JPS61250540A - Apparatus for confirming floc image in water purification plant - Google Patents

Abstract

PURPOSE:To make it possible to detect the abnormality of illumination equipment by an inexpensive apparatus and to attain to enhance the processing reliability of an image, by providing a check pattern to the backscreen provided at the position opposed to TV in order to enhance the confirmation accuracy of a floc group. CONSTITUTION:A back screen 24 is formed as a black color one in order to confirm a floc group in a high contrast with high accuracy with due regard to that the color of floc 12 is white. By using this back screen 24 to which a white check pattern was provided, image information is not judged to be black as a whole when an illumination apparatus 40 is normal and no floc larger than the size capable of being confirmed on the basis of one picture element is present in image information taken in TV30 and only the part of the check pattern 2 is judged as white. However, when the illumination apparatus is abnormal, the whole is judged to be black and, therefore, it is judged that the illumination apparatus 40 has abnormality. As mentioned above, the abnormality of the illumination apparatus is detected by an inexpensive apparatus.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] □ The present invention is a water floc image recognition device that measures the particle size and distribution of floc in a floc formation pond (mixing pond) of a water treatment plant using image processing technology. Regarding.

[Background of the invention]

In water purification plants, the suspended particles in raw water are small in size, so they are aggregated to form flocs, and the flocs are allowed to settle. For this reason, it is essential to monitor flocs in the floc formation pond (mixing pond).

Conventionally, flocs have been visually monitored several times a day by water treatment plant maintenance managers. Since it relies on visual inspection, the judgment criteria are subjective and qualitative, and the monitoring results have the disadvantage of being difficult to reflect in driving operations.

Furthermore, since the monitoring frequency is discontinuous, countermeasures against agglomeration failure cannot be taken late, which increases troubles.

In contrast, a method has recently been adopted in which an industrial television camera (ITV) is used to monitor the flocs in the floc formation pond. However, even in this case, monitoring remains subjective and discontinuous because it relies on human vision.

In order to improve these drawbacks, Japanese Patent Application Laid-Open No. 54-1432
As described in Japanese Patent No. 96, a method has also been proposed in which a photoelectric conversion device is used to extract an electrical signal according to the shape of the floc. However, in practical use, there are problems such as illuminance and abnormality determination of lighting equipment, and unless these problems are solved, an unmanned monitoring device that performs continuous monitoring using only the monitoring device cannot be put into practical use.

[Purpose of the invention]

An object of the present invention is to provide an inexpensive drinking water floc image recognition device that can objectively and continuously measure the particle size and distribution of floc groups with high reliability.

[Summary of the invention]

A feature of the present invention is that in order to improve the recognition accuracy of flock groups, abnormalities related to the lighting system can be detected only by the image processing device by using a check pattern equipped with check points on the pack screen located opposite the television camera. That's what I did.

[Embodiments of the invention]

Figure 2 is a diagram of the floc formation pond at the water treatment plant. Second
In the figure, 10 is a flocculation pond, 11 is a stirring paddle, 12 is flocs, and 13 is a rectifying wall. A group of minute flocs that have flowed in from a rapid mixing pond (not shown) in the front stage of the flocculation basin 10 are stirred by the stirring paddle 11 in the flocculation basin 10, so that the flocs 12 collide with each other and coagulate. . That is, the particle size of the flocs increases. Usually, the floc formation pond 10 is composed of three ponds, each of which is partitioned by a flow wall 13. The particle size of the floc 12 gradually increases as it passes through the three ponds f.

FIG. 1 shows an embodiment of the present invention in which image recognition of a group of flocs within a floc formation pond 10 is performed.

In FIG. 1, 20 is an airtight container, 21 is an observation window, and 22
is a wiper, 23ti wiper drive device, 24 is a pack screen, 24A and 24B are back screen fixtures, 25A, 25B, and 25C are respective baffle plates, 30
is an industrial television camera (ITv), 31 is a close-up lens,
4o is a lighting device, 32 is an ITV controller, 41 is a lighting device controller, 42 is a light shielding cover, 5o is an image recognition device, and 60 is an image recognition control device.

The purpose of such a water floc image recognition device is to monitor the final degree of floc formation. It is arranged near the exit of the third pond of the Tsuku formation pond 10. Moreover, when monitoring the process of floc formation, it is installed in the first or second pond of the floc formation pond 1o.

The configuration shown in FIG. 1 will be explained in more detail.

1 TV 30 fixed in the airtight container 20 has a close-up lens 3
1. Observation window 21 made of transparent material such as glass
The image of the flocs 12 in the floc formation pond lo is enlarged and recognized through the image. The wiper 22 operates periodically to remove dirt from the surface of the observation window 21.

Moreover, the pack screen 24 is installed as follows. In order to accurately recognize flock groups with high contrast, back screen fixture 24A is installed in 1 airtight container 2o.
and 24B, the pack screen 24 is connected to the observation window 2.
10 Install on the opposite surface. Considering that the flock 120 color is white, the pack screen 24 is desirably black in order to accurately recognize the flock with high contrast.

Next, the lighting device [40 and the light shielding cover 42 will be explained. The floc formation pond 10 is opened to the atmosphere for the purpose of constantly monitoring flocs. For this reason, the floc formation pond 10
The amount of light incident on the area changes over time and is strongly influenced by the weather.

A lighting device 40 is usually installed to constantly monitor the flock. If the purpose is to simply monitor the maintenance personnel's vision, it is sufficient to install the lighting device ff140, and the lighting device 40 may be installed above the floc formation pond 10.

However, even if the illumination device t40 is installed, changes in ambient illuminance strongly affect the degree of image recognition Rnt of the flock group. For example, if the illuminance is low, the flocs will be perceived as small, and conversely, if the illuminance is high, the flocs will be perceived as large. In order to eliminate this influence, it is necessary to avoid being influenced by changes in illuminance as a natural phenomenon.

In this disaster example, a light-shielding cover 42 is provided to darken the surrounding area and provide illuminance under a certain condition using only the lighting device 40.

Next, flock picture 1 imported into I'f'V30
M! A foreign book of the image recognition device 50 that processes information will be explained.

FIG. 4 shows details of the image processing device 50. The image processing devices 1t and 50 perform various calculations such as the particle size and distribution of floc groups in order to extract useful information for water quality management from the image information obtained from the I'l:'V30. implement.

2 in FIG. 4, 100 is a recognition timing control circuit;
01 is an A/D conversion circuit, 102 is a threshold input circuit, 103
104 is a binarization circuit, 104 is a labeling circuit, 105 is a particle size calculation mode designation circuit, 106 is a particle size measurement circuit, 107 is a particle size comparison circuit, and 108A.

108B, 108C, . . . , 108Z are each number storage circuits, 109 is a recognition number control circuit, and 110 is a particle size distribution calculation display circuit.

FIG. 5 shows an image plane of a flock group recognized by the ITV 30. Since the floc group is a grayscale image, the boundary between the floc 12 and water is not actually clear, but for simplicity, only the outline of the floc group is illustrated. The brightness level of the flock group is high and white, while the back screen 2 is in the background.
4, the brightness level of the water is low and black.

In Figure 4, i! ! ! The identification timing control circuit 100 includes an ITV 30 and an ITV controller 32 shown in FIG.
The time interval (frequency) at which the image information is captured is controlled. The time interval for importing may be a time sufficient for the floes previously imported by the ITV 30 to move out of the processing range. Next, the A/D conversion circuit 101 receives an analog signal of the obtained luminance information, for example, the screen signal shown in FIG. 5, and converts the signal one by one into a digital signal. The converted digital signal is processed based on the threshold specified by the threshold input circuit 102.
It is binarized in a binarization circuit 103.

For example, FIG. 6 shows a case where the screen of FIG. 5 is scanned along line AA' to display the distribution of brightness levels. Here, the brightness level is displayed in 8 bits (256 levels), with lower brightness toward the top of the vertical axis and higher brightness toward the bottom. Since the flock 12 is white, its brightness is high. In other words, the downwardly valleyed portion represents the floc.

In this luminance distribution, each pixel is binarized by the binarization circuit 103 based on a threshold value specified by the threshold input circuit 102, for example, the luminance specified by the BB' line.

The threshold value specified by the threshold value input circuit 102 is maintained constant under constant illuminance, but can also be operated by an operator.

In the binarization circuit 103, pixels whose luminance level is higher than the threshold value are set to the '1' level, while pixels whose luminance level is higher than a predetermined value are set to the '0' level.As shown in FIG. The part corresponding to flocs is at the "1" level, and the part corresponding to water is at the @θ" level.

FIG. 8 shows an example of the results of floc recognition performed in this manner.

Once the flocs are recognized, the representative grain size of the flocs is calculated, but before that a nanno (-) is added to each floc.

That is, the labeling circuit 104 independently recognizes each flock and assigns a number to each flock. Then, each of the nano-flocs (-) is transferred to the particle size measuring circuit 10.
6. Calculate the representative particle size.

Representative particle diameters include directional diameter DCS, maximum diameter Dmaxs, minimum diameter Dmins, area circle equivalent diameter Dcir, and equivalent peripheral length Del. The grain size calculation mode designating circuit 105 designates a representative grain size to be adopted from among these representative grain sizes. In this way, seven particle sizes are calculated for each floc in accordance with the specified standard of representative particle size. The particle size comparison circuit 107 compares the particle sizes of each floc and stores the number of flocs having each particle size in the corresponding storage location, that is,
Number storage circuits 108A, 108B, 108C.

. . . The numbers are stored in 108z. Since the image of the floc is binarized, the minimum unit for measuring the particle size is one pixel. Therefore, if the number of flocs corresponding to each particle size is tl-Ni, for example, the number storage circuit 108 stores the number N1 of flocs whose particle size corresponds to one pixel, and the number storage circuit 108B stores the number N1 of flocs whose particle size corresponds to one pixel. The number N2 of flocks corresponding to two pixels is stored.

By the way, in order to accurately determine the particle size distribution of flocs,
It is better to obtain the particle size distribution from a large amount of information by capturing the screen multiple times. The number of recognition times control circuit 109 is designated with the number of times the flock image is to be recognized, and if the number of times the flock image is to be recognized is less than the number of times of recognition, the process returns to the recognition timing control circuit 100 . The recognition timing control circuit captures the image at the specified timing, repeats the operations described above, calculates the particle size of the flocs, and stores the number of each floc in the number storage circuit 108A.
It is stored in 108B, 108C, . . . 108Z. On the other hand, when the number of recognition times specified by the recognition number control circuit 109 is reached, the particle size distribution calculation and display circuit 110 outputs the number storage circuits 108A and 108B.

1080, Mountain... 1 Calculates the number concentration distribution for each particle size based on the oszO value and outputs the result.

Next, abnormality detection in lighting equipment will be explained. Abnormalities in lighting equipment include power outage, disconnection, burnout of lamps,
This may be caused by something blocking the light attached to the lamp itself.

First, in order to simplify the explanation, a shading 6I<-42 is provided, the surroundings are darkened, and the illuminance is provided under a certain condition only by the illumination device 40. If an abnormality occurs, the illuminance will be lost, so IT
The image information entered by *υ from V30 is determined to be entirely black. The image information processed by the binarization circuit 103 is as shown in FIG. On the other hand, even if the illumination device is normal, if there is no floc larger than the size that can be recognized by one pixel in the image information captured by I'I'V30, the entire image information will be black. It is judged that.

Therefore, the image information processed by the binarization circuit 103 becomes as shown in FIG. This is because, considering that the color of the flock 12 is white, the color is black in order to recognize the flock with high contrast and accuracy.

From the above, it cannot be determined that the lighting equipment is abnormal only from the image information that the entire image is black. In the present invention, a board 24 as shown in FIG. 3 is used.

In the backboard 24, a checkmark is placed in the entire image information 1 taken in from the ITV 30, and a turn 2 is turned off. Check turn 2 is a white color and the area is 1 pixel.
It is sufficient if the area of the TV 30 is equal to or larger than M#.

By using a check board (equipped with a check mark turn 2 as shown in Figure 3), the illumination device Jlf is normal and the image information captured by the ITV30 rarely has a size larger than that which can be recognized by one pixel. If there were no large flocks, the image information would not be judged as black, and the check/
Only the part of the turn is considered white. For this reason, when a check markter like that shown in FIG. 3 is used, the image information processed by the dichotomization circuit 103 becomes as shown in FIG. 10. However, if the lighting equipment is abnormal, the image information is determined to be black (because it is judged as black), the image information processed by the binarization circuit 103 becomes as shown in FIG. If it is black, it can be determined that there is an abnormality in the lighting equipment.

If the entire image information is black, it can be determined that there is an abnormality. There are various ways to determine that the entire image information is black, but to give an example, it is determined by the lighting equipment abnormality detection circuit 3 in the next stage of the binarization circuit in the image recognition device 50. The lighting equipment abnormality detection circuit 3 is labeled.

Similarly to the programming circuit 104, a number is assigned to each flock and the number N of flocks is extracted. Here, check pattern 2 is also labeled as a flock. When it is determined that the number of flocs is Nt - zero as a result of labeling, it is determined that the lighting equipment is abnormal, and the process moves to the alarm output determination circuit 4. If the number N of flocs is 1 or more as a result of labeling, it is determined that the lighting equipment is normal, and the processing moves to the labeling circuit 104. However, labeling circuit 10
In step 4, processing is performed with the check pattern 20 portion excluded from the labeling target.

Ending the process after the lighting equipment abnormality detection circuit 3 detects an abnormality only once tends to result in erroneous detection. For example, a case may be considered in which a black object overlaps check pattern 2 (441 mu rumors + 1st day of opening, 1 day of opening, 441 mu rumors, 1 day of opening, 1,000 yen), etc.). For this reason, it is necessary to operate the wiper or the like, capture the image, and re-judge the image based on new image information. The alarm output judgment circuit 4 stores the number of abnormality detections, and if the number of tries exceeds the set value, the process is transferred to the abnormality alarm output circuit 5. If the number of retries is within the set value, the process is transferred to the wiper drive circuit 6, and the wiper drive circuit operates the wiper driver 23 to operate the wiper <-22. A wiper is attached to the device and operated.Then, the processing is transferred to the recognition timing control circuit 100, a new image is taken in, and the above-mentioned processing is performed.The flow of the abnormality processing described above is shown in FIG.

Note that in order to simplify the explanation of abnormality detection in lighting equipment, the light-shielding cover 42 was provided, the surroundings were darkened, and the illuminance under certain conditions was explained only by the lighting device 40. However, without the light-shielding cover 42, Even when there is an influence of light, it is true that there is a difference in the brightness level of the check turn 20 section taken from ITv30 when the lighting equipment is normal and when it is abnormal.

Of course. By setting an appropriate threshold value in the threshold input circuit 102, if the lighting equipment is abnormal, check pattern 20 elements: 2 (can be removed by the fi conversion circuit 103. The converted image information is as shown in Fig. 9. Based on this binary information, even if it is affected by the shadow 4 of external light, the lighting equipment abnormality detection circuit 3
It is possible to detect abnormalities.

〔Effect of the invention〕

As explained above, according to the present invention, anomalies can be detected without the need for special equipment for detecting anomalies in the lighting equipment of an image processing device. It enables highly reliable image processing and continuous unattended processing.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an embodiment of the present invention, Fig. 2 is a block diagram of a floc formation pond in a water purification plant, Fig. 3 is a perspective view of a backboard with a check pattern, and Fig. 4 is the same as that shown in Fig. 1. Detailed configuration diagram of the image processing device, Figures 5 to 7 are waveform diagrams showing the signal processing process, Figure 8 is a binary image diagram obtained by implementing the present invention, and Figure 9 is a binary image at abnormal times. Figures 10 and 10 are binary image diagrams at normal times, and Figure @11 is an abnormality processing 70 diagram of the present invention. 2... Check pattern, 3... Lighting equipment abnormality detection circuit, 4... Fresh alarm output judgment circuit, 5... Abnormal J alarm output circuit, 6... Wiper drive circuit, 10... Flock Formation pond, 11... Stirring paddle, 12... Flock, 20... Airtight container, 21... Observation window, 24...
...Back screen, 25A, 25B, 25C...
Baffle board, 30...Industrial television camera, 40...
Illumination device, 42... Light shielding cover, 50... Image recognition device, 60... Image recognition control device, 100... Recognition timing control circuit, 101... A/D conversion circuit, 1
02... Threshold input circuit, 103... Binarization circuit, 1
04...Labeling circuit, 105...Particle size calculation mode designation circuit, 106...Particle size measurement circuit, 107...
Particle size comparison circuit, 108A. 108B, 108C,...108Z...Number storage circuit, 109...Recognition number control circuit, 11o
...Particle size distribution calculation display circuit, 200, 201...2
Value image, 203...Check pattern.

Claims (1)

[Claims] 1. A lighting device that irradiates light onto a group of flocs in a flocculation pond of a water purification plant, and an image recognition device that performs image processing by photoelectrically converting image information obtained by capturing an image of the flocs received by the illumination device with a television camera; A water flock image recognition device comprising a back screen disposed opposite to the television camera, and a check pattern provided on the back screen.