Artificial intelligence early warning method and system for filamentous bacterium expansion

Through artificial intelligence image recognition technology, the filamentous bacteria in the microscopic examination pictures of sewage plant are identified, and their expansion data are calculated. Combined with the sludge settlement performance indicators, accurate and timely warning of the expansion of filamentous bacteria is achieved, and the poor accuracy in the existing technology is solved. and the problem of hysteresis improves the stability of the operation and management of the sewage treatment plant.

CN118918582APending Publication Date: 2024-11-08SDIC XINKAI WATER ENVIRONMENT INVESTMENT CO LTD

Patent Information

Application Number
CN202410931610.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

In the prior art, the early warning method for filamentous bacterial swelling is poor, especially when filamentous bacterial swelling develops rapidly, it is difficult to provide accurate and stable feedback, and the SVI data has a lag and cannot provide timely judgment.

Method used

Use artificial intelligence image recognition tool to identify filamentous bacteria in the microscopic examination pictures taken by sewage plant sludge. The degree of expansion is determined by calculating the filamentous bacteria expansion data, and early warning of the risk of filamentous bacteria expansion, and use the training identification model and gray The degree threshold algorithm automatically segments, determines the filamentous body length and area ratio of the total protruding bacterial fuses, and combines the sludge settlement performance index SVI and SV30 for correlation prediction.

Benefits of technology

It improves the accuracy and timeliness of filamentous bacterial expansion judgment, can provide more timely and accurately judgments on filamentous bacterial expansion, and early warnings, provide more response time for sewage plants to avoid deterioration of sludge settlement.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918582A_ABST
    Figure CN118918582A_ABST
Patent Text Reader

Abstract

The invention provides an artificial intelligence early warning method and system for filamentous bacterium expansion, and relates to the technical field of sewage treatment.The method comprises the steps that S1, a microscopic examination picture is obtained, and filamentous bacteria in the microscopic examination picture are recognized; s2, calculating filamentous bacterium expansion data, and judging the filamentous bacterium expansion degree; and S3, according to the filamentous bacterium expansion degree, early warning the filamentous bacterium expansion risk in advance. According to the invention, the filamentous bacterium expansion can be judged more timely and accurately, and the occurrence of filamentous bacterium expansion can be warned; according to the method, the technical staff of the sewage plant can provide more and more sufficient coping time for coping with sludge bulking in advance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and in particular to an artificial intelligence early warning method and system for filamentous bacterial expansion. Background Art

[0002] With the continuous expansion of urban scale and population growth, water pollution has become a key pollution problem that needs to be solved urgently. During the operation of sewage treatment plants, the state of activated sludge is a key factor affecting the treatment effect. The current sewage discharge standards are increasing day by day, and the monitoring of activated sludge requires more real-time, fast and convenient methods and tools.

[0003] Filamentous bacterial expansion is an undesirable state of activated sludge. Due to the excessive growth of filamentous bacteria in the activated sludge, the settling performance of the activated sludge deteriorates, which eventually leads to the inability of the secondary sedimentation tank in the biochemical system to effectively complete the mud-water separation, destroying the biochemical system. In severe cases, it will cause the biochemical system to collapse and the sewage plant to stop production.

[0004] At the same time, with the birth of new sewage treatment plants such as sinking type, refined operation has become an important direction in the future water treatment field. Among them, in the biochemical treatment section, the monitoring and early warning of filamentous bacteria expansion will be beneficial to the control of the process. In summary, predicting filamentous bacteria expansion in advance will greatly improve the stability of operation management.

[0005] In response to this problem, many scholars at home and abroad have studied the early warning of filamentous fungus expansion and achieved certain improvement results:

[0006] Patent document CN104914213A discloses a method for early warning of activated sludge filamentous bacterial expansion, comprising: 1) continuously monitoring the biomass and the total respiration rate OURCN of activated sludge in a sewage treatment plant for a period of time;

[0007] 2) Obtain the change value of the sludge concentration and the total sludge respiration rate SOURCN value of the sewage treatment plant that changes with time, draw a curve between the change value of the sludge concentration and the total sludge respiration rate OURCN value of each sewage treatment plant, and calculate the sludge total respiration rate SOURCN change curve; 3) Observe whether there is a sudden increase from low to high in the sludge total respiration rate SOURCN change curve. If so, the activated sludge will subsequently undergo filamentous bacterial expansion. This invention can predict whether there is a risk of filamentous bacterial expansion in the activated sludge based on the analysis of the activated sludge SOURCN data. There is no direct correlation between the SOURCN value and the filamentous bacterial expansion. The judgment of this patent is based on the "sudden increase" of the SOURCN value compared with the conventional SOURCN value. The correlation is weak and the accuracy is difficult to grasp.

[0008] Patent document CN112767362A discloses a method for predicting sludge bulking based on phase contrast microscopic images of activated sludge. The specific implementation scheme includes six steps: image acquisition, image fusion, image segmentation, feature extraction, dynamic model, and sludge bulking warning. Image acquisition, collecting phase contrast microscopic images of activated sludge samples; image fusion, using discrete cosine transform to fuse images of activated sludge in different focal planes; image segmentation, using phase consistency method to remove phase difference, and then realizing the segmentation of activated sludge filamentous bacteria and flocs through low-pass filtering, phase stretching transformation, threshold segmentation, opening operation and gyration radius features; feature extraction, extracting morphological characteristic parameters of flocs and filamentous bacteria; combined with SVI value, configuring network model SVI warning model, and outputting SVI value.

[0009] SV30 and SVI are traditional methods for evaluating sludge characteristics and sedimentation performance. SV30 refers to the volume percentage of sludge in the aeration tank mixed liquid after static sedimentation in the measuring cylinder for 30 minutes; the sludge volume index SVI refers to the volume of precipitated sludge formed by each unit mass of dry sludge after the aeration tank mixed liquid has settled for 30 minutes. Both indicators can directly and intuitively reflect the sedimentation of sludge, but in actual engineering control, monitoring SVI and SV30 often has a lag and cannot achieve an early warning effect. The paper "Application of Multi-dimensional Coupling Filamentous Bacteria Monitoring Method in Sludge Bulking" (China Water and Wastewater, 2017, Vol. 33, No. 15. Fan Niansi, Wang Runfang, Qi Rong, et al.) describes the relationship between the characteristics of filamentous bacteria and sludge sedimentation: before the sludge sedimentation has changed significantly, the characteristics of filamentous bacteria have undergone a precursory change, that is, the characteristics of filamentous bacteria themselves can provide a warning effect and change far earlier than the SVI value. Therefore, the monitoring of filamentous fungi characteristics can be carried out ahead of changes in SVI values, providing early warning and predicting changes in sedimentation before sedimentation changes occur.

[0010] The existing related technologies have the following defects: (1) The SOURCN method has a weak correlation with filamentous fungal expansion and poor accuracy, and it is difficult to provide accurate and stable feedback when filamentous fungal expansion develops rapidly; (2) SVI data has a lag and cannot provide timely judgment. Summary of the invention

[0011] In view of the defects in the prior art, the present invention provides an artificial intelligence early warning method and system for filamentous fungus expansion.

[0012] According to the present invention, an artificial intelligence early warning method for filamentous fungus expansion is provided, comprising:

[0013] Step S1: obtaining a microscopic examination image, and identifying filamentous bacteria in the microscopic examination image based on the obtained microscopic examination image;

[0014] Step S2: calculating the filamentous bacteria expansion data and determining the filamentous bacteria expansion degree;

[0015] Step S3: providing early warning of the risk of filamentous fungus expansion according to the degree of filamentous fungus expansion.

[0016] Preferably, the microscopic images include: images of one or more of samples stained by Gram staining, samples stained by Neisser staining, and unstained samples, taken under a preset magnification bright field using a microscope.

[0017] Preferably, the identifying filamentous fungi in the microscopic examination picture based on the acquired microscopic examination picture comprises: identifying filamentous fungi in the microscopic examination picture using an image recognition method;

[0018] Step S1.1: cutting the microscopic image into fixed pixel images and reading them in grayscale;

[0019] Step S1.2: enhancing the grayscale image;

[0020] Step S1.3: Use gray threshold algorithm to perform automatic segmentation to obtain a binary image;

[0021] Step S1.4: Targets with a projection area greater than a preset value and a radius greater than a preset value are determined to be filamentous fungi.

[0022] Preferably, the identifying filamentous fungi in the microscopic examination picture based on the acquired microscopic examination picture comprises: using a trained filamentous fungi identification model to identify the filamentous fungi in the acquired microscopic examination picture;

[0023] Wherein, the trained filamentous fungus recognition model includes:

[0024] Label the filamentous fungi in the microscopic images to form training sets and test sets;

[0025] The training set is used to train the filamentous fungi recognition model, and the test set is used to verify the trained filamentous fungi recognition model until the trained filamentous fungi recognition model meets the preset requirements.

[0026] Preferably, the calculation of the filamentous fungus expansion data includes: calculating one or more of the following indicators: the total length of filaments protruding from the bacterial floc, the area ratio occupied by the filamentous fungi, the ratio of bridging filamentous fungi, and the ratio of filamentous fungi protruding from the bacterial floc.

[0027] Preferably, the early warning of filamentous fungus expansion risk includes:

[0028] When the total length of the extended bacterial floc filaments is greater than 10 7μm / mL, the activated sludge is determined to be in a state of risk of expansion; and / or when the total protruding floc filament length value increases by more than 150% within 5 days, the activated sludge is determined to be in a state of risk of expansion.

[0029] Preferably, the early warning of filamentous fungus expansion risk also includes:

[0030] The degree of filamentous bacteria expansion and sludge settling performance SVI, SV 30 Indicators are correlated in a system with time as the horizontal axis to predict SVI and SV in advance 30 Changes in apparent data.

[0031] According to the present invention, an artificial intelligence early warning system for filamentous fungus expansion is provided, comprising:

[0032] Module M1: obtaining microscopic images, and identifying filamentous bacteria in the microscopic images based on the obtained microscopic images;

[0033] Module M2: Calculate the expansion data of filamentous bacteria and determine the expansion degree of filamentous bacteria;

[0034] Module M3: Provide early warning of filamentous fungus expansion risks based on the degree of filamentous fungus expansion.

[0035] Preferably, the microscopic examination images include: images of one or more of samples stained by Gram staining, samples stained by Neisser staining, and unstained samples, taken under a preset magnification bright field under a microscope;

[0036] The identifying of filamentous fungi in the microscopic examination picture based on the acquired microscopic examination picture includes: identifying the filamentous fungi in the microscopic examination picture using an image recognition method;

[0037] Module M1.1: Cut the microscopic image into fixed pixel images and read them in grayscale;

[0038] Module M1.2: Enhanced grayscale image;

[0039] Module M1.3: Use gray threshold algorithm to perform automatic segmentation to obtain a binary image;

[0040] Module M1.4: Targets with a projection area greater than a preset value and a radius greater than a preset value are identified as filamentous fungi;

[0041] Alternatively, the identifying filamentous fungi in the microscopic examination picture based on the obtained microscopic examination picture comprises: using a trained filamentous fungi identification model to identify the filamentous fungi in the obtained microscopic examination picture;

[0042] Wherein, the trained filamentous fungus recognition model includes:

[0043] Label the filamentous fungi in the microscopic images to form training sets and test sets;

[0044] The training set is used to train the filamentous fungi recognition model, and the test set is used to verify the trained filamentous fungi recognition model until the trained filamentous fungi recognition model meets the preset requirements.

[0045] Preferably, the calculation of the filamentous fungus expansion data includes: calculating one or more of the following indicators: the total length of the filaments protruding from the bacterial floc, the area ratio occupied by the filamentous fungi, the proportion of the bridging filamentous fungi, and the proportion of the filamentous fungi protruding from the bacterial floc;

[0046] The early warning filamentous fungus expansion risks include:

[0047] When the total length of the extended bacterial floc filaments is greater than 10 7 μm / mL, the activated sludge is judged to be in a state of bulking risk; and / or when the total length of the protruding bacterial floc filaments increases by more than 150% within 5 days, the activated sludge is judged to be in a state of bulking risk;

[0048] The early warning filamentous fungus expansion risk also includes:

[0049] The degree of filamentous bacteria expansion and sludge settling performance SVI, SV 30 Indicators are correlated in a system with time as the horizontal axis to predict SVI and SV in advance 30 Changes in apparent data.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. The present invention utilizes artificial intelligence image recognition tools to identify filamentous bacteria in microscopic images of sludge samples taken from sewage treatment plants.

[0052] 2. The present invention accumulates data and scientifically studies key parameter data of filamentous fungus expansion, thereby improving the accuracy and timeliness of the judgment of filamentous fungus expansion and judging filamentous fungus expansion in advance.

[0053] 3. The present invention can provide more timely and accurate judgment of filamentous bacterial expansion and warn of the occurrence of filamentous bacterial expansion; through this method, technical staff of the sewage treatment plant can be provided with more sufficient response time to deal with sludge expansion in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0055] Figure 1 It is the overall flow chart of the present invention;

[0056] Figure 2 For microscopic examination pictures;

[0057] Figure 3 To automatically identify and annotate filamentous fungi on original microscopic images;

[0058] Figure 4 This is a picture of the individually extracted filamentous fungi;

[0059] Figure 5 Reason Figure 4 Picture of filamentous fungi obtained after secondary processing;

[0060] Figure 6 is automatically derived by the software Figure 5 The total extended filament length (TEFL) of filamentous fungi, the area ratio occupied by filamentous fungi, the number of extended filamentous fungi, the number of filamentous fungi bridging filamentous fungi, etc.

[0061] Figure 7 It is a picture showing the values ​​of SVI, TEFL, the proportion of area occupied by filamentous bacteria, the number of filamentous bacteria extending out of flocs, and the number of filamentous bacteria bridging flocs of activated sludge on a timeline. DETAILED DESCRIPTION

[0062] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0063] Example 1

[0064] The embodiment of the present invention provides an artificial intelligence early warning method for filamentous fungus expansion, such as Figure 1 As shown, the method includes:

[0065] Step S1: obtaining a microscopic image and identifying filamentous fungi in the microscopic image;

[0066] Step S2: calculating the filamentous bacteria expansion data and determining the filamentous bacteria expansion degree;

[0067] Step S3: providing early warning of the risk of filamentous fungus expansion according to the degree of filamentous fungus expansion.

[0068] The present invention realizes the recognition of all filamentous bacteria in a microscopic examination picture of microbial activated sludge through an image recognition method.

[0069] The microscopic examination images include: images of one or more of samples stained by Gram staining, samples stained by Neisser staining, and unstained samples, which are taken under a preset magnification bright field under a microscope.

[0070] Among them, filamentous bacteria refer to bacteria species in activated sludge that may cause poor settling of activated sludge due to their long mycelium.

[0071] The degree of filamentous bacteria expansion refers to the extent to which excessive growth of filamentous bacteria in activated sludge will cause internal dispersion of the activated sludge, thereby resulting in poor sedimentation performance.

[0072] Early warning means predicting the possibility of filamentous fungal expansion in the future through data on the degree of filamentous fungal expansion, and issuing a warning when filamentous fungal expansion is likely to occur.

[0073] The automatic recognition function can be achieved through artificial intelligence, and through continuous deep learning by computers, the recognition accuracy can be continuously improved.

[0074] Identifying filamentous fungi in microscopic images includes: using image recognition methods to identify filamentous fungi in microscopic images and marking them, referring to Figure 2 and Figure 3 Specifically:

[0075] Step S1.1: cutting the microscopic image into fixed pixel images and reading them in grayscale;

[0076] Step S1.2: enhancing the grayscale image;

[0077] Step S1.3: Use gray threshold algorithm to perform automatic segmentation to obtain a binary image;

[0078] Step S1.4: Targets with a projection area greater than 200 pixels and a radius greater than 1 are determined to be filamentous fungi.

[0079] Identification of filamentous fungi in microscopic images also includes:

[0080] The trained filamentous fungi recognition model is used to identify the filamentous fungi in the acquired microscopic images;

[0081] Wherein, the trained filamentous fungus recognition model includes:

[0082] Label the filamentous fungi in the microscopic images to form training sets and test sets;

[0083] The training set is used to train the filamentous fungi recognition model, and the test set is used to verify the trained filamentous fungi recognition model until the trained filamentous fungi recognition model meets the preset requirements.

[0084] The degree of expansion of filamentous fungi is determined by one or more of the following indicators: the total length of filaments protruding from the bacterial floc, the proportion of the area occupied by the filamentous fungi, the proportion of bridging filamentous fungi, and the proportion of filamentous fungi protruding from the bacterial floc.

[0085] Specifically, the early warning function is realized based on the characteristics of filamentous bacteria obtained through long-term microscopic examination. After long-term experience accumulation, early prediction can be achieved before changes in sludge settleability occur.

[0086] Specifically, the early warning of filamentous fungus expansion risk adopts method 1 and / or method 2:

[0087] Method 1: When the total length of the extended bacterial floc filaments is greater than 10 7 μm / mL, the activated sludge is judged to be in a state of bulking risk;

[0088] Method 2: When the total length of the protruding bacterial floc filaments increases at a rate greater than 150% within 5 days, the activated sludge is judged to be in a state of bulking risk.

[0089] Methods for early warning of filamentous bacterial expansion can be increased: filamentous bacterial expansion degree and sludge settling performance SVI, SV 30 Indicators are correlated in a system with time as the horizontal axis to predict SVI and SV in advance 30 The change of apparent data can predict the change of sedimentation 3-5 days in advance than the change of SVI value, and a higher judgment accuracy can be achieved after accumulating a certain amount of data.

[0090] The present invention can predict the expansion of filamentous bacteria in advance and make predictions before the sludge settleability changes, so as to prevent irreversible deterioration of the sludge settleability after it deteriorates. The characteristics of filamentous bacteria are no longer limited to the length, width and other features, and the filamentous bacteria are refined and classified into "filamentous bacteria bridging flocs", "filamentous bacteria extending out of flocs" and "filamentous bacteria in free-floating flocs", which increases the accuracy of judgment and early warning.

[0091] Example 2

[0092] Embodiment 2 is a preferred embodiment of Embodiment 1

[0093] 1. Identification of filamentous fungi:

[0094] (1) Obtain microscopic images taken with a microscope, such as Figure 2 As shown, from the magnification of the microscope, it can be seen that the sampling magnification B is 400X, and the scale is 1:100μm;

[0095] (2) Open the image using Photoshop and extract filamentous fungi according to the following steps:

[0096] ① Copy the layer; ② Copy the layer in reverse; ③ Adjust the brightness to 150 and the contrast to -50; ④ Select - Color Range - Shadows, select 10% for color tolerance and 57 for range, confirm and press the delete key; ⑤ Select - Color Range - Highlights, select 75% for color tolerance and 200 for range, confirm and press the delete key;

[0097] At this time, the copied layer is judged as "filamentous fungi", as shown in the attached Figure 3-4 shown.

[0098] 2. Measuring the length of filamentous fungi:

[0099] Figure 4 All image sections identified as filamentous fungi are shown. Use ImageJ software to open the image and follow the steps below to optimize the image for length determination:

[0100] ① In Process→Enhance Contrast, select 1% to enhance the contrast; ② In Image→Type→HSBStack, select Image→Stacks→Stack to Images, open the image in H, S, and B modes, and use the Saturation map; ③ In Process→Binary→Make Binary, realize the binary system of the image. At this time, the image is as shown in the attached picture. Figure 5 shown.

[0101] Click Analyze→Set Scale, enter the scale, and then click Analyze→Measure to measure the total length of all filamentous fungi, which is 2048μm.

[0102] 3. Measuring the area proportion of filamentous fungi

[0103] like Figure 6 As shown, the diameter of the filamentous fungi is X μm, and the product of the diameter and the length of the filamentous fungi is X μm to obtain the area of ​​the filamentous fungi, which is then divided by the area of ​​the entire picture to obtain the proportion of the area occupied by the filamentous fungi.

[0104] 4. Measure the proportion of bridging filamentous fungi, the total length of filaments protruding from the bacterial flocs, and the proportion of filamentous fungi protruding from the bacterial flocs

[0105] Define "bridging filamentous fungi" as "filamentous fungi with both ends of the filaments bridged between flocs". At the same time, define "filamentous fungi extending out of flocs" as "filamentous fungi with only one end in the floc". After determination, calculate the number and length respectively, and based on this, the total length of filaments extending out of flocs and the length of bridging filamentous fungi can be obtained, and then divided by the total length of filamentous fungi to obtain the proportion of bridging filamentous fungi and the proportion of filamentous fungi extending out of flocs.

[0106] 5. Monitor the sludge bulking degree of activated sludge, the length of filamentous bacteria, the area ratio of filamentous bacteria, the ratio of bridging filamentous bacteria, and the ratio of filamentous bacteria protruding from flocs;

[0107] During the operation of the reactor, the SVI value of the activated sludge was monitored every 2 days;

[0108] During this period, the data of the length of filamentous bacteria, the area ratio of filamentous bacteria, the ratio of bridging filamentous bacteria, and the ratio of filamentous bacteria protruding from flocs of activated sludge were obtained. The frequency can be set to 100 times / day, and the average value within 2 days was taken.

[0109] The monitoring is conducted for 3-6 months, and the monitoring can be conducted within the time range when the sludge settling properties change greatly. The activated sludge should be kept in an expanded state for a period of time to ensure the data volume.

[0110] 6. Draw a point-line graph distributed over time

[0111] Draw a relationship diagram with TEFL as the horizontal axis and SVI as the vertical axis, such as Figure 7 shown.

[0112] With time as the horizontal axis, the average value of the length of filamentous fungi, the proportion of area occupied by filamentous fungi, the proportion of bridging filamentous fungi, the proportion of filamentous fungi protruding from the floc, and the SVI value as the vertical axis, a dot-line graph is drawn, such as Figure 7 shown.

[0113] Observe the relationship between TEFL and SVI values, and the relationship between filamentous fungi-related characteristics and SVI value changes in the point-line graph, and set up early warning strategies for filamentous fungi-related characteristics.

[0114] Example 3

[0115] Embodiment 3 is a preferred embodiment of Embodiment 1

[0116] (1) Use the EasyDL image tool in Baidu Brain AI open platform and the object segmentation tool;

[0117] (2) After importing the microscopic images, manually annotate the filamentous bacteria on the images;

[0118] (3) After accumulating the annotations of filamentous fungi in multiple images (e.g., 30 images), artificial intelligence learning can be performed within the platform, and the model can achieve a certain recognition effect;

[0119] (4) Continue to accumulate annotations of filamentous fungi (e.g., 100-300 images), and the accuracy of determination can be improved. At this time, the model has completed training and can realize the automatic recognition function.

[0120] The embodiments of the present invention provide an artificial intelligence early warning method and system for filamentous bacterial expansion, which utilizes artificial intelligence image recognition tools to identify filamentous bacteria in microscopic images of sludge samples taken from sewage treatment plants; through data accumulation, key parameter data of filamentous bacterial expansion are scientifically analyzed to judge filamentous bacterial expansion in advance.

[0121] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0122] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. An artificial intelligence early warning method for filamentous fungus expansion, characterized in that: include: Step S1: obtaining a microscopic examination image, and identifying filamentous bacteria in the microscopic examination image based on the obtained microscopic examination image; Step S2: calculating the filamentous bacteria expansion data and determining the filamentous bacteria expansion degree; Step S3: providing early warning of the risk of filamentous fungus expansion according to the degree of filamentous fungus expansion.

2. The artificial intelligence early warning method for filamentous fungus expansion according to claim 1, characterized in that: The microscopic examination images include: images of one or more of samples stained by Gram staining, samples stained by Neisser staining, and unstained samples, which are taken under a preset magnification bright field under a microscope.

3. The artificial intelligence early warning method for filamentous fungus expansion according to claim 1, characterized in that: The identifying of filamentous fungi in the microscopic examination picture based on the acquired microscopic examination picture includes: identifying the filamentous fungi in the microscopic examination picture using an image recognition method; Step S1.1: cutting the microscopic image into fixed pixel images and reading them in grayscale; Step S1.2: enhancing the grayscale image; Step S1.3: Use gray threshold algorithm to perform automatic segmentation to obtain a binary image; Step S1.4: Targets with a projection area greater than a preset value and a radius greater than a preset value are determined to be filamentous fungi.

4. The artificial intelligence early warning method for filamentous fungus expansion according to claim 1, characterized in that: The identifying of filamentous fungi in the microscopic examination picture based on the obtained microscopic examination picture includes: using a trained filamentous fungi identification model to identify the filamentous fungi in the obtained microscopic examination picture; Wherein, the trained filamentous fungus recognition model includes: Label the filamentous fungi in the microscopic images to form training sets and test sets; The training set is used to train the filamentous fungi recognition model, and the test set is used to verify the trained filamentous fungi recognition model until the trained filamentous fungi recognition model meets the preset requirements.

5. The artificial intelligence early warning method for filamentous fungus expansion according to claim 1, characterized in that: The calculation of the filamentous fungus expansion data includes: calculating one or more of the following indicators: the total length of the filaments protruding from the fungus group, the area ratio occupied by the filamentous fungi, the ratio of the bridging filamentous fungi, and the ratio of the filamentous fungi protruding from the fungus group.

6. The artificial intelligence early warning method for filamentous fungus expansion according to claim 5, characterized in that: The early warning filamentous fungus expansion risks include: When the total length of the extended bacterial floc filaments is greater than 10 7 μm / mL, the activated sludge is determined to be in a state of risk of expansion; and / or when the total protruding floc filament length value increases by more than 150% within 5 days, the activated sludge is determined to be in a state of risk of expansion.

7. The artificial intelligence early warning method for filamentous fungus expansion according to claim 1, characterized in that: The early warning filamentous fungus expansion risk also includes: The degree of filamentous bacteria expansion and sludge settling performance SVI, SV 30 Indicators are correlated in a system with time as the horizontal axis to predict SVI and SV in advance 30 Changes in apparent data.

8. An artificial intelligence early warning system for filamentous fungus expansion, characterized in that: include: Module M1: obtaining microscopic images, and identifying filamentous bacteria in the microscopic images based on the obtained microscopic images; Module M2: Calculate the expansion data of filamentous bacteria and determine the expansion degree of filamentous bacteria; Module M3: Provide early warning of filamentous fungus expansion risks based on the degree of filamentous fungus expansion.

9. The filamentous fungus expansion artificial intelligence early warning system according to claim 8, characterized in that: The microscopic examination images include: images of one or more of samples stained by Gram staining, samples stained by Neisser staining, and unstained samples, taken under a preset magnification bright field using a microscope; The identifying of filamentous fungi in the microscopic examination picture based on the acquired microscopic examination picture includes: identifying the filamentous fungi in the microscopic examination picture using an image recognition method; Module M1.1: Cut the microscopic image into fixed pixel images and read them in grayscale; Module M1.2: Enhanced grayscale image; Module M1.3: Use gray threshold algorithm to perform automatic segmentation to obtain a binary image; Module M1.4: Targets with a projection area greater than a preset value and a radius greater than a preset value are identified as filamentous fungi; Alternatively, the identifying filamentous fungi in the microscopic examination picture based on the obtained microscopic examination picture comprises: using a trained filamentous fungi identification model to identify the filamentous fungi in the obtained microscopic examination picture; Wherein, the trained filamentous fungus recognition model includes: Label the filamentous fungi in the microscopic images to form training sets and test sets; The training set is used to train the filamentous fungi recognition model, and the test set is used to verify the trained filamentous fungi recognition model until the trained filamentous fungi recognition model meets the preset requirements.

10. The artificial intelligence early warning system for filamentous fungus expansion according to claim 8, characterized in that: The calculation of the filamentous fungus expansion data includes: calculating one or more of the following indicators: the total length of the filaments protruding from the fungus group, the area ratio occupied by the filamentous fungi, the ratio of the bridging filamentous fungi, and the ratio of the filamentous fungi protruding from the fungus group; The early warning filamentous fungus expansion risks include: When the total length of the extended bacterial floc filaments is greater than 10 7 μm / mL, the activated sludge is judged to be in a state of bulking risk; and / or when the total length of the protruding bacterial floc filaments increases by more than 150% within 5 days, the activated sludge is judged to be in a state of bulking risk; The early warning filamentous fungus expansion risk also includes: The degree of filamentous bacteria expansion and sludge settling performance SVI, SV 30 Indicators are correlated in a system with time as the horizontal axis to predict SVI and SV in advance 30 Changes in apparent data.

Citation Information

Patent Citations

  • Method for early warning of activated sludge filamentous bacterium expansion

    CN104914213A

  • Sludge bulking prediction method based on activated sludge phase difference microscopic image

    CN112767362A

Cited By

  • Method and system for early warning and controlling sludge bulking of sewage treatment system in coal chemical industry

    CN119930060A