CN118655116B - Automatic analysis method for full algae based on fluorescent quantum dots - Google Patents
Automatic analysis method for full algae based on fluorescent quantum dots Download PDFInfo
- Publication number
- CN118655116B CN118655116B CN202411152399.5A CN202411152399A CN118655116B CN 118655116 B CN118655116 B CN 118655116B CN 202411152399 A CN202411152399 A CN 202411152399A CN 118655116 B CN118655116 B CN 118655116B
- Authority
- CN
- China
- Prior art keywords
- representing
- training
- sample
- hyperspectral data
- algae
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 241000195493 Cryptophyta Species 0.000 title claims abstract description 59
- 239000002096 quantum dot Substances 0.000 title claims abstract description 18
- 238000004458 analytical method Methods 0.000 title claims abstract description 17
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 78
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000005259 measurement Methods 0.000 claims abstract description 16
- 238000001514 detection method Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 239000013598 vector Substances 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 22
- 229930002868 chlorophyll a Natural products 0.000 claims description 22
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 claims description 22
- 230000003595 spectral effect Effects 0.000 claims description 20
- 238000002835 absorbance Methods 0.000 claims description 10
- 238000002310 reflectometry Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 8
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 claims description 5
- 229910052698 phosphorus Inorganic materials 0.000 claims description 5
- 239000011574 phosphorus Substances 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 59
- 239000011159 matrix material Substances 0.000 description 7
- 238000010521 absorption reaction Methods 0.000 description 4
- 238000002798 spectrophotometry method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000002795 fluorescence method Methods 0.000 description 2
- 239000007850 fluorescent dye Substances 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- UPYKUZBSLRQECL-UKMVMLAPSA-N Lycopene Natural products CC(=C/C=C/C=C(C)/C=C/C=C(C)/C=C/C1C(=C)CCCC1(C)C)C=CC=C(/C)C=CC2C(=C)CCCC2(C)C UPYKUZBSLRQECL-UKMVMLAPSA-N 0.000 description 1
- 108010053210 Phycocyanin Proteins 0.000 description 1
- 108010004729 Phycoerythrin Proteins 0.000 description 1
- 150000001746 carotenes Chemical class 0.000 description 1
- 235000005473 carotenes Nutrition 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- KBPHJBAIARWVSC-RGZFRNHPSA-N lutein Chemical compound C([C@H](O)CC=1C)C(C)(C)C=1\C=C\C(\C)=C\C=C\C(\C)=C\C=C\C=C(/C)\C=C\C=C(/C)\C=C\[C@H]1C(C)=C[C@H](O)CC1(C)C KBPHJBAIARWVSC-RGZFRNHPSA-N 0.000 description 1
- 229960005375 lutein Drugs 0.000 description 1
- ORAKUVXRZWMARG-WZLJTJAWSA-N lutein Natural products CC(=C/C=C/C=C(C)/C=C/C=C(C)/C=C/C1=C(C)CCCC1(C)C)C=CC=C(/C)C=CC2C(=CC(O)CC2(C)C)C ORAKUVXRZWMARG-WZLJTJAWSA-N 0.000 description 1
- 235000012680 lutein Nutrition 0.000 description 1
- 239000001656 lutein Substances 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- KBPHJBAIARWVSC-XQIHNALSSA-N trans-lutein Natural products CC(=C/C=C/C=C(C)/C=C/C=C(C)/C=C/C1=C(C)CC(O)CC1(C)C)C=CC=C(/C)C=CC2C(=CC(O)CC2(C)C)C KBPHJBAIARWVSC-XQIHNALSSA-N 0.000 description 1
- NCYCYZXNIZJOKI-UHFFFAOYSA-N vitamin A aldehyde Natural products O=CC=C(C)C=CC=C(C)C=CC1=C(C)CCCC1(C)C NCYCYZXNIZJOKI-UHFFFAOYSA-N 0.000 description 1
- FJHBOVDFOQMZRV-XQIHNALSSA-N xanthophyll Natural products CC(=C/C=C/C=C(C)/C=C/C=C(C)/C=C/C1=C(C)CC(O)CC1(C)C)C=CC=C(/C)C=CC2C=C(C)C(O)CC2(C)C FJHBOVDFOQMZRV-XQIHNALSSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/55—Specular reflectivity
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6428—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6428—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes"
- G01N2021/6439—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" with indicators, stains, dyes, tags, labels, marks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- General Engineering & Computer Science (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Analytical Chemistry (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Software Systems (AREA)
- Optics & Photonics (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Databases & Information Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computational Mathematics (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention relates to the technical field of algae analysis and discloses a full-algae automatic analysis method based on fluorescent quantum dots, which comprises the following steps of S101, acquiring hyperspectral data and water quality parameters of an experimental water sample; the method comprises the steps of S102, preprocessing hyperspectral data and water quality parameters of an experimental water sample through a sliding window to generate a characteristic sequence, S103, acquiring a fluorescent signal of the experimental water sample through a fluorescence quantum detection method, training a generation model based on the characteristic sequence and the fluorescent signal, S104, acquiring an actual measurement value of the algae concentration of the experimental water sample, and training a prediction model based on the fluorescent signal and the actual measurement value of the algae concentration, wherein the hyperspectral data and the water quality parameters are utilized to generate a fluorescent signal approaching to the fluorescent signal acquired through the fluorescence quantum detection method through the generation model, and a nonlinear mapping relation between the fluorescent signal and the algae concentration is established through the prediction model, so that the precision of measuring the algae concentration is improved.
Description
Technical Field
The invention relates to the technical field of algae analysis, in particular to a full-algae automatic analysis method based on fluorescent quantum dots.
Background
The existing algae analysis method mainly comprises the steps of 1, directly observing morphological structure characteristics of algae through a microscope, identifying and counting types, 2, spectrophotometry, namely, marking specific components in algae cells by taking fluorescent quantum dots as fluorescent probes by using the fluorescent quantum dots as fluorescent probes through proper equipment (such as a fluorescent microscope, a flow cytometer and the like) to detect the absorbance at specific wavelengths (generally 665 nanometers), and 3, fluorescence method, namely, fluorescent intensity of the chlorophyll a is directly proportional to the concentration of the chlorophyll a under the irradiation of excitation light at the specific wavelengths, and the concentration of the chlorophyll a is calculated according to a relation between the fluorescent intensity and the concentration of the chlorophyll a by a fluorescent photometer or a portable fluorescent photometer, wherein the spectrophotometry is used for marking the specific components in the algae cells.
However, both the microscopic counting method and the fluorescence quantum detection method need to be carried out in an experimental environment, have serious time delay and cannot measure the concentration of algae in real time, and the spectrophotometry and the fluorescence method can measure the concentration of algae in real time, but are easily influenced by fluorescence generated by other substances in a water sample, so that the measurement accuracy is not high.
Disclosure of Invention
The invention provides a full algae automatic analysis method based on fluorescent quantum dots, which solves the technical problems in the background technology.
The invention provides a full algae automatic analysis method based on fluorescent quantum dots, which comprises the following steps:
step S101, hyperspectral data and water quality parameters of an experimental water sample are obtained;
The hyperspectral data are represented by spectral reflectivities at wavelengths of 350 nanometers to 850 nanometers;
The water quality parameters comprise chlorophyll a concentration, total phosphorus content, suspended matter concentration and turbidity;
Step S102, preprocessing hyperspectral data and water quality parameters of an experimental water sample through a sliding window to generate a characteristic sequence;
The characteristic sequence comprises K sequence units, and each sequence unit corresponds to a combination vector of a sliding window;
step S103, obtaining a fluorescence signal of an experimental water sample by a fluorescence quantum detection method, and training a generation model based on a characteristic sequence and the fluorescence signal;
generating an input of a model as a characteristic sequence and outputting as a fluorescent signal;
The fluorescence signal is represented by fluorescence intensity with a wavelength of M nm to N nm;
Step S104, obtaining an algae concentration actual measurement value of the experimental water sample, and training a prediction model based on the fluorescence signal and the algae concentration actual measurement value;
The input of the prediction model is a fluorescence signal, and the output is an algae concentration predicted value;
Step 105, generating a characteristic sequence of the water sample to be detected according to the step 101 and the step 102, inputting the characteristic sequence into a trained generation model, outputting a fluorescence signal, inputting the fluorescence signal into a trained prediction model, and outputting an algae concentration predicted value of the water sample to be detected.
Further, M and N are custom parameters, K is determined by the size and step size of the sliding window, and the calculation formula is as follows:
;
Wherein length represents the wavelength length of the hyperspectral data, the wavelength length of the hyperspectral data is 850 nanometers minus 350 nanometers, which is equal to 500 nanometers, and size and step represent the size and step size of the sliding window respectively, wherein the size and step size of the sliding window are self-defined parameters.
Further, the absorbance value of the experimental water sample at the wavelength of 663 nanometers and 635 nanometers is measured by a portable spectrophotometer, and the chlorophyll a concentration of the experimental water sample is obtained by a chlorophyll a concentration calculation formula, wherein the chlorophyll a concentration calculation formula is as follows:
;
Wherein Chla represents the chlorophyll a concentration, Represents the absorbance value at a wavelength of 663 nm,The absorbance at 635 nm is expressed, vol is the volume, and Weight is the Weight.
Further, the hyperspectral data and the water quality parameters of the experimental water sample are preprocessed through the sliding window to generate a characteristic sequence, and the method comprises the following steps:
Step S201, extracting characteristic values of hyperspectral data in a sliding window;
The characteristic values of the hyperspectral data comprise the maximum value, the minimum value, the average value, the standard deviation, the median, the skewness and the kurtosis of the spectral reflectivity of the hyperspectral data;
the calculation formula of the skewness Skewness of the spectral reflectance of the hyperspectral data is as follows:
;
The Kurtosis of the spectral reflectance of the hyperspectral data is calculated as follows:
;
where size represents the size of the sliding window, Representing the spectral reflectance corresponding to the ith wavelength of the hyperspectral data within the sliding window,AndRespectively representing the average value and standard deviation of the spectral reflectivity of the hyperspectral data in the sliding window;
Step S202, normalizing the characteristic values of hyperspectral data in a sliding window and water quality parameters by a minimum and maximum normalization method;
and step S203, splicing the characteristic value of the hyperspectral data in the normalized sliding window and the normalized water quality parameter to obtain a combination vector.
Further, the generating model comprises K hidden layers, the kth hidden layer inputs the kth sequence unit of the characteristic sequence and outputs an updating vector, wherein K is more than or equal to 1 and less than or equal to K;
The update vector output by the Kth hidden layer is input to Q classifiers, and the classification spaces of the Q classifiers respectively represent the fluorescence intensities corresponding to the Q wavelengths of the fluorescence signal, wherein Q=M-N.
Further, the calculation formula for generating the model includes:
;
;
;
;
;
Wherein the method comprises the steps of AndRepresenting the update vectors output by the kth and kth-1 hidden layers respectively,Is assigned a value of 0 for each dimension value,Size and dimensions of (2)Is of the same size as the (a), when k-1=0, thenEach of the dimension values of (c) is 0,A kth sequence element representing a feature sequence of a kth hidden layer input,、、AndRepresenting the correlation coefficient, the concealment vector, the reset gate and the update gate of the kth concealment layer, respectively,、、、、、AndRespectively representing a first weight parameter, a second weight parameter, a third weight parameter, a fourth weight parameter, a fifth weight parameter, a sixth weight parameter and a seventh weight parameter of a kth hidden layer,、、AndRespectively representing a first bias parameter, a second bias parameter, a third bias parameter and a fourth bias parameter of the kth hidden layer,Representing point-wise multiplication, T representing a transpose operation, concat representing a splice operation, pile representing a stack operation, sigmoid representing a sigmoid activation function, tanh representing a hyperbolic tangent activation function, reLU representing a ReLU activation function.
Further, repeating the feature sequences generated in the steps S101 to S102 as training data for training the training samples of the generated model, wherein the fluorescent signals obtained through the fluorescence quantum detection method are used as sample labels for training the generated model, a discriminator is connected in the training process of the generated model, and the value output by the discriminator represents the probability value that the fluorescent signals output by the generated model belong to the sample labels of the training samples;
Loss function of discriminator The calculation formula of (2) is as follows:
;
Wherein the method comprises the steps of Training data representing a u-th training sample,A sample tag representing the u-th training sample,The sample label representing the ith training sample is input to the discriminator, the probability value that the sample label of the ith training sample belongs to the sample label of the ith training sample is output,Training data representing the nth training sample is input to the fluorescence signal output by the generation model,Training data representing the ith training sample is input to the probability value that the fluorescence signal output by the generation model belongs to the sample label of the ith training sample.
Further, the fluorescence signal is divided into P wave bands according to the equal wavelength interval, and the average value of fluorescence intensities corresponding to all wavelengths in each wave band is calculated to be used as the input of a prediction model, wherein P is a self-defined parameter, and the calculation formula of the prediction model is as follows:
;
Where P represents the number of bands, con represents the algae concentration predicted value output by the prediction model, Represents the average of the fluorescence intensities corresponding to all wavelengths in the p-th band,Represents the p-1 power of the average of the fluorescence intensities corresponding to all wavelengths in the p-th band,And the weight parameter corresponding to the p-th wave band is represented.
Further, nonlinear fitting is carried out on the fluorescence signal and the algae concentration actual measurement value through a least square method to obtain a prediction model.
The invention provides a full algae automatic analysis system based on fluorescent quantum dots, which comprises:
The data acquisition module is used for acquiring hyperspectral data and water quality parameters of the experimental water sample;
the characteristic sequence generation module is used for preprocessing hyperspectral data and water quality parameters of the experimental water sample through a sliding window to generate a characteristic sequence;
the generation model construction module is used for acquiring a fluorescent signal of the experimental water sample through a fluorescent quantum detection method and training a generation model based on the characteristic sequence and the fluorescent signal;
The prediction model construction module is used for acquiring the algae concentration actual measurement value of the experimental water sample and training the prediction model based on the fluorescence signal and the algae concentration actual measurement value.
The method has the advantages that the hyperspectral data and the water quality parameters are utilized to generate the fluorescent signal which approximates to the fluorescent signal obtained by the fluorescent quantum detection method through the generation model, and the nonlinear mapping relation between the fluorescent signal and the algae concentration is established through the prediction model, so that the precision of measuring the algae concentration is improved, and the hyperspectral data and the water quality parameters can be acquired in real time, so that the real-time determination of the algae concentration is realized.
Drawings
FIG. 1 is a flow chart of a method for automatically analyzing full algae based on fluorescent quantum dots according to the present invention;
FIG. 2 is a flow chart of the invention for preprocessing hyperspectral data and water quality parameters of an experimental water sample through a sliding window to generate a characteristic sequence;
Fig. 3 is a schematic diagram of a full algae automatic analysis system based on fluorescent quantum dots according to the present invention.
In the figure, a data acquisition module 301, a feature sequence generation module 302, a generation model construction module 303 and a prediction model construction module 304 are shown.
Detailed Description
The subject matter described herein will now be discussed with reference to example embodiments. It is to be understood that these embodiments are merely discussed so that those skilled in the art may better understand and implement the subject matter described herein and that changes may be made in the function and arrangement of the elements discussed without departing from the scope of the disclosure herein. Various examples may omit, replace, or add various procedures or components as desired. In addition, features described with respect to some examples may be combined in other examples as well.
It is to be noted that unless otherwise defined, technical or scientific terms used in one or more embodiments of the present invention should be taken in a general sense as understood by one of ordinary skill in the art to which the present invention belongs. The use of the terms "first," "second," and the like in one or more embodiments of the present invention does not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or articles listed after the word are included in the word or "comprising", and equivalents thereof, but does not exclude other elements or articles "connected" or "connected", and the like, are not limited to physical or mechanical connections, but may include electrical connections, both direct and indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
As shown in fig. 1 to 3, the automatic analysis method of the whole algae based on the fluorescent quantum dots comprises the following steps:
step S101, hyperspectral data and water quality parameters of an experimental water sample are obtained;
The hyperspectral data are represented by spectral reflectivities at wavelengths of 350 nanometers to 850 nanometers;
The water quality parameters comprise chlorophyll a concentration, total phosphorus content, suspended matter concentration and turbidity;
Step S102, preprocessing hyperspectral data and water quality parameters of an experimental water sample through a sliding window to generate a characteristic sequence;
The characteristic sequence comprises K sequence units, and each sequence unit corresponds to a combination vector of a sliding window;
step S103, obtaining a fluorescence signal of an experimental water sample by a fluorescence quantum detection method, and training a generation model based on a characteristic sequence and the fluorescence signal;
generating an input of a model as a characteristic sequence and outputting as a fluorescent signal;
The fluorescence signal is represented by fluorescence intensity with a wavelength of M nm to N nm;
Step S104, obtaining an algae concentration actual measurement value of the experimental water sample, and training a prediction model based on the fluorescence signal and the algae concentration actual measurement value;
The input of the prediction model is a fluorescence signal, and the output is an algae concentration predicted value;
Step 105, generating a characteristic sequence of the water sample to be detected according to the step 101 and the step 102, inputting the characteristic sequence into a trained generation model, outputting a fluorescence signal, inputting the fluorescence signal into a trained prediction model, and outputting an algae concentration predicted value of the water sample to be detected.
It should be noted that, the chlorophyll a concentration has a strong absorption peak between 650 nm and 750 nm, the total phosphorus content has a strong absorption peak between 380 nm and 780 nm, the suspended matter concentration has a strong absorption peak between 400 nm and 700 nm, and the turbidity has a strong absorption peak between 400 nm and 700 nm, so that the hyperspectral data is represented by the spectral reflectance of 350 nm to 850 nm, which can indirectly reflect the water quality characteristics of the water sample, and can reduce the data calculation amount to a certain extent, thereby improving the calculation speed of the generation model and the prediction model.
In one embodiment of the invention, the hyperspectral data is obtained by a handheld or stationary hyperspectral instrument.
In one embodiment of the present invention, M and N are custom parameters, preferably, M is set to 500, N is set to 750, k is determined by the size and step size of the sliding window, and the calculation formula is as follows:
;
Wherein length represents the wavelength length of the hyperspectral data, the wavelength length of the hyperspectral data is 850 nanometers minus 350 nanometers, which is equal to 500 nanometers, size and step represent the size and step length of the sliding window respectively, wherein the size and step length of the sliding window are self-defined parameters, preferably, the size of the sliding window is set to 25, the step length of the sliding window is set to 25, and K is 20.
It should be noted that M and N are set according to emission wavelengths of different substances in algae in the fluorescent signal, for example, emission wavelengths of chlorophyll a in algae in the fluorescent signal are about 650 to 670 nm, emission wavelengths of carotene in algae in the fluorescent signal are about 500 to 600 nm, emission wavelengths of lutein in algae in the fluorescent signal are about 500 to 600 nm, emission wavelengths of phycoerythrin in algae in the fluorescent signal are about 560 to 570 nm, and emission wavelengths of phycocyanin in algae in the fluorescent signal are about 630 to 650 nm.
In one embodiment of the invention, absorbance values of the experimental water sample at the wavelengths of 663 nanometers and 635 nanometers are measured by a portable spectrophotometer, and chlorophyll a concentration of the experimental water sample is obtained by a chlorophyll a concentration calculation formula;
The chlorophyll a concentration was calculated as follows:
;
Wherein Chla represents the chlorophyll a concentration, Represents the absorbance value at a wavelength of 663 nm,The absorbance at 635 nm is expressed, vol is the volume, and Weight is the Weight.
In one embodiment of the present invention, as shown in fig. 2, the hyperspectral data and the water quality parameters of the experimental water sample are preprocessed through a sliding window to generate a characteristic sequence, which comprises the following steps:
Step S201, extracting characteristic values of hyperspectral data in a sliding window;
The characteristic values of the hyperspectral data comprise the maximum value, the minimum value, the average value, the standard deviation, the median, the skewness and the kurtosis of the spectral reflectivity of the hyperspectral data;
the calculation formula of the skewness Skewness of the spectral reflectance of the hyperspectral data is as follows:
;
The Kurtosis of the spectral reflectance of the hyperspectral data is calculated as follows:
;
where size represents the size of the sliding window, Representing the spectral reflectance corresponding to the ith wavelength of the hyperspectral data within the sliding window,AndRespectively representing the average value and standard deviation of the spectral reflectivity of the hyperspectral data in the sliding window;
Step S202, normalizing the characteristic values of hyperspectral data in a sliding window and water quality parameters by a minimum and maximum normalization method;
and step S203, splicing the characteristic value of the hyperspectral data in the normalized sliding window and the normalized water quality parameter to obtain a combination vector.
It should be noted that, the combination vector of one sliding window is expressed as:
;
Wherein the method comprises the steps of Respectively representing the characteristic values of the hyperspectral data in the normalized sliding window, namely the maximum value, the minimum value, the average value, the standard deviation, the median, the skewness and the kurtosis of the spectral reflectivity of the hyperspectral data,The normalized water quality parameters, namely chlorophyll a concentration, total phosphorus content, suspended matter concentration and turbidity, are respectively shown.
In one embodiment of the invention, the generated model comprises K hidden layers, the kth hidden layer inputs the kth sequence unit of the characteristic sequence and outputs an update vector, wherein K is more than or equal to 1 and less than or equal to K;
The update vector output by the Kth hidden layer is input to Q classifiers, and the classification spaces of the Q classifiers respectively represent the fluorescence intensities corresponding to the Q wavelengths of the fluorescence signal, wherein Q=M-N.
It should be noted that, the wavelength interval between Q wavelengths of the fluorescent signal may be set in a self-defined manner, for example, M is set to 500, n is set to 750, Q is 750-500=250, the number of classifiers is too large, the calculation complexity of generating the model is increased, the model convergence is not favored, the wavelength interval may be set to 5nm, Q is (750-500)/5=50, and the wavelength interval may be set to 10 nm, and Q is (750-500)/10=25.
In one embodiment of the invention, the computational formula for generating the model includes:
;
;
;
;
;
Wherein the method comprises the steps of AndRepresenting the update vectors output by the kth and kth-1 hidden layers respectively,Is assigned a value of 0 for each dimension value,Size and dimensions of (2)Is of the same size as the (a), when k-1=0, thenEach of the dimension values of (c) is 0,A kth sequence element representing a feature sequence of a kth hidden layer input,、、AndRepresenting the correlation coefficient, the concealment vector, the reset gate and the update gate of the kth concealment layer, respectively,、、、、、AndRespectively representing a first weight parameter, a second weight parameter, a third weight parameter, a fourth weight parameter, a fifth weight parameter, a sixth weight parameter and a seventh weight parameter of a kth hidden layer,、、AndRespectively representing a first bias parameter, a second bias parameter, a third bias parameter and a fourth bias parameter of the kth hidden layer,Representing point-wise multiplication, T representing a transpose operation, concat representing a splice operation, pile representing a stack operation, sigmoid representing a sigmoid activation function, tanh representing a hyperbolic tangent activation function, reLU representing a ReLU activation function.
It should be noted that, the weight parameter and the bias parameter of the generated model are both learnable parameters, for example, the kth sequence unit of the feature sequence input by the kth hidden layer, that is, the combined vector corresponding to the kth sequence unit, the size of the combined vector may be 1×11, the size of the update vector output by the kth and the kth-1 hidden layer may be designed to be 1×16, if the vector size obtained by splicing the combined vector and the update vector is 1×27, the sixth weight parameter and the seventh weight parameter of the kth hidden layer may be designed to be vectors with a size of 27×1, and the reset gate and the update gate corresponding to the multiplication of the two may be real numbers.
It should be noted that the number of the substrates,Vector size and (v) of (v)If the vector sizes of the two are the same, or the two cannot be stacked, for example, the two cannot be stacked into a matrix with the size of 2×64, then the third weight parameter can be designed into a matrix with the size of 64×1, the matrix with the size of 2×64 is multiplied by the matrix with the size of 64×1 to obtain a matrix with the size of 2×1, then the transpose operation is performed to obtain a matrix with the size of 1×2, the fourth weight parameter can be designed into a matrix with the size of 2×1, and the correlation coefficient obtained by multiplying the two is a real number.
In one embodiment of the present invention, the feature sequences generated in steps S101 to S102 are repeated as training data for training the training samples of the generated model, the fluorescent signal obtained by the fluorescence quantum detection method is used as a sample label for training the generated model, a discriminator is connected in the training process of the generated model, and the value output by the discriminator represents the probability value that the fluorescent signal output by the generated model belongs to the sample label of the training samples;
Loss function of discriminator The calculation formula of (2) is as follows:
;
Wherein the method comprises the steps of Training data representing a u-th training sample,A sample tag representing the u-th training sample,The sample label representing the ith training sample is input to the discriminator, the probability value that the sample label of the ith training sample belongs to the sample label of the ith training sample is output,Training data representing the nth training sample is input to the fluorescence signal output by the generation model,Training data representing the ith training sample is input to the probability value that the fluorescence signal output by the generation model belongs to the sample label of the ith training sample.
In one embodiment of the present invention, the fluorescence signal is divided into P bands according to the equal wavelength interval, and an average value of fluorescence intensities corresponding to all wavelengths in each band is calculated as an input of a prediction model, where P is a custom parameter, preferably, P is set to 10, and a calculation formula of the prediction model is as follows:
;
Where P represents the number of bands, con represents the algae concentration predicted value output by the prediction model, Represents the average of the fluorescence intensities corresponding to all wavelengths in the p-th band,Represents the p-1 power of the average of the fluorescence intensities corresponding to all wavelengths in the p-th band,And the weight parameter corresponding to the p-th wave band is represented.
In one embodiment of the invention, the predictive model is obtained by nonlinear fitting of the fluorescence signal and the algae concentration actual measurement by least squares.
In one embodiment of the present invention, as shown in fig. 3, the present invention provides a full algae automatic analysis system based on fluorescent quantum dots, comprising:
the data acquisition module 301 is used for acquiring hyperspectral data and water quality parameters of an experimental water sample;
the characteristic sequence generation module 302 is used for preprocessing hyperspectral data and water quality parameters of the experimental water sample through a sliding window to generate a characteristic sequence;
The generation model construction module 303 is used for acquiring a fluorescence signal of the experimental water sample through a fluorescence quantum detection method and training a generation model based on the characteristic sequence and the fluorescence signal;
The prediction model construction module 304 is configured to obtain an actual algae concentration value of the experimental water sample, and train the prediction model based on the fluorescence signal and the actual algae concentration value.
The embodiment has been described above with reference to the embodiment, but the embodiment is not limited to the above-described specific implementation, which is only illustrative and not restrictive, and many forms can be made by those of ordinary skill in the art, given the benefit of this disclosure, are within the scope of this embodiment.
Claims (4)
1. The automatic analysis method of the full algae based on the fluorescent quantum dots is characterized by comprising the following steps of:
step S101, hyperspectral data and water quality parameters of an experimental water sample are obtained;
The hyperspectral data are represented by spectral reflectivities at wavelengths of 350 nanometers to 850 nanometers;
The water quality parameters comprise chlorophyll a concentration, total phosphorus content, suspended matter concentration and turbidity;
Step S102, preprocessing hyperspectral data and water quality parameters of an experimental water sample through a sliding window to generate a characteristic sequence;
The characteristic sequence comprises K sequence units, and each sequence unit corresponds to a combination vector of a sliding window;
step S103, obtaining a fluorescence signal of an experimental water sample by a fluorescence quantum detection method, and training a generation model based on a characteristic sequence and the fluorescence signal;
generating an input of a model as a characteristic sequence and outputting as a fluorescent signal;
The fluorescence signal is represented by fluorescence intensity with a wavelength of M nm to N nm;
Step S104, obtaining an algae concentration actual measurement value of the experimental water sample, and training a prediction model based on the fluorescence signal and the algae concentration actual measurement value;
The input of the prediction model is a fluorescence signal, and the output is an algae concentration predicted value;
step 105, generating a characteristic sequence of the water sample to be detected according to the step 101 and the step 102, inputting the characteristic sequence into a trained generation model, outputting a fluorescence signal, inputting the fluorescence signal into a trained prediction model, and outputting an algae concentration prediction value of the water sample to be detected;
m and N are self-defined parameters, K is determined by the size and step length of the sliding window, and the calculation formula is as follows:
;
wherein length represents the wavelength length of the hyperspectral data, the wavelength length of the hyperspectral data is 850 nanometers minus 350 nanometers, which is equal to 500 nanometers, size and step represent the size and step length of the sliding window respectively, wherein the size and step length of the sliding window are self-defined parameters;
Preprocessing hyperspectral data and water quality parameters of an experimental water sample through a sliding window to generate a characteristic sequence, wherein the method comprises the following steps of:
Step S201, extracting characteristic values of hyperspectral data in a sliding window;
The characteristic values of the hyperspectral data comprise the maximum value, the minimum value, the average value, the standard deviation, the median, the skewness and the kurtosis of the spectral reflectivity of the hyperspectral data;
the calculation formula of the skewness Skewness of the spectral reflectance of the hyperspectral data is as follows:
;
The Kurtosis of the spectral reflectance of the hyperspectral data is calculated as follows:
;
where size represents the size of the sliding window, Representing the spectral reflectance corresponding to the ith wavelength of the hyperspectral data within the sliding window,AndRespectively representing the average value and standard deviation of the spectral reflectivity of the hyperspectral data in the sliding window;
Step S202, normalizing the characteristic values of hyperspectral data in a sliding window and water quality parameters by a minimum and maximum normalization method;
Step S203, splicing the characteristic value of the hyperspectral data in the normalized sliding window and the normalized water quality parameter to obtain a combination vector;
The generating model comprises K hidden layers, the kth hidden layer inputs the kth sequence unit of the characteristic sequence and outputs an updating vector, wherein K is more than or equal to 1 and less than or equal to K;
The update vector output by the Kth hidden layer is input to Q classifiers, and the classification spaces of the Q classifiers respectively represent the fluorescence intensities corresponding to the Q wavelengths of the fluorescence signals, wherein Q=M-N;
The calculation formula for generating the model comprises:
;
;
;
;
;
Wherein the method comprises the steps of AndRepresenting the update vectors output by the kth and kth-1 hidden layers respectively,Is assigned a value of 0 for each dimension value,Size and dimensions of (2)Is of the same size as the (a), when k-1=0, thenEach of the dimension values of (c) is 0,A kth sequence element representing a feature sequence of a kth hidden layer input,、、AndRepresenting the correlation coefficient, the concealment vector, the reset gate and the update gate of the kth concealment layer, respectively,、、、、、AndRespectively representing a first weight parameter, a second weight parameter, a third weight parameter, a fourth weight parameter, a fifth weight parameter, a sixth weight parameter and a seventh weight parameter of a kth hidden layer,、、AndRespectively representing a first bias parameter, a second bias parameter, a third bias parameter and a fourth bias parameter of the kth hidden layer,Representing point-wise multiplication, T representing a transpose operation, concat representing a splice operation, pile representing a stack operation, sigmoid representing a sigmoid activation function, tanh representing a hyperbolic tangent activation function, reLU representing a ReLU activation function;
Repeating the characteristic sequences generated in the steps S101 to S102 to serve as training data of training samples for training the generated model, wherein fluorescent signals obtained through a fluorescent quantum detection method serve as sample labels for training the generated model, a discriminator is connected in the training process of the generated model, and the value output by the discriminator represents the probability value of the fluorescent signals output by the generated model belonging to the sample labels of the training samples;
Loss function of discriminator The calculation formula of (2) is as follows:
;
Wherein the method comprises the steps of Training data representing a u-th training sample,A sample tag representing the u-th training sample,The sample label representing the ith training sample is input to the discriminator, the probability value that the sample label of the ith training sample belongs to the sample label of the ith training sample is output,Training data representing the nth training sample is input to the fluorescence signal output by the generation model,Training data representing a nth training sample is input to a probability value that a fluorescence signal output by the generation model belongs to a sample label of the nth training sample;
Dividing a fluorescent signal into P wave bands according to equal wavelength intervals, and calculating an average value of fluorescent intensities corresponding to all wavelengths in each wave band as an input of a prediction model, wherein P is a self-defined parameter, and the calculation formula of the prediction model is as follows:
;
Where P represents the number of bands, con represents the algae concentration predicted value output by the prediction model, Represents the average of the fluorescence intensities corresponding to all wavelengths in the p-th band,Represents the p-1 power of the average of the fluorescence intensities corresponding to all wavelengths in the p-th band,And the weight parameter corresponding to the p-th wave band is represented.
2. The automatic analysis method of full algae based on fluorescent quantum dots according to claim 1, wherein absorbance values of the experimental water sample at wavelengths of 663 nm and 635 nm are measured by a portable spectrophotometer, and chlorophyll a concentration of the experimental water sample is obtained by a chlorophyll a concentration calculation formula, wherein the chlorophyll a concentration calculation formula is as follows:
;
Wherein Chla represents the chlorophyll a concentration, Represents the absorbance value at a wavelength of 663 nm,The absorbance at 635 nm is expressed, vol is the volume, and Weight is the Weight.
3. The automatic analysis method of full algae based on fluorescent quantum dots according to claim 1, wherein the prediction model is obtained by nonlinear fitting of the fluorescence signal and the algae concentration actual measurement value by a least square method.
4. A fluorescent quantum dot-based automatic analysis system for total algae, wherein the method for automatically analyzing total algae based on fluorescent quantum dots according to any one of claims 1 to 3 is performed, comprising:
The data acquisition module is used for acquiring hyperspectral data and water quality parameters of the experimental water sample;
the characteristic sequence generation module is used for preprocessing hyperspectral data and water quality parameters of the experimental water sample through a sliding window to generate a characteristic sequence;
the generation model construction module is used for acquiring a fluorescent signal of the experimental water sample through a fluorescent quantum detection method and training a generation model based on the characteristic sequence and the fluorescent signal;
The prediction model construction module is used for acquiring the algae concentration actual measurement value of the experimental water sample and training the prediction model based on the fluorescence signal and the algae concentration actual measurement value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411152399.5A CN118655116B (en) | 2024-08-21 | 2024-08-21 | Automatic analysis method for full algae based on fluorescent quantum dots |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411152399.5A CN118655116B (en) | 2024-08-21 | 2024-08-21 | Automatic analysis method for full algae based on fluorescent quantum dots |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118655116A CN118655116A (en) | 2024-09-17 |
| CN118655116B true CN118655116B (en) | 2024-12-13 |
Family
ID=92705870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411152399.5A Active CN118655116B (en) | 2024-08-21 | 2024-08-21 | Automatic analysis method for full algae based on fluorescent quantum dots |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118655116B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115270951A (en) * | 2022-07-22 | 2022-11-01 | 浙江树人学院 | Self-adaptive dissolved organic carbon online detection method based on multi-source spectrum fusion |
| CN117723497A (en) * | 2024-02-08 | 2024-03-19 | 上海阿夸斯科技有限公司 | Intelligent follow-up water quality monitoring system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114112941A (en) * | 2021-12-14 | 2022-03-01 | 江苏省地质勘查技术院 | Aviation hyperspectral water eutrophication evaluation method based on support vector regression |
| CN117470776A (en) * | 2023-10-31 | 2024-01-30 | 山东省城市供排水水质监测中心 | Hyperspectral remote sensing water quality chlorophyll a inversion method based on SSA-XGBOOST algorithm |
-
2024
- 2024-08-21 CN CN202411152399.5A patent/CN118655116B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115270951A (en) * | 2022-07-22 | 2022-11-01 | 浙江树人学院 | Self-adaptive dissolved organic carbon online detection method based on multi-source spectrum fusion |
| CN117723497A (en) * | 2024-02-08 | 2024-03-19 | 上海阿夸斯科技有限公司 | Intelligent follow-up water quality monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118655116A (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cheng et al. | Deriving leaf mass per area (LMA) from foliar reflectance across a variety of plant species using continuous wavelet analysis | |
| Bendel et al. | Evaluating the suitability of hyper-and multispectral imaging to detect foliar symptoms of the grapevine trunk disease Esca in vineyards | |
| CN112712108A (en) | Raman spectrum multivariate data analysis method | |
| CN117723497B (en) | Intelligent follow-up water quality monitoring system | |
| Harrison et al. | Improved estimates of phytoplankton community composition based on in situ spectral fluorescence: use of ordination and field-derived norm spectra for the bbe FluoroProbe | |
| CN109187443A (en) | Water body bacterial micro-organism based on multi-wavelength transmitted spectrum accurately identifies method | |
| CN115424006A (en) | Multi-source multi-level data fusion method applied to crop phenotypic parameter inversion | |
| CN117470776A (en) | Hyperspectral remote sensing water quality chlorophyll a inversion method based on SSA-XGBOOST algorithm | |
| EP2366991A1 (en) | Probabilistic scoring for components of a mixture | |
| CN115565028A (en) | Transformer oil aging detection method, detection system, electronic equipment and readable storage medium | |
| CN106126879B (en) | A kind of soil near-infrared spectrum analysis prediction technique based on rarefaction representation technology | |
| CN118655116B (en) | Automatic analysis method for full algae based on fluorescent quantum dots | |
| CN116992293B (en) | Intelligent data processing method for chemiluminescent instrument | |
| CN119198596A (en) | A method, device, equipment, medium and product for detecting surface water quality parameters | |
| CN110887798B (en) | Nonlinear full-spectrum water turbidity quantitative analysis method based on extreme random tree | |
| CN117556245B (en) | Method for detecting filtered impurities in tetramethylammonium hydroxide production | |
| CN113376107A (en) | Water quality monitoring system and method based on cloud platform | |
| CN119025927A (en) | Rapid water quality inversion method, device, equipment and storage medium | |
| CN106383103A (en) | Time-resolved oil fluorescence identification method based on main component analysis | |
| Hou et al. | Rapid identification of rice seed based on inverse Fourier transform of laser-induced breakdown spectroscopy | |
| CN115512144A (en) | Automatic XRF spectrogram classification method based on convolution self-encoder | |
| CN107179297A (en) | A kind of intelligent category of annatto authentication method and its platform | |
| Avhad et al. | Soil Nutrients Analysis Techniques and Crop/Fertilizers Prediction—A Review | |
| Lounis et al. | Hybridisation of fuzzy systems and genetic algorithms for water quality characterisation using remote sensing data | |
| Liu et al. | Quantification of brown tide algae using EGAF coding and DFF feature fusion based on LED-induced fluorescence spectroscopy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |