JPH0424913B2 - - Google Patents

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an image recognition method for activated sludge, and is particularly applicable to an aeration tank in equipment for purifying wastewater using the activated sludge method in sewage treatment, etc. The present invention relates to an image recognition processing method for recognizing and evaluating the properties of activated sludge.

[Conventional technology]

The activated sludge method is a water treatment process that is used in a wide range of fields such as sewage treatment. The requirements for carrying out this method are to ensure the decomposition and assimilation reaction of organic pollutants by microorganisms in the presence of oxygen, and to ensure effective sedimentation and separation of the generated activated sludge flocs. Depending on the fluctuation,
Situations where good precipitation separation cannot be ensured occur from time to time, and conventionally it has been considered one of the water treatment processes that is relatively difficult to manage.

In order to ensure good sedimentation separation, the activated sludge flocs produced must be dense and have a large particle size, that is, have a high sedimentation rate. Conventionally,
In terms of total activity (macro), indicators such as SVI have been used to evaluate sedimentation and provide information for operation management. However, since the SVI index itself was based on manual separation operations, it lacks real-time performance, and there is no clear threshold value for the numerical value, so it is difficult to judge the quality of precipitation separation based on experience. It had characteristics that made it difficult to automate or bring it online, such as requiring the intervention of subjective judgment.

Against this background, in recent years, attempts have been made to obtain microscopic information about activated sludge by importing enlarged images of activated sludge using a microscope, etc. into an image processing device using a television camera, and processing the images. . However, these are recognition counts for protozoa on the one hand and filamentous bacteria that interfere with sedimentation on the other hand, and do not deal with the density of the activated sludge floc itself.

[Problem to be solved by the invention]

As mentioned above, in the prior art, no technology has been developed to deal with the density of activated sludge flocs.

Therefore, an object of the present invention is to provide an image recognition method for quantifying and evaluating the density of activated sludge flocs and enabling stable operation control of activated sludge flocs.

[Means to solve the problem]

The present inventors believe that the factor governing sedimentation separation is not only the amount of filamentous bacteria but also the density of the activated sludge floc itself.As a result of intensive research, the inventors have quantitatively determined the density of the activated sludge floc. This invention has achieved the above objectives by inventing a novel image recognition processing method that can be expressed.

That is, the present invention provides a method for enlarging activated sludge imaging and image processing, which includes: (1) a first step of calculating a difference between a brightness value of original image data and a brightness value of background image data; a second step of binarizing the image obtained in step 1; (3) a third step of binarizing the image obtained in the first step after line detection processing; (4) a fourth step of calculating the sum of the brightness values of the image data obtained in the second step and the image data obtained in the third step to obtain a mask image; (5) the first step; (6) calculating the sum of the brightness values of the image data obtained in the fourth step and the image data obtained in the fourth step to obtain an enhanced image; (6) calculating the image obtained in the fifth step; a sixth step of performing binarization processing; (7) a seventh step of reducing and enlarging the image obtained in the sixth step to obtain only an image of activated sludge flocs; (8) a seventh step. an eighth step of integrating the number of pixels of the image obtained in the first step and calculating the area and/or volume of the activated sludge floc; (9) combining the image obtained in the first step with the seventh step; a ninth step of calculating the logical product of the obtained images; (10) a tenth step of binarizing the image obtained in the ninth step using a plurality of brightness values; (11) a tenth step. (12) For each of the plurality of images obtained in the 11th step, the number of pixels is integrated, and the area and area of the activated sludge flocs are calculated. (13) the ratio of each area of the activated sludge flocs obtained in the twelfth step to the area of the activated sludge flocs obtained in the eighth step; No.
(14) a thirteenth step of calculating at least one of the ratios of each of the volumes of activated sludge flocs obtained in the twelve steps to the volume of the activated sludge flocs obtained in the eighth step; A fourteenth step of determining the density of the activated sludge flocs based on the calculated value obtained in the step.

In the above, activated sludge is a general term for those formed by floc-forming bacteria, filamentous bacteria, protozoa, etc., and the floc-like objects formed by floc-forming bacteria are called activated sludge flocs. Filamentous objects that are free independently or extend radially from the flocs are called filamentous bacteria, and the present invention quantitatively recognizes activated sludge flocs.

[Effect]

In the image recognition method according to the present invention, by performing predetermined image calculation processing on an enlarged image of activated sludge flocs generated in an activated sludge process such as sewage treatment, its density can be quantified and evaluated. This enables diagnosis of the quality of activated sludge flocs, thereby enabling effective operation management and automatic control.

〔Example〕

Hereinafter, the present invention will be explained in more detail with reference to the drawings, but the present invention is not limited to this embodiment.

First, explanation will be given using FIG. FIG. 1 is an explanatory diagram showing an example of operation of the activated sludge image recognition processing method according to the present invention.

The raw water 1 is pretreated if necessary and then introduced into the aeration tank 2, where it is mixed with activated sludge 12 returned from the settling tank 4. Air is supplied into the aeration tank 2 from a blower 3 in order to supply oxygen necessary for activated sludge activity. Activated sludge decomposes and assimilates organic substances in wastewater while remaining in the aeration tank. The effluent from the aeration tank 2 is sent to a settling tank 4, where the activated sludge is separated by sedimentation to obtain clear treated water 7. A portion of the precipitated activated sludge becomes return sludge 12, and the remainder is discharged outside the system as surplus sludge 13.

An underwater microscope 8 suitable for obtaining enlarged images of activated sludge is immersed in the aeration tank 2, and images of the activated sludge are periodically captured and transmitted to an image recognition system 9 according to the present invention. . This underwater microscope 8 may be used, for example, in the patent application filed in 1982, which the present applicant previously filed.
A device suitable for this purpose is selected, such as that described in No. 279056.

The image recognition processing system 9 is composed of an image processing device, a computer, a monitor, a television, etc., and performs the necessary calculation processing on the video signal from the underwater microscope 8, and calculates an index for evaluating and determining the density of activated sludge flocs. do. The calculation results in the system 9 are transmitted via the controller 10, and the aeration blower 3, return sludge pump 5, and surplus sludge pump 6 are controlled as necessary.

Next, a procedure for processing image data in the image recognition system 9 will be described using a process diagram showing an example of the flow of the image recognition processing method of the present invention shown in FIG.

Original image data 21 input from an underwater microscope, etc.
is smoothed by multiple integral inputs if necessary, and then in order to remove uneven illumination of the input image,
A difference in brightness value is calculated 23 between the background image data 22 that does not include the target object (activated sludge flocs, etc.) (first step). Next, for the image obtained in the first step, follow the two processing procedures A and B.
Perform image processing.

In route A processing, the image obtained in the first step is subjected to binarization processing 24 at a predetermined brightness value (second step), and the image obtained in the first step is intermediate between the image obtained in the first step and the image obtained in the first step. After line detection processing 25 is performed using a Laplacian operator to emphasize bright areas (filamentous bacteria, areas around activated sludge flocs, etc.), binarization processing 26 is performed (third step) to create an image. The sum of the luminance values is calculated 27 between (fourth step).

Here, the size of the Laplacian operator is 3
A ×3 matrix or more, preferably a 9 × 9 matrix, is better, and the weighting coefficients are preferably arranged in about 12 different ways so that line segments with random directions can be emphasized in any direction. It's good to have things ready. After filtering using 12 different operators, a logical OR operation is performed to determine the brightness value of each pixel. By adding the obtained image and the image obtained in the second step, an image to be used as a mask is created. Note that the sum of the obtained brightness values is the maximum brightness value (in the case of 8 bits,
255) is considered the highest brightness value.

Next, the sum of brightness values between this mask image and the image obtained in the first step is calculated 29 in the same manner as in the fourth step to obtain an enhanced image (fifth step). At this time, invert the mask image in black and white,
After compositing, all the brightness values of the parts without objects are treated as the background and the brightness is converted. By binarizing the image obtained in the fifth step 30 at a predetermined level (sixth step), noise, uneven illumination, etc. are removed, and unclear filamentous bacteria and outlines around the flocs are removed. An accurate binarized image of the original image with clarity can be obtained.

Subsequently, in route A, in order to remove filamentous bacteria from the image and extract only flocs, reduction processing and enlargement processing 31 are performed as many times as necessary (seventh step). This number of times may be sufficient to remove the bacteria, taking into consideration the magnification of the sample and the thickness of the filamentous bacteria. The number of pixels of the obtained image is integrated and the area of the activated sludge floc is calculated 32 (eighth step).

For example, this can be calculated based on the number of pixels in the flock section and the unit pixel area (area of a square made of four pixels), or the number of pixels in the flock section relative to the total number of pixels on the screen. It may also be calculated as a ratio.

Next, the processing procedure for route B will be explained. B
In the route, an AND operation 33 is performed for each corresponding luminance value between the image obtained in the first step and the image obtained in the seventh step (ninth step). Here, the logical AND operation means that if the corresponding brightness values of both images are a certain value and 0 (zero), then the certain value is saved, and
Then, say the operation that saves 255. This makes it possible to remove filamentous bacteria and noise. Subsequently, the image obtained in the ninth step is subjected to binarization processing 34 using a plurality of brightness values (tenth step). The brightness value for this binarization process may be set arbitrarily based on knowledge regarding the distribution of brightness values of the activated sludge. For example, calculate the maximum diameter of the targeted activated sludge flocs, calculate the maximum brightness value and minimum brightness value from the brightness value distribution obtained by scanning along the maximum diameter direction, and calculate the maximum brightness value and minimum brightness value. It is preferable to set the brightness value as a value and at least one brightness value therebetween. As a result, a plurality of binarized images having different brightness values are obtained through the binarization process.

For each of the plurality of binarized images, in order to remove filamentous bacteria and the like, image reduction and enlargement processing 35 is performed a necessary number of times to extract only activated sludge flocs (eleventh step). This number of times may be sufficient to remove the bacteria, taking into account the magnification of the sample and the thickness of the filamentous bacteria. For each of the plurality of extracted images, the area of the activated sludge flocs is calculated and measured 36 in the same manner as in the eighth step (twelfth step).

Finally, for the plurality of images obtained on route B, the ratio 37 of the calculated value of flock area obtained on route A to the calculated value of flock area is calculated (13th step). Using the obtained ratio value, the degree of dispersion and aggregation of activated sludge flocs, that is, the density, is determined (fourteenth step).

The above content will be explained in more detail using FIG. 3. FIG. 3 is a process diagram showing the principle of the image recognition processing method of the present invention.

Sample S1 is an activated sludge floc in which microorganisms are loosely aggregated inside the floc from the periphery to the center, and sample S2 is an activated sludge floc in which microorganisms are densely aggregated.

Figure 3-1 is a schematic diagram showing a cross-section of a microscope preparation narrowed on both sides by glass or the like. The value distribution curve is as shown in Figure 3-2. in this way,
Activated sludge flocs are both understood as objects with ring spacing with high luminance values from the center to the periphery, and depending on how to process the intermediate luminance values in the periphery, This will depend on whether you can accurately recognize objects.

Therefore, in route A, an emphasis process is performed using a Laplacian filter on the intermediate luminance value corresponding to the peripheral area, and the ring interval in the peripheral area is exaggerated, thereby obtaining an accurate image of the object. procedures are adopted. That is, the flock image obtained by the procedure of route A is obtained as the one closest to the true image.

Next, processing is performed according to the procedure according to route B. As shown in Figure 3-2, the image of the flock can be obtained as the difference in the brightness value distribution obtained by scanning the cross section of the flock, so it can be binarized by the brightness values A to D. When the processing is carried out, the area of the flock portion is determined differently depending on each case, as shown in FIG. 3-3. Figures 3-4 plot these areas as a ratio to the flock area obtained by arithmetic processing according to the procedure of route A for each brightness value. In other words,
The more sparse the internal activated sludge flocs are, the larger the change in the extracted floc area will be due to the difference in brightness values in the binarization process, and the degree of this change can be used as an index to determine the density of the flocs. .

In addition, when using volume instead of area,
All you have to do is multiply the obtained area by the thickness between the slides.

The degree of change may be determined, for example, by considering it as a straight line (calculating a least squares correlation straight line) and determining its slope as shown in FIGS. 3-4.
In other words, in sample S2, the change in the recognition rate of area or volume due to the difference in brightness value is almost horizontal and smaller than in sample S1, indicating that the degree of microbial aggregation is dense up to the inside of the floc. It can be seen that the density of the floc can be evaluated using this method.

By accumulating a large amount of such data and referring to the database each time an observation is made, it becomes possible to determine the floc state.

〔Effect of the invention〕

As described above, according to the present invention, it becomes possible to quantify and evaluate the density of activated sludge flocs, which could not be done conventionally, and it is possible to achieve stable operation control of the activated sludge process.

[Brief explanation of drawings]

FIG. 1 is a schematic explanatory diagram showing an example of operation of the activated sludge image recognition processing method of the present invention, FIG. 2 is a process diagram showing an example of the flow of the present invention, and FIG. 3 is a diagram illustrating the principle of the present invention. FIG. 1... Raw water, 2... Aeration tank, 3... Aeration blower, 4... Sedimentation tank, 5... Return sludge pump, 6...
... Surplus sludge pump, 7 ... Treated water, 8 ... Underwater microscope, 9 ... Image recognition system, 10 ... Controller, 11 ... Precipitated sludge, 12 ... Return sludge,
13... Surplus sludge.

Claims (1)

[Claims] 1. A method for enlarging activated sludge and image processing, comprising: (1) a first step of calculating a difference between a brightness value of original image data and a brightness value of background image data; (2) a second step of binarizing the image obtained in the first step; (3) a third step of binarizing the image obtained in the first step after line detection processing; , (4) a fourth step of calculating the sum of the brightness values of the image data obtained in the second step and the image data obtained in the third step to obtain a mask image; a fifth step of calculating the sum of the brightness values of the image data obtained in the step and the image data obtained in the fourth step to obtain an enhanced image; (6) the image obtained in the fifth step; (7) a seventh step of reducing and enlarging the image obtained in the sixth step to obtain only an image of the activated sludge floc; an eighth step of integrating the number of pixels of the image obtained in step 7 and calculating the area and/or volume of the activated sludge floc; (9) the image obtained in the first step and the seventh step; (10) a tenth step of binarizing the image obtained in the ninth step using a plurality of brightness values; (11) a ninth step of calculating the logical product of the images obtained in An 11th step in which the images obtained in step 10 are reduced and enlarged; and/or a twelfth step of calculating the volume; and (13) the ratio of each area of the activated sludge flocs obtained in the twelfth step to the area of the activated sludge flocs obtained in the eighth step; or 12th
(14) a thirteenth step of calculating at least one of the ratios of the volumes of activated sludge flocs obtained in the eighth step to the volumes of the activated sludge flocs obtained in the eighth step; A fourteenth step of determining the density of activated sludge flocs based on the calculated value obtained in step 1.