixed mold, in which a series of injection molding processes are performed at predetermined timing by a drive means (not shown).

Reference numeral 3 denotes a monitoring video camera, and in this embodiment, it is installed to monitor the movable mold 1 for abnormalities. Then, the monitoring video camera 3 transmits a frame image signal of a still image obtained by imaging the movable mold 1 at a period of point-sequential scanning to the signal processing unit 4. 5
A display control section further processes data formed by performing predetermined signal processing to be described later in the signal processing section 4, and causes the display monitor 6 to display the processing results.

Reference numeral 7 denotes an operation panel, which the operator uses when adjusting the threshold level or instructing the imaging conditions of the surveillance video camera 3 by remote control. 8 is a timing control unit that generates a synchronization signal for synchronizing the molding operation of the molding machine and the image processing of the image processing device, and also performs signal processing when the signal processing unit 4 determines that an abnormality has occurred, which will be described later. A command signal from section 4 is transferred to the molding machine to perform controls such as stopping the molding operation.

FIG. 2 shows the configuration of the signal processing section 4 in more detail. That is, the signal processing unit 4 receives an image signal from the surveillance video camera 3, and the A/D conversion unit 41 converts each pixel signal into digital data ( The data corresponding to each pixel signal is hereinafter referred to as pixel data) and is stored in the image memory 42 formed of a random access memory (RAM) or the like according to the pixel arrangement. Therefore, image data corresponding to one frame of still image from the monitoring video camera 3 is stored, and this is used as image data to be determined.

Reference numeral 43 denotes a reference data memory for storing reference image data, which is formed of the same random access memory (RAM) as the image memory 42 and has a storage capacity equivalent to one frame image. 44 is a calculation unit consisting of a microcomputer or the like having a calculation function, and performs image processing on the image data stored in the image memory 42 and the reference data memory 43 to detect abnormalities in the movable mold 1 in the molding machine. Determine the presence or absence of. Furthermore, the threshold level data applied to this determination process is stored in advance in an internal register or the like, and can be changed as appropriate by instructions from the operation panel 7.

Here, in the reference data memory 43, n×m pixel data corresponding to point sequential scanning, d11,
d12, ..., d1m, d21, d22, ...
, d2m, ..., dn1, dn2, ..., dn
m is stored, and n×m pixel data corresponding to point sequential scanning are also stored in the image memory 42, D11, D12, ...,
D1m, D21, D22, ..., D2m, ...
, Dn1, Dn2, ..., Dnm are stored, the calculation unit 44 stores |d11-
D11|, |d12-D12|,..., |dnm-D
Subtraction of nm| and calculation of the absolute value are performed (hereinafter, this data will be referred to as difference data), and the resultant data is supplied to the display control section 5.

FIG. 3 shows the configuration of the display control section 5 in more detail. That is, in FIG. 3, 51 is a data buffer, which temporarily holds each difference data transferred from the arithmetic unit 44 in a dot-sequential scanning period. A color table 52 outputs data corresponding to the level of the difference data held in the data buffer 51. That is, the color table 52 has the function of a so-called look-up table formed of a random access memory (RAM), and the display colors corresponding to the difference data at the current threshold level are written therein. When the threshold level TH is changed, the display color is changed by rewriting the data in the color table 52. Then, the level of the difference data is divided into, for example, four levels, and when the difference data is in the range of the first level 0 to L1, the switching data DB specifying blue is output, and the level of the difference data is in the range of the first level 0 to L1. If it is within the range of L2, switching data DG specifying green is output; if it is within the range of third levels L2 to L3, switching data DY specifying yellow is output; If the range exceeds , switching data DR specifying red is output.

Reference numeral 53 denotes a display timing generator that generates a synchronization signal synchronized with the period of dot sequential scanning, and in synchronization with this synchronization signal, the data buffer 51 converts the difference data from the calculation section 44 into the timing of dot sequential scanning. is received in synchronization with the color table 52, and is further supplied to the color table 52.
2 also converts the difference data into switching data under the above conditions in synchronization with this synchronization signal and outputs it. Therefore, switching data converted for each pixel is output in time series in synchronization with the period of dot sequential scanning.

54 are three analog switches SW1, S
It is a switching unit having W2 and SW3, and the input side contacts of the respective analog switches SW1, SW2, and SW3 are supplied with pixel signals transferred in time series from the surveillance video camera 3 in synchronization with point-sequential scanning. and analog switch S
The output side contact of W1 is connected to the R (red) input terminal of monitor 6, the output side contact of analog switch SW2 is connected to the G (green) input terminal, and the output side contact of analog switch SW3 is connected to the B (blue) input terminal. .

Further, when the switching data DR is supplied, only the analog switch SW1 is conductive, and when the switching data DG is supplied, only the analog switch SW2 is conductive.
When the switching data DB is supplied, only the analog switch SW3 is conductive, and when the switching data DY is supplied, the analog switches SW1 and SW2 are conductive, and the instruction data is generated by the synchronization signal from the display timing generator 53. Because they are synchronized, the pixel signals transferred from the surveillance video camera 3 are transmitted to the monitor 6 without being distorted.
supplied to

The display timing generating section 53 also supplies the monitor 6 with a synchronization signal Sync for reproducing the image of the object based on the pixel signals supplied to the monitor 6. Next, the operation of the signal processing device having such a configuration will be explained. First, the movable mold 1 in a normal state is imaged by the monitoring video camera 3, and each pixel signal is transferred to the A/D converter 4.
1, it is converted into pixel data, and the data is transferred to the image memory 42 and the calculation unit 4.
4 to the reference data memory 43. Through this process, still image data corresponding to one frame image is stored in the reference data memory 43. In addition, since images are taken according to the monitoring items, for example, when monitoring the presence or absence of abnormalities in the molded product upon completion of mold opening, the state in which the molded product is present in the movable mold 1 is captured and imaged as image data. When storing or monitoring whether there is any abnormality in the movable mold 1 after the molded product has been removed from the movable mold 1 at the start of mold clamping, an image of only the movable mold 1 must be prepared in advance. Preprocessing is performed according to the monitoring item, such as storing data in the reference data memory 43.

Next, the molding machine is operated, and the monitoring video camera 3 takes an image and the image processing device performs processing periodically under the same conditions as the image taken during the pre-processing. This processing timing is set by the operator from the operation panel 7. Then, when the pixel signal output from the surveillance video camera 3 is converted into pixel data by the A/D converter 41 and stored in the image memory 42, the arithmetic unit 44 converts the reference data memory 43 and the image memory into pixel data. The subtraction processing described above is performed on each of the 42 pixel data, and this difference data is sequentially transferred to the display control unit 5 in synchronization with the timing of point-sequential scanning of the monitoring video camera 3.

The display control section 5 converts the difference data into switching data according to a predetermined level in the color table 52, and sequentially transfers the data to the switching section 54, which switches the analog switches SW1 to SW1 according to the switching data. SW3 is selected and switched to a conductive state. As a result, monitor 6 has
Depending on the level of the pixel signal supplied from the surveillance video camera 3, images colored red (R), yellow (Y), blue (B), and green (G) are reproduced, and the coloring will change depending on the difference between the level of the difference data and the threshold value TH.

The relationship between the difference between the reference image data and the judged image data and the threshold level for the colored image reproduced on the monitor 6 is set as shown in FIG. By changing the magnitude of the threshold level with respect to the absolute value of the difference in image data, the coloring of the monitor 6 can be changed to red, yellow, blue, or green even if the difference is the same. This correlation data is stored in the color table 52 as a lookup table. Furthermore, this coloring process is shown in Figure 5.
- Explanation will be made based on FIG. 8.

For example, the absolute value of the difference between the pixel data for each pixel stored in the reference data memory 43 and the pixel data for each pixel temporarily stored in the image memory 42 is
As shown in FIG. 5, it is assumed that the distribution is as shown by the Z-axis value with respect to the imaging plane shown by the XY coordinates. Then, assuming that the threshold level is set to a certain low value th1, the coloring conditions at the threshold value th1 are written into the color table 52 based on the data in the color table 52 based on a command from the calculation unit 44. Since the color table 52 outputs switching data based on coloring conditions set in advance for the threshold level th1, the monitor 6 displays red (R) as shown in FIG.
, yellow (Y), blue (B), and green (G).

Furthermore, when the threshold level is set to a higher level th2, the data in the color table 52 is rewritten to the coloring conditions for the threshold value th2. Then, based on the coloring conditions set in advance for the threshold level th2, the monitor 6 is colored with a different color, as shown in FIG. 7, for example. Furthermore, the threshold level is set to a higher level t
When set to h3, the data in the color table 52 is rewritten to the coloring conditions for the threshold th3. and,
Based on coloring conditions preset for the threshold level th3, the monitor 6 is colored with a different color, as shown in FIG. 8, for example.

[0030] By appropriately changing the threshold level in this manner, the difference data based on the threshold level is colored and displayed on the monitor 6 in real time. In this embodiment, the operator can instruct the calculation unit 44 from the operation panel 7 to change the threshold level TH as appropriate, so that by adjusting the threshold level TH, the reproduced image can be colored. can be changed for each part.

Therefore, for example, if a lot of noise occurs during monitoring, by setting the threshold TH at a level slightly higher than the noise level while watching the coloring of the reproduced image on the monitor 6, the noise component can be reduced. The value of the difference data obtained by adjusting it so as not to be influenced or by subtracting processing in the calculation unit 44 will be smaller in the non-abnormal part than in the abnormal part, so for example, if the difference between these values is By appropriately adjusting the threshold value TH to the level, it is possible to obtain an effect such as making the abnormal part and the normal part colored differently to make them easier to identify.

[0032] It should be noted that such adjustment of the threshold level TH is just an example, and the operator can set the coloring conditions as appropriate.
It becomes possible to perform highly accurate image processing. In addition, since the coloring of the reproduced image changes in real time each time the threshold level TH is changed, it is easy to see how the threshold level TH adjusted by the operator tends to correspond to the coloring conditions desired by the operator. Since it can be recognized, adjustment is extremely easy and allows highly accurate adjustment.

Next, another embodiment will be explained based on FIG. 9. Differences from the first embodiment described above will be described. The overall schematic configuration is the same as that in FIGS. 1 and 2, but as shown in FIG. 9, the configuration and function of the display control section 5 are different. That is, in this embodiment, as shown in FIG. 9, no image is reproduced for the pixel signals from the surveillance video camera 3;
Red (R), blue (B), green (G), yellow (Y) switching data DR, DB, DG, DY are directly input to the RGB input terminal of monitor 6 as binary data of “L” and “H”. It is designed to supply

As a result, the switching section 54 shown in FIG. 3 becomes unnecessary, and the configuration can be simplified. In addition, the first example is as follows.
Regarding the pixel signal from the surveillance video camera 3, red (R
), blue (B), green (G), or yellow (Y) to reproduce the image, it is possible to reproduce the image clearly with the object to be determined colored. However, in this embodiment, since the gradation control of the reproduced image is not performed, the sharpness is lowered, but depending on the field of application, it is possible to provide an image processing apparatus that does not have any practical problems.

[0035] In these two examples, the four colors of red (R), blue (B), green (G), and yellow (Y) were used, but the invention is not limited to this. By adding or changing the configuration of the color table 52 and the switching unit 54 so as to correspond to a large number of colors, it is also possible to display a larger number of colors.

[0036]

As explained above, according to the present invention, the value of the threshold level can be appropriately changed by adjustment, etc., and the difference between the image data to be judged and the reference image data can be adjusted according to the above-described method. By displaying a plurality of different colors corresponding to the magnitude of the difference on the display means based on the threshold level, the magnitude of the difference can be confirmed visually and intuitively. Therefore, it is possible to easily adjust the threshold level to determine the presence or absence of abnormalities and differences from reference image data in an optimal state while looking at the colored display, and at the same time, it is possible to improve the adjustment accuracy. can.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a block diagram showing in detail the configuration of a signal processing section in FIG. 1;

FIG. 3 is a block diagram showing in detail the configuration of a display control section in FIG. 1;

4 is an explanatory diagram for explaining the function of a color table 52 in FIG. 3. FIG.

FIG. 5 is an explanatory diagram for explaining the operation of one embodiment.

FIG. 6 is an explanatory diagram for further explaining the operation of one embodiment.

FIG. 7 is an explanatory diagram for further explaining the operation of one embodiment.

FIG. 8 is an explanatory diagram for further explaining the operation of one embodiment.

FIG. 9 is a block diagram showing the configuration of a display control section, which is a different part, for explaining the configuration of another embodiment of the image processing apparatus according to the present invention.

[Explanation of symbols]

1; Movable side mold 2; Fixed side mold 3; Surveillance video camera 4; Signal processing section 5; Display control section 6; Monitor 7; Control panel 8; Timing control section 41; A/D conversion section 42; Image memory 43; Memory for reference data 44; Arithmetic unit 51; data buffer 52; Color table 53;Display timing 54; switching section

Claims (1)

[Claims] Claim 1: Store reference image data of an object to be determined in advance, and compare it with image data to be determined obtained by capturing an image of the object during image processing, and calculate the difference between the reference image data and the image data to be determined. In an image processing apparatus that determines a change in an object by detecting a location where the threshold value exceeds a predetermined threshold level, means for appropriately changing the value of the threshold level;
The size of the difference between the judgment image data and the reference image data is
Signal processing means for detecting based on the threshold level,
An image processing apparatus characterized by comprising display control means for causing a display means to display a plurality of different colors in correspondence with each other according to the magnitude of the difference obtained by the signal processing means.