JPH05146791A - Microbe recognizer - Google Patents

Abstract

PURPOSE:To accurately recognize a microbe in the recognizer for sewage disposal by obtaining a binary picture from the video signal of the microbe in liq., measuring the size, number and occurence frequency of the microbe, classifying the length of a specified microbe and appropriately diagnosing the binary value to correct the binarization threshold. CONSTITUTION:An image pickup device 200 is dipped in the sewage in an aeration tank 110 to pick up the image of the microbe, and the video signal is inputted to a job side monitor 340 through a job site control panel 330 and simultaneously inputted to a picture processor 310 through a central control panel 300. The video signal is binarized by the processor 310, the identification size, number, occurence frequency, etc., of the microbe are calculated, and the microbes are classified. During the process, the processed image and calculated value are indicated on a control monitor 320, the state of the microbe is judged, and an abnormality signal is sent to a control circuit 500, and the factor affecting the propagation of filamentous microganism is controlled by the control circuit 500 based on the abnormality signal and the measured value of a water quality measuring means 510.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a microorganism recognition device for monitoring microorganisms in a sewage treatment process or the like.

[0002]

2. Description of the Related Art In a sewage treatment plant, microorganisms are allowed to ingest organic matter in the inflowing water in an aeration tank and the microorganisms are settled in a sedimentation tank. In order to improve the quality of water discharged from sewage treatment plants, it is necessary to improve the shape of microorganisms. Microorganisms are roughly classified into agglutinating microorganisms and filamentous microorganisms, but if filamentous microorganisms proliferate too much, sedimentation becomes poor. The deterioration of the sedimentation property is called the bulking phenomenon. When the sedimentation property deteriorates, microorganisms flow out from the sedimentation basin, resulting in deterioration of discharged water quality. Therefore, it is important not to propagate filamentous microorganisms. As filamentous microorganisms, for example, Sphaeroterus (Sph
aerotilus). In order to prevent the filamentous microorganisms from multiplying too much, it is necessary to continuously and quantitatively measure the types of microorganisms and their appearance amounts or concentrations and reflect them in operation management. In this case, it is important to obtain an accurate state of the microorganism without disturbing the state of aggregation of the microorganism and its habitat environment. Conventionally, in order to monitor the state of microorganisms, a microorganism recognition device has been proposed which images the microorganisms by an imaging means and uses an image processing technique. Specifically, the image signal (grayscale image) picked up by the image pickup means is binarized by a threshold value, and then the microorganisms are extracted and recognized. A microorganism recognition device utilizing image processing technology is described in, for example, Japanese Patent Application Laid-Open No. 62-50608. In this prior art document, the threshold is uniquely determined based on the histogram of the grayscale image.

[0003]

When the threshold is uniquely determined by the histogram, the histogram (especially the brightness) of the imaged video signal changes due to the dirt of the lens of the underwater microscope as the image pickup means and the change of the illumination. Usually, the brightness of the histogram of the video signal decreases. Therefore, it becomes impossible to accurately extract and recognize the microorganism. Moreover, since the filamentous microorganisms and the flocculating microorganisms have different brightness,
The optimum threshold for binarization is also different. However, the thresholds are close. Therefore, there is a problem that the filamentous microorganisms and the flocculating microorganisms cannot be accurately extracted and recognized unless the thresholds for both microorganisms are optimized. In view of the above circumstances, an object of the present invention is to provide a microorganism recognition device capable of accurately recognizing microorganisms.

[0004]

An object of the present invention is to provide a microorganism comprising an image pickup device for picking up an image of microorganisms in a liquid, an image processing device for processing an image based on a video signal of the image pickup device, and a monitor device for displaying the image. In the recognition device, a binarized image obtained by binarizing the video signal is obtained, and the size, number, appearance frequency, etc. of the microorganisms are measured from the binarized image, and the length of the specific microorganism among the microorganisms is determined. This is achieved by performing classification, performing appropriate diagnosis of the binarization threshold value based on the classified information, and correcting the binarization threshold value so that the classification number is within a predetermined range. In this specification, the frequency distribution of the length of the microorganism is referred to as the length classification.

[0005]

The classification number of microorganisms is within a predetermined range regardless of the state of the sewage treatment plant, and the optimum binarized image of microorganisms can be obtained by changing the threshold value according to the classification number.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows the configuration of an embodiment of the present invention. In FIG. 1, the inflowing sewage is subjected to sedimentation and removal of large dust in a sedimentation basin 100, and then the aeration tank 11
It flows into 0. Further, return sludge (microorganisms) is supplied from the final settling tank 150 to the aeration tank 110 via the sludge return pipe 160. Air tank 11
At 0, the air sent from the proofer 140 through the air pipe 130 is diffused by the diffuser 120. As a result, the returned sludge and wastewater supplied into the air ratio tank 110 are mixed by stirring. The returned sludge, that is, activated sludge, is a lump (a floc) having a particle size of about 0.1 to 1.0 mm in which microorganisms are aggregated, and includes several tens of kinds of microorganisms, but when roughly classified, it is composed of aggregated microorganisms and filamentous microorganisms. .. Activated sludge absorbs oxygen in the supplied air and decomposes organic matter in wastewater into carbon dioxide and water. A part of the organic matter is applied to the bacterial growth of the activated sludge. The mixed liquid of activated sludge and wastewater is guided to the final settling tank 150, where the activated sludge gravity settles. The supernatant is usually discharged after chlorine sterilization.
On the other hand, the settled sludge is returned to the aeration tank 110 as return sludge from the sludge return pipe 160. The image pickup device 200 is arranged so as to be immersed in the liquid in the air ratio tank 110, and images the microorganisms in the air ratio tank. As the imaging device 200, for example, an underwater microscope, which the applicant of the present invention has previously filed as JP-A-2-182630, is used. The video signal (grayscale image) imaged by the imaging device 200 is input to the on-site monitor 340 via the on-site operation panel 330, and is also input to the image processing device 310 via the central operation panel 300. Site control panel 330
Supplies power to the image pickup apparatus 200, gives operation signals (site operation panel) such as a television camera installed in the image pickup apparatus 200, lighting, washing, and sampling, and also from the central operation panel 300. A control signal (central operation panel) is given to the imaging device 200 via the on-site operation panel 330. The on-site monitor 340 always displays the enlarged microorganism image from the imaging device 200. When operating the on-site operation panel 330, the operation is performed while referring to the microorganism enlarged image on the on-site monitor 340. This operation includes, for example, adjustment of illumination intensity. The central control panel 300 is a field control panel 330.
The video signal transmitted from the image processing device 3 is received.
10 and a device for transmitting a control signal for remotely operating the imaging device 200 to the site operation panel 330. Here, the control signal means, for example, the imaging device 20.
0 This is a signal such as a power ON / OFF command for a TV camera, an illumination device, a cleaning sampling start command, a camera diaphragm, a focus operation command, a motor control command, and the like. Image processing device 31
0 performs image processing of the video signal of the microorganism enlarged image from the image pickup device 200, and identifies the microorganism, the size, the number of the microorganism,
The appearance frequency and the like are calculated, the microorganisms are classified based on these calculated values, the processed image array calculated values obtained in each processing step are displayed on the central monitor 320, and the state of the microorganisms is determined, and The abnormal signal is transmitted to the control circuit 500. The abnormal signal includes, for example, excessive breeding of yarn (bulking) and a precursor of bulking. A conventional image processing device binarizes a video signal transmitted from an imaging device with a uniquely determined threshold value, and calculates the size, number, etc. of microorganisms based on the binarized image. These problems have been described in, for example, "Problems to be solved by the invention" in the present specification. The image processing apparatus of the present invention measures the length of microorganisms from a binarized image and corrects the threshold value for binarization so as to obtain a classification number within a predetermined range. This predetermined range is a result of preliminarily binarizing a plurality of images with an optimum threshold for each screen, measuring the characteristic amount of the microorganism, and obtaining and analyzing the frequency distribution of the length of the microorganism. An example of this predetermined range is shown in FIG.
As shown in (b), regardless of the state of the filamentous microorganism, the number distribution (frequency distribution) of each exists in the range where the curve E is the lower limit and the curve D is the upper limit. The image processing apparatus of the present invention can obtain the optimum binarized signal of the microorganism by changing the threshold value according to the classification number. Control circuit 5
00 controls factors affecting the reproduction of filamentous microorganisms based on the abnormal signal transmitted from the image processing device 310 and the measurement value of the water quality measuring means 510. This factor is, for example,
There are organic matter load, DO, pH, N and P balance, returned sludge amount, water temperature, excess sludge amount, inflow water amount, sludge retention time and the like. The control method, for example, adjusts the air flow rate for bulking that is considered to be caused by low DO, and adjusts the mixing ratio of return sludge and inflow water for bulking that is caused by low organic matter load. The water quality measuring unit 510 has measuring functions such as DO, MLSS, pH, alkalinity, and flow rate, and the measured values thereof are transmitted to the control circuit 500. The central monitor 320 displays the original image and the processed image transmitted from the image processing device 310, and the calculated value.

FIG. 2 is a diagram showing the structure of the image processing apparatus 310. The image signal from the image pickup device 200 is processed by the image processing processor 350 of the image processing unit 370, the identification of the microorganism, the size, the number of the microorganism, the appearance frequency, etc. are calculated, and these calculated values are passed through the system bus 507. , To the central processing unit 520. The image processing memory 360 stores the grayscale image from the imaging device 200, and also stores the processed image obtained from each processing step of the image processing processor 350. The central processing unit 520 is an image processing processor 350.
Based on the calculated value output from, the microorganism classification process, the threshold appropriate diagnosis, the threshold correction, and the microorganism state diagnosis are performed. Information necessary for these processes and programs and data are stored in the main memory 600 or the auxiliary memory 601.
Or stored in the disk 602.

FIG. 3 shows the relationship between the threshold and the image processing result. For example, FIG. 3A is an image obtained by performing image processing with an optimum threshold value T and extracting and recognizing a microorganism. On the other hand, FIG.
(B) is an image that has been image-processed with a threshold slightly higher than this threshold T, and FIG. 3 (c) is an image that has been processed with a threshold slightly lower than this threshold T. The respective threshold values in FIGS. 3B and 3C are values near the threshold value T in FIG. 3A. In the image of FIG. 3B, a large amount of noise n, which is not a filamentous microorganism, is generated and may be extracted and recognized as a filamentous microorganism. In the image of FIG. 3 (c), thread discontinuity appears, and filamentous microorganisms cannot be extracted as filamentous microorganisms.
In addition, the aggregating microorganism may be recognized as smaller than the real thing. As shown in the figure, it can be seen that even a slight deviation from the optimum threshold will give inaccurate results for image processing. Further, since the filamentous microorganisms and the flocculating microorganisms have different brightness, the optimum threshold value for binarization is also different. However, the thresholds are close. Therefore, unless the thresholds for both microorganisms are optimized, the filamentous microorganisms and the aggregating microorganisms cannot be accurately extracted and recognized. Therefore, it is desired to perform image processing with the optimal thresholds to extract and recognize the microorganisms.

FIG. 4 shows an example of the number distribution of filamentous microorganisms. The horizontal axis of FIG. 4 is the classification level, and the vertical axis is the number of filamentous microorganisms. The classification level refers to classification based on the length of filamentous microorganisms. In this classification method, for example, filamentous microorganisms having a length of 20 pixels or less are classified as level 1 and those having 20 to 50 pixels are classified as level 2. Note that FIG. 4 shows an example in which the longer the filamentous filament yarn, the higher the classification level. In FIG. 4A, a curve A is an example of the number distribution of filamentous microorganisms image-processed with a threshold slightly lower than the optimum threshold, and a curve B is
1 of the number distribution of filamentous microorganisms image-processed with the optimal threshold
This is an example, and the curve C is an example of the number distribution of filamentous microorganisms image-processed with a threshold slightly higher than the optimum threshold. The thresholds of the curves A and C are values near the optimum threshold.
Regarding the number distributions of the curves A, B, and C, it can be seen that the difference in the number of filamentous microorganisms is large at the low classification level, but the difference in the number of filamentous microorganisms is small at the high classification level. Further, FIG. 4B shows an example of a result of image processing of various images from the image pickup apparatus 200 for each screen with an optimum threshold value and obtaining the number distribution of filamentous microorganisms. FIG. 4B is a detailed view of the curve B in FIG. 4A. In FIG. 4B, the curve D is an example of the upper limit of the number distribution of filamentous microorganisms image-processed with the optimum threshold value. Yes,
Curve E is an example of the lower limit of the number distribution of filamentous microorganisms image-processed with an optimum threshold. In a plurality of images, regardless of the state of the filamentous microorganism, the number distribution of each exists in a range in which the curve E is the lower limit and the curve D is the upper limit.
Therefore, by analyzing the number distribution of filamentous microorganisms, the optimum threshold value can be determined or corrected to a threshold value falling between the curves E and D. Here, a similar distribution curve is obtained for the entire length of the filamentous microorganisms at the classification level instead of the number of filamentous microorganisms.

5 and 6 show an embodiment for determining the optimum threshold value in the present invention. FIG. 5 illustrates the image processing unit 370.
Shows the flow of. First, the image signal from the imaging device 200 is stored in the image memory 360 as a grayscale image. In the average brightness obtaining circuit 351, the density frequency distribution of the grayscale image stored in the image memory 360 is calculated, and the average brightness of the screen is obtained. Next, in the temporary threshold value determination circuit 352, a predetermined constant is subtracted from the average brightness transmitted from the average brightness determination circuit 351 to obtain the temporary threshold value of the filamentous microorganisms or the flocculating microorganisms. The predetermined constant is set by each processing plant. Subsequently, in the binarization processing 353, the grayscale images stored in the image memory 360 are binarized separately with the provisional threshold values of the filamentous microorganisms and the flocculating microorganisms, and the obtained binarized image is imaged. Send to memory 360. The recognition circuit 354 separately extracts and recognizes filamentous microorganisms and aggregating microorganisms based on the binarized image stored in the image memory 360. Known methods of extraction and recognition include degeneration, expansion, and thinning. Then, the measurement circuit 35
5, the size, number, appearance frequency, etc. of the microorganisms extracted and recognized by the recognition circuit 354 are calculated, and these calculated values are transmitted to the central processing unit 520 via the system bus 507.

FIG. 6 shows a flow of the central processing unit 520. According to the conventional technique, the central processing unit 520 has only the function of transmitting the calculated value transmitted from the arithmetic circuit to the storage device. However, according to the present invention, in the central processing unit 520, the central processing unit 520 has a classifying means and a proper diagnosis of the threshold. By providing a threshold value correction circuit and the like, the optimum threshold value is determined. First, the classification circuit 52
In No. 1, the calculated value from the image processing memory 360 is classified based on the length of the filamentous microorganism, the number distribution thereof is calculated, and stored in the main memory 600. As shown in FIG. 4 (b), the number distribution of filamentous microorganisms is 2 D and E.
Since it exists in the area range surrounded by the curved line, the upper limit curve D and the lower limit curve E of the area range can be used to perform appropriate diagnosis of the threshold value. Therefore, in the appropriate diagnosis 522 of the threshold value, the diagnosis is performed based on the number distribution of the filamentous microorganisms from the classification circuit 521. In this diagnosis, for example, when the number distribution is smaller than the predetermined reference curve E, it is judged that the threshold value of the filamentous microorganism is low, and the abnormal signal 1 is corrected by the threshold value correction circuit 529.
If the number distribution is larger than the predetermined reference curve D, it is judged that the threshold value of the filamentous microorganism is high and the abnormal signal 2 is transmitted to the threshold value correction circuit 529. The reference curves E and D are set by each processing plant. If the threshold appropriateness diagnosis 522 determines normal, an update process 523 is performed. On the other hand, in the threshold correction circuit 529, the threshold proper diagnosis 5
If it is determined to be abnormal in step 22, the respective threshold values of the abnormal signal 1 and the abnormal signal 2 are corrected, and the corrected threshold values are passed through the system bus 507 and binarized 353.
Send to. This threshold correction is performed by, for example, the abnormal signal 1
Is added to the threshold value of the filamentous microorganisms as a correction threshold value, and the abnormal signal 2 is subtracted from the threshold value of the filamentous microorganisms as a correction threshold value. In this case, the threshold value of the aggregating microorganisms remains unchanged or changes according to the threshold value variation of the filamentous microorganisms. Here, the binarization processing 353
Is a correction threshold value transmitted from the threshold value correction circuit 529, binarizes the grayscale image again, stores it in the image memory 360, and the recognition circuit 354 uses the binarized image stored in the image memory 360 based on the binarized image. Then, the filamentous microorganisms and the flocculating microorganisms are separately extracted and recognized, and the measuring circuit 355 calculates the identification of the microorganisms, the size and the number of the microorganisms, and transmits the calculated values to the central processing unit 520. The central processing unit 520 again performs classification processing of microorganisms, appropriate diagnosis of threshold values, correction of threshold values, and the like, based on the transmitted calculated values. By repeating this, it becomes possible to determine the optimum threshold value. The update processing 523 restarts the memory for storing the microorganism measurement result and updates the information of the microorganism measured at the final threshold value.

FIG. 7 shows another flow of the central processing unit 520. First, in the classification circuit 521, classification is performed based on the length of the filamentous microorganisms calculated from the image processing memory 360, and the number distribution thereof is calculated. The numbers of short threads and long threads of filamentous microorganisms are separately statisticized and stored in the main memory 600 in the order of measurement. A short thread is, for example,
Classifying levels 0 to 3 and lengths of filamentous microorganisms of not more than a predetermined reference value. A long thread is, for example,
Excludes short threads, and refers to filamentous microorganisms having a length equal to or greater than a predetermined reference value. For the storage order, for example, for short yarns, the number measured with the temporary threshold is stored in a, the number measured with the first correction threshold is stored in b, and the number measured with the second correction threshold is stored in c. These are stored in the main memory 600. The predetermined reference value is set by each processing plant. Next, in the appropriate diagnosis circuit 522 of the threshold value, diagnosis is performed based on the number distribution of the microorganisms from the classification circuit 521.
This diagnosis is, for example, as described in FIG.
If the number of short threads is smaller than the predetermined reference value E, it is determined that the threshold value of the filamentous microorganism is low, the abnormal signal 1 is transmitted to the threshold value correction circuit 529, and the number of short threads is set to the predetermined reference value. When it is larger than D, it is determined that the threshold value of the filamentous microorganism is high, and the abnormal signal 2 is transmitted to the threshold value correction circuit 529. Here, the correction threshold appropriateness diagnosis 524 has a function of checking whether or not the threshold value appropriateness diagnosis circuit 522 has an appropriate threshold value diagnosis. If the threshold diagnosis is not correct,
As shown in (b), a large amount of noise n, which is not a filamentous microorganism, is generated in the image, which is extracted and recognized as a filamentous microorganism, and as shown in FIG. 3 (c),
The image shows thread discontinuities, filamentous microorganisms cannot be extracted as filamentous microorganisms, and flocculating microorganisms are perceived smaller than the real one. Therefore, in the appropriate diagnosis 524 of the correction threshold value, the difference in the number of microorganisms is obtained from the number of the microorganisms measured twice, and the diagnosis is performed based on the difference in the number. The number difference is, for example, the difference between b and a,
It is the difference between c and b. From the microorganism number distribution graph shown in FIG. 4A, it can be seen that the difference in the number of filamentous microorganisms is large at a low classification level, and appropriate diagnosis of the correction threshold value is performed using the difference. The proper diagnosis of this correction threshold is
For example, in the threshold appropriate diagnosis circuit 522, when the provisional threshold value of the filamentous microorganism is determined to be higher than the reference value D, 1 is subtracted from the provisional threshold value of the filamentous microorganism to obtain the correction threshold value, and the correction threshold value is set to the system bus 507. To the binarization processing 353 of FIG. After that, the processing is performed according to the flow of FIG. As a result, in the proper diagnosis 524 of the correction threshold, the difference in the number of short yarns measured twice is obtained, and when the absolute value of this difference is larger than the predetermined reference value P, that is, the noise If no occurrence occurs, it is determined that the threshold correction of the threshold appropriateness diagnosis circuit 522 is proper, and the update processing 523 is performed. On the other hand, when the absolute value of the number difference is smaller than the predetermined reference value Q, it is determined that the threshold value correction of the threshold value appropriate diagnosis circuit 522 is inappropriate, and the abnormal signal 3 is transmitted to the threshold value correction circuit 529, and the correction signal is transmitted. Since 1 is subtracted from the provisional threshold value to obtain the correction threshold value, 1 is added to the corrected threshold value to restore the pre-correction threshold value. Then, when it is determined to be normal in the threshold appropriateness diagnosis 524, the update processing 523 is performed. that's all,
Although the threshold appropriate diagnosis circuit 522 has described the case where the temporary threshold value of the filamentous microorganism is determined to be higher than the reference value D, the correction threshold appropriate diagnosis 524 is such that the temporary threshold value of the filamentous microorganism is lower than the reference value E. The same is true when it is determined that The reference values P and Q are set by each processing plant. Further, the appropriate diagnosis of the threshold value based on the number of long yarns instead of the number of short yarns has the same effect. The threshold value correction circuit 529 performs threshold value correction on each of the abnormal signals 1, 2, and 3, and sends the corrected threshold value to the binarization processing 353 via the system bus 507.
For the threshold value correction, for example, for the abnormal signal 1, 1 is added to the threshold value of the filamentous microorganism to obtain a correction threshold value, and the abnormal signal 2
For example, there is a method in which 1 is subtracted from the threshold value of the filamentous microorganism to obtain a correction threshold value, and for the abnormal signal 3, the corrected threshold value is returned to the threshold value before correction. In this case, the threshold value of the aggregating microorganisms remains unchanged or changes according to the threshold value variation of the filamentous microorganisms.

As described in the above examples, the microorganisms in the activated sludge are imaged, image-processed, classified according to the length of the filamentous microorganisms, the number distribution thereof is determined, and the number distribution of the filamentous microorganisms is calculated. The optimum threshold value can be determined by appropriately diagnosing the threshold value and correcting the threshold value from the size of the aggregating microorganism. Microbes can be accurately recognized by the optimum threshold, and the amount of microbes can be quantitatively and quickly measured to accurately grasp the microbe state, and the bulking phenomenon can be detected early to suppress it. Also, it becomes possible to consider countermeasures at an early stage. Although the above examples have been described as examples applied to the sewage treatment process, the present invention, other bioprocess for culturing microorganisms, microorganisms such as lake bluegrass, monitoring of microscopic objects in liquids such as marine plankton. It is natural that the method can be applied to the above, and it can be applied to the abnormality diagnosis of the mixed liquid and the process control using the image measurement value of the minute object as an index.

[0014]

EFFECTS OF THE INVENTION According to the present invention, it is possible to perform image processing with an optimum threshold value and accurately recognize microorganisms in activated sludge, and to quantitatively measure the amount of microorganisms in a short period of time. You can As a result, the state of microorganisms can be accurately grasped, and it becomes possible to optimally control the operation management of the plant such as sewage treatment.

[Brief description of drawings]

FIG. 1 is a microorganism observation system for a sewage treatment process showing an embodiment of the present invention.

FIG. 2 shows a configuration of an image processing apparatus.

FIG. 3 shows a relationship between a threshold and an image processing result.

FIG. 4 shows the number distribution of filamentous microorganisms.

FIG. 5 shows a flow of the present invention.

FIG. 6 shows a flow of the present invention.

FIG. 7 shows another flow of the present invention.

[Explanation of symbols]

 110 Aeration tank 200 Imaging device 300 Central control panel 310 Imaging device 330 Field operation panel 350 Image processing processor 351 Average brightness determination circuit 352 Temporary threshold value determination circuit 353 Binarization processing 354 Recognition circuit 355 Measurement circuit 360 Image processing memory 370 image Processing unit 500 Control circuit 520 Central processing unit 521 Classification circuit 522 Threshold proper diagnosis circuit 523 Update processing 524 Corrected threshold proper diagnosis circuit 529 Threshold correction circuit

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁵ Identification number Internal reference number FI Technical display location G01N 21/84 Z 2107-2J G06F 15/62 395 9287-5L 15/64 400 L 8840-5L ( 72) Inventor Ichiro Enjo, Ichiro 4025, Kuji-cho, Hitachi City, Ibaraki Prefecture Hitachi Research Laboratory, Ltd. (72) Inventor, Tomonori Kaneko 5-2-1 Omika-cho, Hitachi City, Ibaraki Hitachi Ltd. (72) Inventor Naoki Hara 52-1 Omika-cho, Hitachi-shi, Ibaraki Prefecture Hitachi Ltd. Omika factory

Claims (7)

[Claims] 1. An imaging device for imaging microorganisms in a liquid,
An image processing device that performs image processing based on the video signal,
In a microorganism recognition device including a monitor device for displaying this image, a binarized image obtained by binarizing the video signal is obtained, and the size, number, appearance frequency, etc. of microorganisms are measured from the binarized image. , Of the microorganisms, the length of a specific microorganism is classified, and proper diagnosis of the binarization threshold is performed based on the classified information, and the threshold for binarization is corrected to have a classification number within a predetermined range set in advance. A microorganism recognition device characterized by: 2. An imaging device for imaging microorganisms in a liquid,
An image processing device that performs image processing based on the video signal,
In a microorganism recognition device including a monitor device for displaying this image, a means for determining a threshold value for binarization processing from the average brightness of the monitor screen, and a binarization processing means for binarizing the video signal using this threshold value. A means for storing a processed image output from the binarization processing means, a means for extracting and recognizing a microorganism based on the processed image, a means for measuring the size, number, appearance frequency, etc. of the microorganism, A means for classifying the length of a specific microorganism among the microorganisms based on the measurement value from the measuring means, a means for appropriately diagnosing the binarization threshold value based on this classification information, and the binarization threshold value is appropriate. A microorganism recognizing device, comprising: a threshold correcting unit that corrects when the liquid is not present, and monitors the state of microorganisms in the liquid. 3. An imaging device for imaging microorganisms in a liquid,
An image processing device that performs image processing based on the video signal,
In the microorganism recognition device comprising a monitor device for displaying this image, when there are two or more types of microorganisms in the liquid, a binarization processing means for performing a binarization processing with different thresholds depending on the object of extraction recognition, A means for storing each processed image output from the binarization processing means, a means for extracting and recognizing each microorganism based on the processed image, and a means for separately measuring the size, number, appearance frequency, etc. of each microorganism. And means for classifying the length of each specific microorganism of each microorganism based on the measurement value from the measuring means, and means for appropriately diagnosing the binarization threshold value for each microorganism based on the classification information. And a threshold value correction means for correcting when the binarization threshold value is not appropriate, and monitoring the state of microorganisms in the liquid. 4. The microorganism recognition device according to claim 2, wherein the classification means calculates a number distribution curve based on the classification information. 5. The binarization threshold appropriate diagnosis means according to claim 2 or 3, when the binarization threshold exceeds a predetermined reference curve of the number distribution curve, the binarization threshold is determined to be high. A microorganism recognizing device which makes a judgment and outputs an abnormal signal, and outputs an abnormal signal by judging that the threshold value for binarization is low when the predetermined reference curve of the number distribution curve is not reached. 6. The correction threshold proper diagnosis means for checking whether or not the correction threshold diagnosed by the binarization threshold proper diagnosis means is proper according to any one of claims 1 to 5. Microbial recognition device. 7. The correction threshold appropriateness diagnosing means according to claim 6, wherein the number difference of the microorganisms measured at least twice based on the number distribution of the microorganisms from the classifying means is obtained, and the absolute value of the number difference is a predetermined value. When the absolute value of the number difference is smaller than the predetermined reference value, it is determined that the threshold correction is inappropriate and an abnormal signal is output to the threshold correction means when the absolute value of the number difference is smaller than the predetermined reference value. A microorganism recognition device characterized by: