Method for detecting concentration of bacteria in water based on image processing technology

Abstract

The invention discloses a method for detecting the concentration of bacteria in water based on an image processing technology, and relates to the technical field of computer image processing. The method for detecting the concentration of bacteria in water based on the image processing technology comprises the following steps: acquiring preliminary detection accuracy; acquiring an area scanning evaluation value; generating a superpixel; and acquiring the concentration of the to-be-detected bacteria. According to the method, the to-be-detected sample data is preliminarily marked through the fluorescent marker to obtain the preliminary detection accuracy, meanwhile, the to-be-detected region division process in the laser layer-by-layer scanning operation process is monitored in real time to obtain the region scanning evaluation value, and then the obtained three-dimensional image is subjected to superpixel segmentation to generate superpixels; finally, the number of the to-be-detected bacteria in the superpixels is counted, and the concentration of the to-be-detected bacteria is obtained in combination with the preset experimental group water sample volume, so that the effect of improving the reliability of detecting the concentration of the bacteria in water based on data processing is achieved, and the problem that the reliability of detecting the concentration of the bacteria in water based on data processing is low in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer image processing, and in particular to a method for detecting bacteria concentration in water based on image processing technology. Background Art

[0002] With the rapid development of industrialization and urbanization, water pollution is becoming increasingly serious, and water quality monitoring has become an important part of ensuring public health and sustainable use of water resources. Although traditional water quality monitoring methods, such as microscopic observation and biochemical analysis, can provide relatively accurate water quality data, they often have problems such as cumbersome operation, long time consumption, and high cost, and are difficult to meet the needs of large-scale, real-time, and efficient water quality monitoring. In recent years, with the rapid development of computer technology and image processing technology, water quality monitoring methods based on image processing have gradually emerged. This method uses high-resolution cameras or microscopes and other equipment to collect water sample images, and uses image processing algorithms to preprocess, extract features, and classify and identify images, thereby achieving rapid and accurate detection of parameters such as bacterial concentration in water.

[0003] The existing technology collects water samples and obtains images of bacteria through image acquisition equipment. At the same time, it uses image processing technology to extract the contour information of bacteria and calculates the characteristic parameters of bacteria based on the extracted contour information. Then, the counting result is obtained by counting the number of bacteria in the image. Finally, the calculated bacterial concentration is compared with the standard method to verify the accuracy of the result, thereby improving the efficiency of bacterial detection.

[0004] For example, the invention patent with announcement number: CN108830149B announces a method and terminal device for detecting target bacteria, including: obtaining an original image of a bacterial slice to be detected; identifying the effective activity area of ​​the target bacteria in the original image, and obtaining a preset number of clear grid images based on the effective activity area; inputting the preset number of clear grid images into a neural network model to obtain an image recognition result; determining the position information of the target bacteria in the original image based on the image recognition result; selecting the effective grid image with the highest clarity among the effective grid images of each layer under the same coordinate according to a clarity evaluation algorithm as the clear grid image of the same coordinate; selecting the clear grid images of each coordinate in turn to obtain a preset number of clear grid images.

[0005] For example, the invention patent with announcement number CN107644210B discloses a method for estimating the number of microorganisms based on image processing, including: reading the original image and graying it to obtain a grayscale image I0; setting the grayscale of the edge contour part to 0, and setting the processed image to I1; denoising I1 using a non-local mean filtering algorithm, and setting the denoised image to I2; binarizing the image I3 after removing the background, and setting the binarized image to I4; selecting 10 representative bacterial image samples A1 to A10, counting the total number of pixels occupied by all bacteria in the entire image, dividing by the average number of pixels per bacterium, and finally obtaining the total number of bacteria in the original image.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, bacteria in water are of various types and shapes, so there may be deviations in the counting results in the process of counting the number of bacteria in the image. Secondly, due to the influence of other substances in the water sample (such as suspended matter, organic matter) and environmental factors, the quality of the bacterial image and the recognition process are interfered, thereby reducing the reliability of the detection. There is a problem of low reliability in the detection of bacterial concentration in water based on data processing. Summary of the invention

[0008] The embodiment of the present application solves the problem of low reliability of bacteria concentration detection in water based on data processing in the prior art by providing a method for detecting bacteria concentration in water based on image processing technology, and improves the reliability of bacteria concentration detection in water based on data processing.

[0009] The embodiment of the present application provides a method for detecting the concentration of bacteria in water based on image processing technology, comprising the following steps: S1, blank marking and fluorescent marking are performed on the acquired control sample data and the sample data to be tested, respectively, and the marking effect of the fluorescent marker on the bacteria to be tested is preliminarily evaluated during the marking process to obtain a preliminary detection accuracy, and the preliminary detection accuracy is used to quantify the marking effect of the fluorescent marker on the bacteria to be tested and the preliminary reliability of the detection; S2, based on the obtained preliminary detection accuracy, it is determined whether to perform a laser layer-by-layer scanning operation, and if so, the process of dividing the area to be tested during the laser layer-by-layer scanning operation is monitored in real time to obtain a regional scanning evaluation value, and at the same time, based on the obtained regional scanning evaluation value, it is determined whether to generate a two-dimensional image of the bacteria to be tested, and the regional scanning evaluation value is used to evaluate the accuracy of the laser layer-by-layer scanning operation in dividing the area to be tested within a preset time period; S3, accumulating the generated two-dimensional images to obtain a three-dimensional image and preprocessing it, and at the same time, superpixel segmentation is performed on the preprocessed three-dimensional image according to the superpixel algorithm to generate superpixels; S4, counting the number of bacteria to be tested after identification and segmentation by the superpixel within a preset time period and combining it with the preset experimental group water sample volume to obtain the concentration of the bacteria to be tested.

[0010] Furthermore, the preliminary detection accuracy is obtained by the following method: obtaining fluorescence data of the bacteria to be tested in the water samples of the experimental group within a preset time period, and simultaneously obtaining reference fluorescence data of the corresponding bacteria to be tested in the water samples of the control group within the same time period, the fluorescence data including fluorescence colorimetry, fluorescence penetration rate and fluorescence labeling efficiency, and the reference fluorescence data including reference fluorescence colorimetry, reference fluorescence penetration rate and reference fluorescence labeling efficiency; processing the obtained fluorescence data and reference fluorescence data respectively to obtain fluorescence colorimetry scores, fluorescence penetration rate scores and fluorescence labeling efficiency scores, thereby obtaining the preliminary detection accuracy. degree; the fluorescence colorimetric score is the ratio of the fluorescence colorimetric to the reference fluorescence colorimetric; the fluorescence permeability score is the ratio of the fluorescence permeability to the reference fluorescence permeability; the fluorescence labeling rate score is the ratio of the fluorescence labeling rate to the reference fluorescence labeling rate; the fluorescence colorimetric is the ratio of the difference between the average fluorescence intensity of the fluorescence area where the bacteria to be tested are located and the reference average fluorescence intensity of the background area to the reference average fluorescence intensity of the background area; the fluorescence permeability is the ratio of the number of labeled bacteria to be tested to the total number of bacteria to be tested; the fluorescence labeling efficiency is the ratio of the number of labeled bacteria to be tested to the reference labeled bacteria to be tested.

[0011] Furthermore, the specific limiting expression of the preliminary detection accuracy is:

[0012]

[0013]

[0014] Wherein, m is the number of the bacterial species to be tested, m=1, 2, 3, m=1 represents Escherichia coli, m=2 represents Staphylococcus aureus, m=3 represents Pseudomonas aeruginosa, t is the number of the preset time period, t=1, 2, ..., T, T is the total number of preset time periods, e is a natural constant, It represents the initial detection accuracy of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the fluorescence colorimetric score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the reference fluorescence colorimetric value of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the average fluorescence colorimetric value of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period. It represents the reference average fluorescence colorimetric value of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the fluorescence penetration rate score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, represents the reference fluorescence penetration rate of the mth test bacteria in the control water sample during the blank labeling process, It represents the number of bacteria labeled in the experimental group water sample during the fluorescence labeling process of the mth bacteria in the water sample in the tth preset time period. It represents the total number of the mth bacteria to be tested in the control group water sample during the blank labeling process. It represents the fluorescence labeling efficiency score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the reference fluorescence labeling rate of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the reference labeled bacteria of the mth tested bacteria in the control group water sample during the blank labeling process.

[0015] Furthermore, the regional scanning evaluation value is obtained by the following method: E1, real-time monitoring of the interruption of the laser wavelength of the laser beam in the scanning channel, judging whether the receiving end of the scanning channel receives the laser beam, if so, it indicates that the corresponding bacteria to be tested in the scanning channel are living cells and E2 is executed, otherwise removal is performed; E2, obtaining the average scanning speed, average laser power and density of the bacteria to be tested of the laser beam in a preset time period, judging whether the density of the bacteria to be tested is within the density threshold range, if so, executing E3, otherwise prompting the preset personnel to dilute the bacteria to be tested; E3, judging whether the average scanning speed is within the maximum allowable scanning speed range , and at the same time determine whether the average laser power is within the maximum allowable laser power range. When the average scanning speed is within the maximum allowable scanning speed range and the average laser power is within the maximum allowable laser power range, execute E4, otherwise no calculation is performed; E4, obtain the scanning speed score, laser power score and fluorescent signal intensity score and combine them with the obtained preliminary detection accuracy to obtain the regional scanning evaluation value, the scanning speed score is the ratio of the average scanning speed to the reference average scanning speed, the laser power score is the ratio of the average laser power to the reference average laser power, and the fluorescent signal intensity score is the ratio of the fluorescent signal intensity to the maximum allowable fluorescent signal intensity.

[0016] Furthermore, the specific steps of performing superpixel segmentation on the preprocessed three-dimensional image according to the superpixel algorithm to generate superpixels include: acquiring image information data in the three-dimensional image and segmenting the three-dimensional image into superpixel regions according to the superpixel algorithm, and identifying the target region in the superpixel region according to similar pixel features in the three-dimensional image; the image information data includes color data, brightness data and spatial domain information data; the superpixel region is used to reflect the structural information of the bacteria to be tested in the three-dimensional image; the similar pixel features include similar color features, similar brightness features and similar texture features.

[0017] Furthermore, the method of obtaining the concentration of the bacteria to be tested also includes obtaining an environmental impact score according to the degree of influence of the detection environmental factors on the concentration of the bacteria to be tested; the detection environmental factors include dissolved oxygen content, light intensity and temperature of the water sample of the experimental group; the specific steps for obtaining the environmental impact score are: L1, real-time monitoring of changes in the detection environmental factors in the fluorescent area where the bacteria to be tested are located during the detection process to obtain the average dissolved oxygen content, the average light intensity and the average temperature of the water sample of the experimental group, and judging whether the average dissolved oxygen content is within the allowable range of the dissolved oxygen content. If so, execute L2, otherwise prompt the preset personnel to adjust the dissolved oxygen conditions of the fluorescent area where the bacteria to be tested are located; L2, judging whether the average light intensity is within the allowable range of the light intensity. If so, execute L3, otherwise prompt the preset personnel to adjust the dissolved oxygen conditions of the fluorescent area where the bacteria to be tested are located; A staff member is provided to adjust the light source conditions of the fluorescent area corresponding to the bacteria to be tested; L3, judge whether the average temperature of the water samples in the experimental group is within the allowable range of the water sample temperature of the experimental group. If so, execute L4, otherwise prompt the preset staff member to adjust the water temperature conditions of the fluorescent area corresponding to the bacteria to be tested; L4, obtain the dissolved oxygen content score, the light intensity score and the temperature score of the water samples in the experimental group and combine the obtained regional scanning evaluation value to obtain the environmental impact score; the dissolved oxygen content score is the ratio of the average dissolved oxygen content in the preset time period to the maximum allowable dissolved oxygen content; the light intensity score is the ratio of the average light intensity in the preset time period to the maximum allowable light intensity; the water sample temperature score of the experimental group is the ratio of the average water sample temperature of the experimental group in the preset time period to the maximum allowable water sample temperature of the experimental group.

[0018] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0019] 1. The test sample data is preliminarily marked with fluorescent markers to obtain preliminary detection accuracy. At the same time, the test area division process during the laser layer-by-layer scanning operation is monitored in real time to obtain the regional scanning evaluation value. Then, the acquired three-dimensional image is superpixel segmented according to the superpixel algorithm to generate superpixels. Finally, the number of bacteria to be tested in the superpixel is counted and the concentration of bacteria to be tested is obtained in combination with the preset experimental group water sample volume, thereby achieving improved superpixel generation accuracy and reliability, and further achieving improved reliability of water bacteria concentration detection based on data processing, effectively solving the problem of low reliability of water bacteria concentration detection based on data processing in the prior art.

[0020] 2. By obtaining the fluorescence data of the bacteria to be tested in the water samples of the experimental group within a preset time period, and simultaneously obtaining the reference fluorescence data of the corresponding bacteria to be tested in the water samples of the control group within the same time period, the obtained fluorescence data and the reference fluorescence data are processed separately to obtain the fluorescence colorimetric score, the fluorescence penetration rate score and the fluorescence labeling efficiency score, and the preliminary detection accuracy is obtained based on this, thereby achieving an improvement in the accuracy of obtaining the preliminary detection accuracy, and then achieving an improvement in the labeling effect of the bacteria to be tested and the detection accuracy and reliability.

[0021] 3. By real-time monitoring of the interruption of the laser wavelength of the laser beam in the scanning channel, when the receiving end of the scanning channel receives the laser beam and the density of the bacteria to be measured is within the density threshold range, the average scanning speed is within the maximum allowable scanning speed range, and the average laser power is within the maximum allowable laser power range, the scanning speed score, laser power score and fluorescence signal intensity score are obtained and combined with the obtained preliminary detection accuracy to obtain the regional scanning evaluation value, thereby achieving an improvement in the accuracy of obtaining the regional scanning evaluation value, and then achieving a more accurate evaluation of the accuracy of the division of the area to be measured. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A flowchart of a method for detecting bacterial concentration in water based on image processing technology provided in an embodiment of the present application;

[0023] Figure 2 A flow chart for detecting the concentration of bacteria to be tested provided in the embodiments of the present application;

[0024] Figure 3 A two-dimensional coordinate diagram of the preliminary detection accuracy-first bacteria coefficient to be detected provided in the embodiment of the present application;

[0025] Figure 4 A two-dimensional coordinate diagram of the preliminary detection accuracy-second bacteria coefficient provided in the embodiment of the present application. DETAILED DESCRIPTION

[0026] The embodiment of the present application solves the problem of low reliability of water bacteria concentration detection based on data processing in the prior art by providing a method for detecting water bacteria concentration based on image processing technology. The obtained control sample data and test sample data are blank-marked and fluorescently marked respectively, and the marking effect of the fluorescent marker on the test bacteria is preliminarily evaluated during the marking process to obtain a preliminary detection accuracy. Then, based on the obtained preliminary detection accuracy, it is determined whether to perform a laser layer-by-layer scanning operation. If so, the test area division process during the laser layer-by-layer scanning operation is monitored in real time to obtain a regional scanning evaluation value. At the same time, based on the obtained regional scanning evaluation value, it is determined whether to generate a two-dimensional image of the test bacteria. The generated two-dimensional image is then accumulated to obtain a three-dimensional image and preprocessed. The preprocessed three-dimensional image is superpixel segmented according to a superpixel algorithm to generate superpixels. Finally, the number of test bacteria identified and segmented by superpixels within a preset time period is counted and combined with the preset experimental group water sample volume to obtain the test bacteria concentration, thereby improving the reliability of water bacteria concentration detection based on data processing.

[0027] The technical solution in the embodiment of the present application is to solve the problem of low reliability of water bacteria concentration detection based on data processing. The overall idea is as follows:

[0028] The test sample data is preliminarily marked with fluorescent markers to obtain preliminary detection accuracy. At the same time, the test area division process during the laser layer-by-layer scanning operation is monitored in real time to obtain the regional scanning evaluation value. Then, the acquired three-dimensional image is superpixel segmented according to the superpixel algorithm to generate superpixels. Finally, the number of bacteria to be tested in the superpixel is counted and combined with the preset experimental group water sample volume to obtain the test bacteria concentration, thereby achieving the effect of improving the reliability of bacterial concentration detection in water based on data processing.

[0029] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0030] like Figure 1As shown, it is a flow chart of a method for detecting bacterial concentration in water based on image processing technology provided by an embodiment of the present application, the method comprising the following steps: S1, obtaining preliminary detection accuracy: blank marking and fluorescent marking are respectively performed on the obtained control sample data and the sample data to be tested, and the marking effect of the fluorescent marker on the bacteria to be tested is preliminarily evaluated during the marking process to obtain preliminary detection accuracy, the blank mark represents the known bacteria species to be tested and the reference fluorescent data, and the fluorescent markers include DAPI and SYBR Green, the control sample data represents the control group water sample data that has not been treated with fluorescent markers, and the test sample data represents the experimental group water sample data that has been treated with fluorescent markers. The preliminary detection accuracy is used to quantify the labeling effect of fluorescent markers on the test bacteria and the preliminary reliability of the detection; S2, obtain the regional scanning evaluation value: based on the obtained preliminary detection accuracy, determine whether to perform the laser layer-by-layer scanning operation. If so, monitor the test area division process during the laser layer-by-layer scanning operation in real time to obtain the regional scanning evaluation value, and at the same time, determine whether to generate a two-dimensional image of the test bacteria based on the obtained regional scanning evaluation value. The laser layer-by-layer scanning operation means performing a color scanning operation on the fluorescent markers in the test bacteria according to the preset laser wavelength. The color scanning operation is used to visualize the three-dimensional structure of the test bacteria in the experimental group water sample. The regional scanning evaluation The value is used to evaluate the accuracy of the laser layer-by-layer scanning operation in dividing the test area within a preset time period, and the two-dimensional image is used to reflect the distribution and density of the test bacteria on the color scanning plane; S3, generate superpixels: accumulate the generated two-dimensional images to obtain a three-dimensional image and perform preprocessing, and at the same time, perform superpixel segmentation on the preprocessed three-dimensional image according to the superpixel algorithm to generate superpixels. The three-dimensional image is used to reflect the distribution of the test bacteria in the color scanning space. The preprocessing includes depth correction and volume rendering optimization. The superpixel represents a set of similar pixels of the test bacteria in the test area; S4, obtain the concentration of the test bacteria: count the number of test bacteria identified and segmented by the superpixel within the preset time period and combine the preset experimental group water sample volume to obtain the test bacteria concentration. The test bacteria concentration is the ratio of the number of test bacteria to the experimental group water sample volume.

[0031] In this embodiment, the blank label here refers to the known bacterial species to be tested and the corresponding reference fluorescence data (including reference fluorescence colorimetry, reference fluorescence penetration and reference fluorescence labeling efficiency), which are usually used to establish a baseline or standard for comparison with subsequent fluorescently labeled fluorescence data (including fluorescence colorimetry, fluorescence penetration and fluorescence labeling efficiency); DAPI (4',6-diamidino-2-phenylindole) mainly binds to adenine and thymine base pairs in DNA and emits strong blue fluorescence, and SYBR Green is a double-stranded DNA binding dye that emits green fluorescence after binding to DNA.

[0032] Among them, the reference fluorescence colorimetry is represented by the sum and average of the historical fluorescence colorimetry of the bacteria to be tested in the historical time period in the preset database; the reference fluorescence permeability is represented by the sum and average of the historical fluorescence permeability of the bacteria to be tested in the historical time period in the preset database, and the reference fluorescence labeling rate is represented by the sum and average of the historical fluorescence labeling rates of the bacteria to be tested in the historical time period in the preset database.

[0033] Specifically, image processing software (such as ImageJ) is used to accumulate each pixel in a two-dimensional image (obtained through a microscope) in a set depth direction (set by a preset person) to generate a three-dimensional image (obtained through a laser scanner). Since there may be depth errors or inconsistencies in the accumulation process of the two-dimensional image, depth correction and volume rendering optimization (a technology for converting three-dimensional data into two-dimensional images) are required: (1) The depth value of each pixel in the three-dimensional image is adjusted through image processing software (such as OpenCV) (such as increasing the depth value to ensure that the entire three-dimensional image is in the depth direction); (2) When the three-dimensional image is not in the set depth direction, the transparency parameter is increased through the image processing software (such as OpenCV) to improve the transparency of the three-dimensional image. In bacterial detection, volume rendering optimization can help scientists see the distribution and morphology of bacteria in three-dimensional space more clearly.

[0034] The superpixel algorithm is an image segmentation method that aims to divide an image into continuous regions with similar textures, colors, and edges, called superpixels (a collection of pixels with similar features). It usually includes the SLIC (Simple Linear Iterative Clustering) algorithm, the SEEDS (Superpixels Extracted via Energy-Driven Sampling) algorithm, the LSC (Linear Spectral Clustering) algorithm, and the ERS (Entropy Rate Superpixel) algorithm. These algorithms are usually also implemented in professional image processing software or programming environments, such as MATLAB, Python (especially using its image processing libraries such as scikit-image or OpenCV), which helps to more clearly reflect the distribution of the bacteria to be tested in the color scanning space, and then more accurately obtain the number of bacteria to be tested within a preset time period, thereby achieving an improvement in the reliability of water bacteria concentration detection based on data processing.

[0035] Furthermore, the preliminary detection accuracy is obtained by the following method: obtaining the fluorescence data of the bacteria to be tested in the water sample of the experimental group within a preset time period, and obtaining the reference fluorescence data of the corresponding bacteria to be tested in the water sample of the control group within the same time period, the fluorescence data including fluorescence colorimetric degree, fluorescence penetration rate and fluorescence labeling efficiency, and the reference fluorescence data including reference fluorescence colorimetric degree, reference fluorescence penetration rate and reference fluorescence labeling efficiency; processing the obtained fluorescence data and the reference fluorescence data respectively to obtain the fluorescence colorimetric degree score (i.e., the value in the limiting expression of the preliminary detection accuracy); ), the fluorescence penetration fraction (i.e., the limiting expression of the initial detection accuracy ) and the fluorescence labeling efficiency score (i.e., the limiting expression of the initial detection accuracy ), based on which the preliminary detection accuracy was obtained.

[0036] Among them, the fluorescence colorimetric score is the ratio of the fluorescence colorimetric to the reference fluorescence colorimetric score; the fluorescence permeability score is the ratio of the fluorescence permeability to the reference fluorescence permeability; the fluorescence labeling rate score is the ratio of the fluorescence labeling rate to the reference fluorescence labeling rate; the fluorescence colorimetric is the ratio of the difference between the average fluorescence intensity of the fluorescence area where the bacteria to be tested are located and the reference average fluorescence intensity of the background area to the reference average fluorescence intensity of the background area; the fluorescence permeability is the ratio of the number of labeled bacteria to be tested to the total number of bacteria to be tested; the fluorescence labeling efficiency is the ratio of the number of labeled bacteria to be tested to the reference labeled bacteria to be tested.

[0037] It should be added that the specific limiting expression of the preliminary detection accuracy is:

[0038]

[0039] Wherein, m is the number of the bacterial species to be tested, m=1, 2, 3, m=1 represents Escherichia coli, m=2 represents Staphylococcus aureus, m=3 represents Pseudomonas aeruginosa, t is the number of the preset time period, t=1, 2, ..., T, T is the total number of preset time periods, e is a natural constant, It represents the initial detection accuracy of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the fluorescence colorimetric score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the reference fluorescence colorimetric value of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the average fluorescence colorimetric value of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period. It represents the reference average fluorescence colorimetric value of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the fluorescence penetration rate score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, represents the reference fluorescence penetration rate of the mth test bacteria in the control water sample during the blank labeling process, It represents the number of bacteria labeled in the experimental group water sample during the fluorescence labeling process of the mth bacteria in the water sample in the tth preset time period. It represents the total number of the mth bacteria to be tested in the control group water sample during the blank labeling process. It represents the fluorescence labeling efficiency score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the reference fluorescence labeling rate of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the reference labeled bacteria of the mth tested bacteria in the control group water sample during the blank labeling process.

[0040] In this embodiment, the average fluorescence intensity is the average value of the fluorescence intensity (obtained by a fluorescence spectrophotometer) within a preset time period, and the reference average fluorescence intensity is represented by the sum and average of the historical average fluorescence intensities of the bacteria to be tested in the historical time period in a preset database; the reference marked bacteria to be tested is the optimal number value marked by the fluorescent marker within the preset time period (set by the preset personnel), and the total number of bacteria to be tested is the total number of the bacteria to be tested before being marked with the fluorescent marker.

[0041] It should be noted that the fluorescent area is marked with red fluorescence and the background area is marked with yellow fluorescence. This example can more comprehensively evaluate the fluorescence characteristics of the bacteria to be tested by simultaneously considering the three dimensions of fluorescence colorimetry, fluorescence penetration and fluorescence labeling efficiency, and obtain corresponding reference values ​​through corresponding data of historical time periods, thereby improving the accuracy and reliability of the data. The introduction of reference values ​​makes data between different times and batches comparable, and achieves more accurate acquisition of preliminary detection accuracy.

[0042] It should be understood that the fluorescent labeling efficiency corresponding to different bacteria to be tested will also be different, and the permeability and permeability strength of the fluorescent markers are also different. Therefore, the type of bacteria to be tested needs to be considered in the calculation of the preliminary detection accuracy. For example, Escherichia coli (scientific name: Escherichia coli), also known as Escherichia coli, is a species of the genus Escherichia in the family Enterobacteriaceae and has a double membrane structure; Staphylococcus aureus (S.aureus), also known as "golden Staphylococcus", belongs to the genus Staphylococcus and is a representative of Gram-positive bacteria with a thicker cell wall and rich cytoplasm; Pseudomonas aeruginosa, whose English name and scientific name are both Pseudomonas aeruginosa, is an important Gram-negative bacillus with flagella and pili structures.

[0043] Specifically, to simplify the analysis, define In the formula, It represents the coefficient of the first bacteria to be tested in the process of fluorescence labeling of the mth bacteria to be tested in the water sample of the experimental group within the tth preset time period, It represents the coefficient of the second bacteria to be tested in the process of fluorescence labeling of the mth bacteria to be tested in the water sample of the experimental group within the tth preset time period. The specific limiting expression of the initial detection accuracy after simplification is:

[0044] like Figure 3 As shown, it is a two-dimensional coordinate diagram of the preliminary detection accuracy-first bacteria coefficient provided in the embodiment of the present application (the second bacteria coefficient to be tested is a fixed value of 1.5 at this time), by Figure 3 It can be seen that the preliminary detection accuracy increases with the increase of the first bacterial coefficient to be detected. Among them, the first bacterial coefficient to be detected increases with the fluorescence colorimetric (i.e., the restriction expression of the fluorescence colorimetric score) ) increases, and the fluorescence colorimetric increases with the increase of the average fluorescence colorimetric.

[0045] like Figure 4 As shown, it is a two-dimensional coordinate diagram of the preliminary detection accuracy-second bacteria coefficient to be tested provided by the embodiment of the present application (at this time, the first bacteria coefficient to be tested is a fixed value of 1.35). Among them, the second bacteria coefficient to be tested increases with the increase of the fluorescence penetration rate score and the fluorescence labeling efficiency score, and the fluorescence penetration rate score and the fluorescence labeling efficiency score increase with the increase of the number of bacteria labels to be tested.

[0046] It should be noted that the fluorescence labeling efficiency score also indirectly affects the value of the fluorescence penetration score. When the fluorescence labeling efficiency score increases (that is, the proportion of labeled test bacteria in the overall bacterial population increases), this provides more test bacteria samples for the detection of fluorescence penetration rate, which means that the number of successfully infiltrated test bacteria increases (that is, the fluorescence penetration rate of the test bacteria also increases accordingly), indicating that more test bacteria can be clearly identified, thereby improving the accuracy of bacterial counting, thereby achieving an improvement in the reliability of water bacteria concentration detection based on data processing, and effectively solving the problem of low reliability of water bacteria concentration detection based on data processing in the prior art.

[0047] Furthermore, the specific process of determining whether to perform a laser layer-by-layer scanning operation based on the obtained preliminary detection accuracy is as follows: determining whether the obtained preliminary detection accuracy is within the preliminary detection accuracy threshold range; if so, performing a laser layer-by-layer scanning operation; otherwise, not performing a laser layer-by-layer scanning operation and re-performing fluorescent marking.

[0048] It should be added that the specific steps of the laser layer-by-layer scanning operation are: according to the preset scanning parameters and the preset scanning path, the bacteria to be tested in the fluorescent area are scanned layer by layer, and the division of the scanning area during the layer-by-layer scanning process is monitored in real time to determine whether the scanning area is consistent with the preset scanning area: if the scanning area is consistent with the preset scanning area, the fluorescence signal is obtained and converted into digital data; if the scanning area is inconsistent with the preset scanning area, the preset scanning parameters are adjusted and the layer-by-layer scanning is performed again until the scanning area is consistent with the preset scanning area.

[0049] Among them, the scanning parameters include scanning speed, laser power and laser wavelength; the fluorescence signal is the fluorescence color development produced by the laser beam exciting the fluorescent marker during the layer-by-layer scanning process; the digital data includes the fluorescence signal intensity, layer-by-layer scanning path and scanning area range.

[0050] In this embodiment, the fluorescence signal intensity is acquired through a fluorescence detector, and the layer-by-layer scanning path is the moving trajectory of the laser beam (monitored in real time by a sensor); a scanning area is preset (acquired by preset boundaries and coordinates defined by personnel). When the scanning area is inconsistent with the preset scanning area, it is necessary to adjust the preset scanning parameters, for example, by increasing the speed gear of the control panel in the scanning instrument to increase the scanning speed, by increasing the power gear of the control panel in the scanning instrument to increase the laser power, and by increasing the wavelength parameter of the control panel in the laser to increase the laser wavelength. It should be noted that each time an adjustment is made, the scanning area needs to be acquired again until the acquired scanning area is consistent with the preset scanning area and the adjustment is stopped, thereby improving the accuracy and efficiency of the laser layer-by-layer scanning operation.

[0051] Furthermore, the regional scanning evaluation value is obtained by the following method: E1, when the obtained preliminary detection accuracy is within the preliminary detection accuracy threshold range (including the maximum and minimum values ​​of the historical preliminary detection accuracy), the interruption of the laser wavelength of the laser beam in the scanning channel is monitored in real time to determine whether the receiving end of the scanning channel receives the laser beam. If so, it indicates that the corresponding bacteria to be tested in the scanning channel are living cells and E2 is executed; otherwise, it indicates that the corresponding cells to be tested in the scanning channel are dead cells and are removed; E2, the average scanning speed, average laser power and density of the bacteria to be tested of the laser beam in a preset time period are obtained to determine whether the density of the bacteria to be tested is within the density threshold range (including the maximum and minimum values ​​of the historical density). If If yes, execute E3, otherwise it indicates that the sensitivity of the corresponding bacteria to be tested to the laser beam does not meet the preset requirements and prompts the preset personnel to dilute the bacteria to be tested; E3, judge whether the average scanning speed is within the maximum allowable scanning speed range (including the maximum and minimum values ​​of the historical scanning speed), and at the same time judge whether the average laser power is within the maximum allowable laser power range (including the maximum and minimum values ​​of the historical laser power). When the average scanning speed is within the maximum allowable scanning speed range and the average laser power is within the maximum allowable laser power range, execute E4, otherwise no calculation is performed and the initial scanning speed and initial laser power of the laser beam are reduced; E4, obtain the scanning speed score (that is, the value in the restriction expression of the regional scanning evaluation value) ), laser power fraction (i.e., the limiting expression of the area scan evaluation value ) and the fluorescence signal intensity score (i.e., the limiting expression of the area scan evaluation value ) and combined with the obtained preliminary detection accuracy (i.e., the restricted expression of the regional scan evaluation value ) to obtain the regional scan evaluation value.

[0052] In this embodiment, the initial scanning speed and initial laser power of the laser beam are reduced by reducing the scanning speed parameters and laser power parameters on the laser control panel. After the reduction, the average scanning speed and average laser power within a preset time period are obtained until the average scanning speed is within the maximum allowable scanning speed range and the average laser power is within the maximum allowable laser power range; the scanning speed score is the ratio of the average scanning speed to the reference average scanning speed, the laser power score is the ratio of the average laser power to the reference average laser power, and the fluorescence signal intensity score is the ratio of the fluorescence signal intensity to the maximum allowable fluorescence signal intensity; wherein, the reference average scanning speed is represented by the result of summing and averaging the historical average scanning speeds of the laser beam in the historical time period in the preset database, the reference average laser power is represented by the result of summing and averaging the historical average laser powers of the laser beam in the historical time period in the preset database, and the maximum allowable fluorescence signal intensity is represented by the result of summing and averaging the maximum values ​​of the historical fluorescence signal intensities of the laser beam in the historical time period in the preset database.

[0053] Specifically, the specific limiting expression of the area scan evaluation value is:

[0054]

[0055] Wherein, m is the number of the bacterial species to be tested, m=1, 2, 3, m=1 represents Escherichia coli, m=2 represents Staphylococcus aureus, m=3 represents Pseudomonas aeruginosa, t is the number of the preset time period, t=1, 2, ..., T, T is the total number of preset time periods, e is a natural constant, It represents the regional scanning evaluation value of the mth type of bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, represents the initial detection accuracy of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, ΔCHU represents the initial detection accuracy threshold range, It represents the scanning speed score of the mth type of bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, It represents the average scanning speed of the mth type of bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, represents the reference average scanning speed of the mth bacteria to be tested, represents the maximum allowable scanning speed range of the mth bacteria to be tested, It represents the laser power fraction of the mth bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, represents the average laser power during the laser scanning operation of the mth type of bacteria to be tested in the water sample of the experimental group during the tth preset time period, represents the reference average laser power of the mth bacteria to be tested, represents the maximum allowable laser power range of the mth bacteria to be tested, It represents the fluorescence signal intensity score of the mth bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, It represents the fluorescence signal intensity of the mth bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, Represents the maximum allowable fluorescence signal intensity of the mth bacteria to be tested.

[0056] Specifically, the preliminary detection accuracy threshold range represents the range corresponding to the maximum and minimum values ​​of the historical preliminary detection accuracy of the bacteria to be tested in the historical time period in the preset database; the density threshold range represents the range corresponding to the maximum and minimum values ​​of the historical density of the bacteria to be tested in the historical time period in the preset database, and the density of the bacteria to be tested is obtained by a flow cytometer; the maximum allowable scanning speed range represents the range corresponding to the maximum and minimum values ​​of the historical scanning speed of the bacteria to be tested in the historical time period in the preset database, and the average scanning speed is the average value of the scanning speed of the laser beam in the preset time period (obtained by a laser confocal microscope); the maximum allowable laser power range represents the range corresponding to the maximum and minimum values ​​of the historical laser power of the bacteria to be tested in the historical time period in the preset database, and the average laser power is the average value of the laser power of the laser beam in the preset time period (obtained by a laser power meter).

[0057] Specifically, assuming that the initial detection accuracy threshold range is between 0.9 and 1.25, the statistical table of changes in the regional scanning evaluation value is shown in Table 1:

[0058] Table 1 Statistics of changes in regional scanning evaluation values

[0059]

[0060] It should be understood that, as shown in Table 1, the regional scanning evaluation value increases with the increase of the preliminary detection accuracy, the scanning speed score, the laser power score and the fluorescence signal intensity score, among which the scanning speed score increases with the increase of the average scanning speed, the laser power score increases with the increase of the average laser power, and the fluorescence signal intensity score increases with the increase of the fluorescence signal intensity. It should be noted that the laser power score also indirectly affects the value of the fluorescence signal intensity.

[0061] When the laser power fraction increases (i.e., the average laser power increases), it means that the number of fluorescent markers excited increases, which in turn leads to an increase in the fluorescence signal intensity (i.e., the fluorescence signal intensity fraction increases). This helps to more accurately balance the number of bacteria to be tested based on data processing, reduces the possibility of erroneous identification and missed identification, and thus achieves an improvement in the reliability of bacteria concentration detection in water based on data processing, effectively solving the problem of low reliability of bacteria concentration detection in water based on data processing in the prior art.

[0062] Furthermore, the specific process of determining whether to generate a two-dimensional image of the bacteria to be tested based on the acquired regional scanning evaluation value is as follows: determining whether the acquired regional scanning evaluation value is greater than a preset regional scanning evaluation value: if the acquired regional scanning evaluation value is greater than the preset regional scanning evaluation value, then generating a two-dimensional image of the bacteria to be tested corresponding to the completed layer-by-layer laser scanning operation; if the acquired regional scanning evaluation value is not greater than the preset regional scanning evaluation value, then adjusting the preset scanning parameters and performing the laser layer-by-layer scanning operation again.

[0063] The preset regional scanning evaluation value is represented by the sum and average of the historical regional scanning evaluation values ​​of the bacteria to be tested in the historical time period in the preset database.

[0064] In this embodiment, when the acquired regional scanning evaluation value is not greater than the preset regional scanning evaluation value, it indicates that the accuracy of the laser layer-by-layer scanning operation in dividing the area to be tested within the preset time period does not meet the preset accuracy requirements (set by the preset personnel). At this time, it is necessary to adjust the preset scanning parameters. The specific adjustment steps are as described in the above embodiment, which realizes the improvement of the accuracy and efficiency of the laser layer-by-layer scanning operation.

[0065] Furthermore, the specific steps of performing superpixel segmentation on the preprocessed three-dimensional image to generate superpixels according to the superpixel algorithm include: acquiring image information data in the three-dimensional image and segmenting the three-dimensional image into superpixel regions according to the superpixel algorithm, and identifying the target region in the superpixel region according to similar pixel features in the three-dimensional image; the image information data includes color data, brightness data and spatial domain information data; the superpixel region is used to reflect the structural information of the bacteria to be tested in the three-dimensional image; similar pixel features include similar color features, similar brightness features and similar texture features.

[0066] In this embodiment, color data (usually including RGB values ​​or HSV values) and brightness data (usually including grayscale values ​​or brightness intensity) are used by the superpixel algorithm to distinguish the types of bacteria to be tested (such as Escherichia coli, Staphylococcus aureus and Pseudomonas aeruginosa) and impurities (such as suspended matter). The spatial domain information data is used by the superpixel algorithm to maintain the continuity between adjacent pixels in the three-dimensional image (that is, the positional relationship between adjacent pixels in three-dimensional space). The target area is the area in the superpixel area with the highest similarity to the structural features of the target bacteria to be tested (the maximum value automatically generated by the superpixel algorithm).

[0067] This example introduces a superpixel algorithm to segment a three-dimensional image into superpixel regions, where the superpixel regions retain key information in the image, such as color, brightness, and texture features, which are essential for identifying bacterial structures. At the same time, the superpixel regions also retain the spatial structural information of the image, which helps to understand the distribution and morphology of bacteria in three-dimensional space (the boundaries of bacterial structures can be more accurately identified), thereby reducing the complexity of subsequent image processing. Secondly, compared with the pixel-level images of the prior art, the superpixel regions provide a higher level of image representation, making the analysis results more intuitive and easy to understand.

[0068] Furthermore, superpixels are generated, and then the generated superpixels are identified and segmented based on a pre-trained convolutional neural network model. The specific process is: the generated superpixels are segmented according to the superpixel segmentation algorithm to obtain segmented superpixels, and at the same time, the segmented superpixels are preliminarily identified based on the pre-trained convolutional neural network model to obtain bacteria images to be tested and bacteria images not to be tested, and the segmented superpixels are automatically generated by the pixel similarity in the three-dimensional image; the similarity between adjacent superpixels in the bacteria image to be tested is obtained and it is determined whether the obtained similarity is within a similarity threshold range (including a maximum value and a minimum value equal to the historical similarity). If so, removal processing is performed to exclude superpixels that are not within the similarity threshold range, otherwise merging processing is performed to obtain characteristic parameters of the bacteria to be tested.

[0069] In this embodiment, the pre-trained convolutional neural network model has been trained on a large-scale image dataset (such as ImageNet), which is usually used for image classification, and usually includes VGG (Visual Geometry Group) and Res Net (Residual Network). It should be noted that the segmented superpixels are image blocks of fixed size. The preprocessed superpixel image blocks are used as input and input into the pre-trained convolutional neural network model one by one, and the probability or label of each superpixel image block belonging to the bacterial image to be tested or the bacterial image not to be tested is output, and then the recognition results of each superpixel are integrated to obtain the distribution of the bacterial images to be tested and the bacterial images not to be tested in the entire image.

[0070] Specifically, the similarity threshold range represents the range corresponding to the maximum and minimum historical similarities between adjacent pixels in the bacterial image to be tested in a preset database. The feature parameters are obtained through computer vision, usually including area (the area of ​​the superpixel area), perimeter (the boundary length of the superpixel area), density (the average value of the pixel values ​​in the superpixel area) and shape factors (such as aspect ratio and circularity), thereby achieving more accurate segmentation and recognition of superpixels.

[0071] Further, the concentration of the bacteria to be tested is obtained, which also includes obtaining the environmental impact score according to the degree of influence of the test environmental factors on the concentration of the bacteria to be tested; the test environmental factors include dissolved oxygen content, light intensity and temperature of the water sample of the experimental group; the specific steps for obtaining the environmental impact score are: L1, when the obtained regional scanning evaluation value is greater than the preset regional scanning evaluation value, real-time monitoring of the changes in the test environmental factors in the fluorescent area where the bacteria to be tested are located during the detection process to obtain the average dissolved oxygen content, the average light intensity and the average temperature of the water sample of the experimental group, and judging whether the average dissolved oxygen content is within the allowable range of the dissolved oxygen content (including being equal to the maximum and minimum values ​​of the historical dissolved oxygen content). If so, executing L 2, otherwise the preset personnel are prompted to adjust the dissolved oxygen conditions in the fluorescent area where the bacteria to be tested are located; L2, determine whether the average light intensity is within the allowable range of light intensity (including equal to the maximum and minimum values ​​of the historical light intensity), if so, execute L3, otherwise the preset personnel are prompted to adjust the light source conditions in the fluorescent area where the bacteria to be tested are located; L3, determine whether the average temperature of the water sample in the experimental group is within the allowable range of the water sample temperature in the experimental group (including equal to the maximum and minimum values ​​of the water sample temperature in the historical experimental group), if so, execute L4, otherwise the preset personnel are prompted to adjust the water temperature conditions in the fluorescent area where the bacteria to be tested are located; L4, obtain the dissolved oxygen content score (that is, the environmental impact score in the restriction expression ), light intensity fraction (i.e., the limiting expression of environmental impact fraction ) and the experimental group water sample temperature score (i.e., the environmental impact score in the restricted expression ) and combined with the obtained regional scan assessment value (i.e., the environmental impact score in the limiting expression ) to obtain the environmental impact score; wherein, the dissolved oxygen content score is the ratio of the difference between the maximum allowable dissolved oxygen content and the average dissolved oxygen content within the preset time period to the maximum allowable dissolved oxygen content; the light intensity score is the ratio of the difference between the maximum allowable light intensity and the average light intensity within the preset time period to the maximum allowable light intensity; the experimental group water sample temperature score is the ratio of the difference between the maximum allowable experimental group water sample temperature and the average temperature of the experimental group water samples within the preset time period to the maximum allowable experimental group water sample temperature.

[0072] In this embodiment, the average dissolved oxygen content is the average value of the dissolved oxygen content in the fluorescent area where the water samples in the experimental group are located (measured by a dissolved oxygen meter), the average light intensity is the average value of the light intensity received by the fluorescent area in the water samples in the experimental group within a preset time period (measured by a light intensity meter), and the average temperature of the water samples in the experimental group is the average value of the temperature of the water samples in the experimental group (measured by a thermometer) within the preset time period.

[0073] The allowable range of dissolved oxygen content represents the range corresponding to the maximum value (maximum allowable dissolved oxygen content) and the minimum value of the historical dissolved oxygen content of the experimental group water samples in the preset database within the historical time period. The allowable range of light intensity represents the range corresponding to the maximum value (maximum allowable light intensity) and the minimum value of the historical light intensity of the experimental group water samples in the preset database within the historical time period. The allowable range of experimental group water sample temperature represents the range corresponding to the maximum value (maximum allowable experimental group water sample temperature) and the minimum value of the historical experimental group water sample temperature of the experimental group water samples in the preset database within the historical time period.

[0074] It should be understood that when the average dissolved oxygen content is not within the allowable range of dissolved oxygen content, the oxygen content in the water samples of the experimental group is increased by introducing oxygen or nitrogen is introduced to reduce the oxygen content in the water samples of the experimental group. When the average light intensity is not within the allowable range of light intensity, the light intensity of the fluorescent area where the bacteria to be tested are located is increased by moving closer to the light source or is reduced by moving away from the light source. When the average temperature of the water samples in the experimental group is not within the allowable range of the temperature of the water samples in the experimental group, a heater is used to increase the temperature of the water samples in the experimental group or a cooler is used to reduce the temperature of the water samples in the experimental group.

[0075] Specifically, the specific limiting expression of the environmental impact score is:

[0076]

[0077] Wherein, m is the number of the bacterial species to be tested, m=1, 2, 3, m=1 represents Escherichia coli, m=2 represents Staphylococcus aureus, m=3 represents Pseudomonas aeruginosa, t is the number of the preset time period, t=1, 2, ..., T, T is the total number of preset time periods, e is a natural constant, It represents the environmental impact score of the mth bacteria to be tested in the water sample of the experimental group in the tth preset time period, represents the regional scanning evaluation value of the mth type of bacteria to be tested in the water sample of the experimental group during the laser layer-by-layer scanning operation in the tth preset time period, QU 0 Indicates the preset area scan evaluation value, It represents the average dissolved oxygen content of the mth bacteria to be tested in the water sample of the experimental group in the tth preset time period, represents the maximum allowable dissolved oxygen content of the mth bacteria to be tested, ΔY1 represents the allowable range of dissolved oxygen content, It represents the average light intensity of the mth bacteria to be tested in the water sample of the experimental group in the tth preset time period, represents the maximum allowable light intensity of the mth species of bacteria to be tested, ΔY2 represents the allowable range of light intensity, It represents the average temperature of the water sample of the experimental group for the mth species of bacteria to be tested in the water sample of the experimental group in the tth preset time period, It represents the maximum allowable experimental group water sample temperature of the mth bacteria to be tested, and ΔY3 represents the allowable range of the experimental group water sample temperature.

[0078] It should be understood that the environmental impact score decreases with the increase of the regional scanning assessment value, the average dissolved oxygen content, the average light intensity and the average temperature of the experimental group water samples. Among them, the average light intensity also indirectly affects the value of the average dissolved oxygen content. Assume that in a closed water system, it contains algae, plankton and bacteria. In this system, algae produce oxygen through photosynthesis, while plankton and bacteria consume oxygen through respiration.

[0079] When the average light intensity increases, algae can photosynthesize more efficiently, thereby producing more oxygen, which will lead to an increase in the dissolved oxygen content in the water body. On the contrary, when the average light intensity decreases, photosynthesis weakens and the oxygen produced decreases. At the same time, since the respiration of plankton and bacteria continues and consumes oxygen, the dissolved oxygen content in the water body will also decrease. By considering the above-mentioned indirect impact mechanism, it is helpful to more accurately evaluate the living environment and concentration of the details to be tested in the water body, improve the accuracy of obtaining the environmental impact score, and then realize the improvement of the reliability of water bacteria concentration detection based on data processing, which effectively solves the problem of low reliability of water bacteria concentration detection based on data processing in the prior art.

[0080] Furthermore, an environmental impact score is obtained, and then it is also included to determine whether to re-test the bacteria to be tested based on the concentration of the bacteria to be tested obtained under the detection environment factors. The specific judgment process is: Step 1, determine whether the obtained concentration of the bacteria to be tested is within a preset concentration range (including the maximum and minimum values ​​equal to the historical concentrations). If so, it indicates that the current detection environment of the bacteria to be tested meets the bacteria detection conditions, and the bacteria to be tested is not re-tested at this time. Otherwise, it indicates that the current detection environment of the bacteria to be tested does not meet the bacteria detection conditions. At this time, it is necessary to adjust the detection environment of the water sample of the experimental group and re-test the bacteria to be tested; Step 2, determine whether the obtained concentration of the bacteria to be tested is greater than the reference concentration of the bacteria to be tested in the water sample of the control group. If so, it indicates that the fluorescent marker in the water sample of the experimental group is effectively marked, otherwise it indicates that the fluorescent marker in the water sample of the experimental group is invalidly marked and the fluorescent marker parameters are adjusted.

[0081] In the present embodiment, it should be supplemented that it is also possible to determine whether it is necessary to adjust the detection environment of the water samples in the experimental group by judging whether the obtained environmental impact score is within the environmental impact threshold range (including the range corresponding to the maximum and minimum values ​​of the historical environmental impact score). The specific judgment process is as follows: if the obtained environmental impact score is within the environmental impact threshold range, it indicates that the detection environment in the water samples in the experimental group has no effect on the concentration of the bacteria to be tested. If the obtained environmental impact score is not within the environmental impact threshold range, it indicates that the detection environment in the water samples in the experimental group has an effect on the concentration of the bacteria to be tested. In this case, it is necessary to adjust the detection environment of the water samples in the experimental group and re-test the bacteria to be tested. The environmental impact threshold range represents the range corresponding to the maximum and minimum values ​​of the historical environmental impact scores of the bacteria to be tested in the historical time period in the preset database.

[0082] Specifically, adjusting the detection environment of the experimental group water samples usually includes adjusting the pH value (using sodium hydroxide to increase or decrease the pH of the experimental group water samples) and adjusting nutrients (adding carbon sources or nitrogen sources to the experimental group water samples); adjusting the fluorescent marker parameters usually includes adjusting the marker concentration (increasing the marker concentration by the preset personnel helps to fully label, and reducing the marker concentration helps to avoid fluorescent marker interference in the background area), adjusting the marker type (changing DAPI with SYBR Green or changing SYBR Green with DAPI) and adjusting the marker time (increasing or shortening the contact time between the fluorescent marker and the bacteria to be tested by the preset personnel).

[0083] Specifically, the preset concentration range represents the range corresponding to the maximum and minimum values ​​of the historical concentrations of the bacteria to be tested in the preset database within the historical time period, and the reference bacteria concentration to be tested (which is the optimal experimental result) is represented by the result of summing and averaging the historical experimental concentrations of the bacteria to be tested in the preset database; this example can quickly determine whether the current detection environment meets the bacteria detection conditions by judging whether the concentration of the bacteria to be tested is within the preset concentration range. If the conditions are met, there is no need to re-test, which greatly saves time and resources; by comparing the bacterial concentrations of the experimental group and the control group, the effectiveness of the fluorescent marker can be quickly determined, thereby avoiding invalid detection and further improving detection efficiency.

[0084] In summary, the embodiment of the present application uses fluorescent markers to preliminarily mark the test sample data to obtain preliminary detection accuracy, and at the same time monitors the test area division process during the laser layer-by-layer scanning operation in real time to obtain the regional scanning evaluation value, and then performs superpixel segmentation on the acquired three-dimensional image according to the superpixel algorithm to generate superpixels. Finally, the number of bacteria to be tested in the superpixels is counted and the concentration of bacteria to be tested is obtained in combination with the preset experimental group water sample volume, thereby achieving improved superpixel generation accuracy and reliability, and further achieving improved reliability of water bacteria concentration detection based on data processing, effectively solving the problem of low reliability of water bacteria concentration detection based on data processing in the prior art.

[0085] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0087] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0089] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0090] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for detecting bacterial concentration in water based on image processing technology, characterized in that: The following steps are involved: S1, blank marking and fluorescent marking are performed on the acquired control sample data and the sample data to be tested, respectively, and during the marking process, the marking effect of the fluorescent marker on the bacteria to be tested is preliminarily evaluated to obtain a preliminary detection accuracy, wherein the preliminary detection accuracy is used to quantify the marking effect of the fluorescent marker on the bacteria to be tested and the preliminary reliability of the detection; S2, judging whether to perform a laser layer-by-layer scanning operation based on the obtained preliminary detection accuracy, and if so, monitoring the process of dividing the area to be tested during the laser layer-by-layer scanning operation in real time to obtain a regional scanning evaluation value, and judging whether to generate a two-dimensional image of the bacteria to be tested based on the obtained regional scanning evaluation value, wherein the regional scanning evaluation value is used to evaluate the accuracy of the laser layer-by-layer scanning operation in dividing the area to be tested within a preset time period; S3, accumulating the generated two-dimensional images to obtain a three-dimensional image and performing preprocessing, and performing superpixel segmentation on the preprocessed three-dimensional image according to a superpixel algorithm to generate superpixels; S4, counting the number of bacteria to be tested after being identified and segmented by the superpixel within a preset time period and obtaining the concentration of the bacteria to be tested in combination with the preset volume of the experimental group water sample.

2. A method for detecting bacterial concentration in water based on image processing technology as claimed in claim 1, characterized in that: The preliminary detection accuracy is obtained by the following method: Obtaining fluorescence data of the bacteria to be tested in the water sample of the experimental group within a preset time period, and simultaneously obtaining reference fluorescence data of the corresponding bacteria to be tested in the water sample of the control group within the same time period, wherein the fluorescence data includes fluorescence colorimetry, fluorescence penetration rate and fluorescence labeling efficiency, and the reference fluorescence data includes reference fluorescence colorimetry, reference fluorescence penetration rate and reference fluorescence labeling efficiency; The acquired fluorescence data and reference fluorescence data are processed respectively to obtain the fluorescence colorimetric score, fluorescence penetration score and fluorescence labeling efficiency score, and the preliminary detection accuracy is obtained based on this; The fluorescence colorimetric score is the ratio of the fluorescence colorimetric to the reference fluorescence colorimetric; The fluorescence permeability score is the ratio of the fluorescence permeability to the reference fluorescence permeability; The fluorescence labeling rate score is the ratio of the fluorescence labeling rate to the reference fluorescence labeling rate; The fluorescence colorimetric index is the ratio of the difference between the average fluorescence intensity of the fluorescence area where the bacteria to be detected are located and the reference average fluorescence intensity of the background area to the reference average fluorescence intensity of the background area; The fluorescence penetration rate is the ratio of the number of bacterial markers to be tested to the total number of bacteria to be tested; The fluorescent labeling efficiency is the ratio of the number of labeled bacteria to be tested to the number of labeled reference bacteria to be tested.

3. A method for detecting bacterial concentration in water based on image processing technology as claimed in claim 2, characterized in that: The specific limiting expression of the preliminary detection accuracy is: Wherein, m is the number of the bacterial species to be tested, m=1, 2, 3, m=1 represents Escherichia coli, m=2 represents Staphylococcus aureus, m=3 represents Pseudomonas aeruginosa, t is the number of the preset time period, t=1, 2, ..., T, T is the total number of preset time periods, e is a natural constant, It represents the initial detection accuracy of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the fluorescence colorimetric score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the reference fluorescence colorimetric value of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the average fluorescence colorimetric value of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period. It represents the reference average fluorescence colorimetric value of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the fluorescence penetration rate score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, represents the reference fluorescence penetration rate of the mth test bacteria in the control water sample during the blank labeling process, It represents the number of bacteria labeled in the experimental group water sample during the fluorescence labeling process of the mth bacteria in the water sample in the tth preset time period. It represents the total number of the mth bacteria to be tested in the control group water sample during the blank labeling process. It represents the fluorescence labeling efficiency score of the mth bacteria to be tested in the water sample of the experimental group during the fluorescence labeling process in the tth preset time period, It represents the reference fluorescence labeling rate of the mth bacteria to be tested in the control water sample during the blank labeling process. It represents the reference labeled bacteria of the mth tested bacteria in the control group water sample during the blank labeling process.

4. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 1, characterized in that: The specific process of determining whether to perform the laser layer-by-layer scanning operation based on the acquired preliminary detection accuracy is as follows: Determine whether the obtained preliminary detection accuracy is within the preliminary detection accuracy threshold range, if so, perform a laser layer-by-layer scanning operation, otherwise do not perform the laser layer-by-layer scanning operation and re-perform fluorescent marking; The specific steps of the laser layer-by-layer scanning operation are: The bacteria to be tested in the fluorescent area are scanned layer by layer according to the preset scanning parameters and the preset scanning path, and the division of the scanning area during the layer-by-layer scanning process is monitored in real time to determine whether the scanning area is consistent with the preset scanning area: If the scanning area is consistent with the preset scanning area, the fluorescence signal is acquired and converted into digital data; If the scanning area is inconsistent with the preset scanning area, adjust the preset scanning parameters and re-scan layer by layer until the scanning area is consistent with the preset scanning area; The scanning parameters include scanning speed, laser power and laser wavelength; The fluorescent signal is the fluorescence coloration generated by the laser beam exciting the fluorescent marker during the layer-by-layer scanning process; The digital data includes fluorescence signal intensity, layer-by-layer scanning path and scanning area range.

5. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 1, characterized in that: The area scan evaluation value is obtained by the following method: E1, real-time monitoring of the interruption of the laser wavelength of the laser beam in the scanning channel, and judging whether the receiving end of the scanning channel receives the laser beam. If so, it indicates that the corresponding bacteria to be tested in the scanning channel are living cells and E2 is executed, otherwise, removal is performed; E2, obtaining the average scanning speed, average laser power and density of the bacteria to be tested within a preset time period, and determining whether the density of the bacteria to be tested is within a density threshold range. If so, executing E3, otherwise prompting the preset personnel to dilute the bacteria to be tested; E3, judging whether the average scanning speed is within the maximum allowable scanning speed range, and judging whether the average laser power is within the maximum allowable laser power range. When the average scanning speed is within the maximum allowable scanning speed range and the average laser power is within the maximum allowable laser power range, executing E4, otherwise no calculation is performed; E4, obtain the scanning speed score, laser power score and fluorescence signal intensity score and combine them with the obtained preliminary detection accuracy to obtain the regional scanning evaluation value, the scanning speed score is the ratio of the average scanning speed to the reference average scanning speed, the laser power score is the ratio of the average laser power to the reference average laser power, and the fluorescence signal intensity score is the ratio of the fluorescence signal intensity to the maximum allowable fluorescence signal intensity.

6. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 1, characterized in that: The specific process of determining whether to generate a two-dimensional image of the bacteria to be tested based on the acquired regional scanning evaluation value is as follows: Determine whether the obtained area scan evaluation value is greater than the preset area scan evaluation value: If the acquired regional scanning evaluation value is greater than the preset regional scanning evaluation value, a two-dimensional image of the bacteria to be tested corresponding to the completed layer-by-layer laser scanning operation is generated; If the obtained regional scanning evaluation value is not greater than the preset regional scanning evaluation value, the preset scanning parameters are adjusted and the laser layer-by-layer scanning operation is performed again.

7. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 1, characterized in that: The specific steps of performing superpixel segmentation on the preprocessed three-dimensional image according to the superpixel algorithm to generate superpixels include: Acquire image information data in a three-dimensional image and segment the three-dimensional image into super-pixel regions according to a super-pixel algorithm, and identify a target region in the super-pixel region according to similar pixel features in the three-dimensional image; The image information data includes color data, brightness data and spatial domain information data; The super pixel region is used to reflect the structural information of the bacteria to be tested in the three-dimensional image; The similar pixel features include similar color features, similar brightness features and similar texture features.

8. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 1, characterized in that: The generation of superpixels further includes identifying and segmenting the generated superpixels based on a pre-trained convolutional neural network model. The specific process is as follows: Segmenting the generated superpixels according to a superpixel segmentation algorithm to obtain segmented superpixels, and preliminarily identifying the segmented superpixels based on a pre-trained convolutional neural network model to obtain a bacterial image to be tested and a bacterial image not to be tested, wherein the segmented superpixel representation is automatically generated by pixel similarity in the three-dimensional image; The similarity between adjacent superpixels in the bacterial image to be tested is obtained and it is determined whether the obtained similarity is within the similarity threshold range. If so, removal processing is performed, otherwise merging processing is performed to obtain characteristic parameters of the bacteria to be tested, and the characteristic parameters include area, perimeter, density and shape factor.

9. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 1, characterized in that: The obtaining of the concentration of the bacteria to be tested also includes obtaining an environmental impact score according to the degree of influence of the detected environmental factors on the concentration of the bacteria to be tested; The detection environmental factors include dissolved oxygen content, light intensity and temperature of the water samples in the experimental group; The specific steps for obtaining the environmental impact score are as follows: L1, real-time monitoring of changes in environmental factors in the fluorescent area where the bacteria to be tested are located during the test process to obtain the average dissolved oxygen content, average light intensity and average temperature of the water samples in the experimental group, and determine whether the average dissolved oxygen content is within the allowable range of dissolved oxygen content. If so, execute L2, otherwise prompt the preset personnel to adjust the dissolved oxygen conditions of the corresponding fluorescent area where the bacteria to be tested are located; L2, determine whether the average light intensity is within the allowable range of light intensity, if so, execute L3, otherwise prompt the preset personnel to adjust the light source conditions corresponding to the fluorescent area where the bacteria to be tested are located; L3, determine whether the average temperature of the water sample of the experimental group is within the allowable range of the water sample temperature of the experimental group. If so, execute L4, otherwise prompt the preset personnel to adjust the water temperature conditions of the corresponding fluorescent area where the bacteria to be tested are located; L4, obtain the dissolved oxygen content score, light intensity score and temperature score of the experimental group water samples and combine them with the obtained regional scanning assessment values ​​to obtain the environmental impact score; The dissolved oxygen content fraction is the ratio of the average dissolved oxygen content within a preset time period to the maximum allowable dissolved oxygen content; The light intensity score is the ratio of the average light intensity in a preset time period to the maximum allowable light intensity; The experimental group water sample temperature score is the ratio of the average temperature of the experimental group water samples within a preset time period to the maximum allowable experimental group water sample temperature.

10. A method for detecting bacteria concentration in water based on image processing technology as claimed in claim 9, characterized in that: The step of obtaining the environmental impact score further includes determining whether to re-test the bacteria under test based on the concentration of the bacteria under test obtained under the test environmental factors. The specific determination process is as follows: Step 1: determine whether the obtained concentration of the bacteria to be tested is within the preset concentration range. If so, the bacteria to be tested will not be tested again. Otherwise, it indicates that the current testing environment of the bacteria to be tested does not meet the bacteria testing conditions. At this time, the testing environment of the water sample of the experimental group needs to be adjusted and the bacteria to be tested needs to be tested again. Step 2: Determine whether the obtained test bacteria concentration is greater than the reference test bacteria concentration in the control group water sample. If so, it indicates that the fluorescent marker in the experimental group water sample is effectively marked. Otherwise, it indicates that the fluorescent marker in the experimental group water sample is invalid and the fluorescent marker parameters are adjusted.

Images