Measuring method and device, electronic device and storage medium

Abstract

The present disclosure relates to a measurement method and device, an electronic device and a storage medium, and belongs to the field of water environment detection. The method includes: obtaining a spectral signal of a sample and a true value of a sediment index, the spectral signal including: a scattering signal, and / or a transmission signal, and / or a reflection signal; for the sample, determining at least one target factor affecting the sediment index; using the spectral signal, constructing a spectral feature characterizing each target factor, and a single sample corresponds to multiple spectral features; using the spectral feature and the true value of the sediment index, constructing first data; based on the first data, determining a sediment index calculation model, the sediment index calculation model is used to calculate the sediment index of the sample to be tested. This method can be used as a generalized method for determining sediment indicators in samples in the full time and space domain, and can improve the accuracy and efficiency of the determined sediment indicators.

Description

Technical Field

[0001] The present disclosure relates to the field of water environment detection, and in particular to a measurement method and device, an electronic device and a storage medium. Background Art

[0002] The sediment content of water bodies is one of the important hydrological parameters. Monitoring the sediment content of water bodies is of great significance to infrastructure construction, water source protection, soil erosion control and many other aspects.

[0003] Existing measurement methods mainly include: (1) Sampling analysis: by collecting samples and then analyzing them in the laboratory. Although this method is accurate, the sampling frequency is low, the cost is high, and it cannot provide real-time data. (2) Optical instruments: such as turbidity meters, which can quickly measure the sediment content in water bodies, but the accuracy is poor in water bodies with high turbidity or containing other suspended matter. (3) Acoustic instruments: such as ultrasonic sediment meters, which estimate the sediment content by the propagation characteristics of sound waves in water bodies, but are greatly affected by factors such as water temperature and salinity.

[0004] Regardless of which of the above methods is used, they are only applicable to certain scenarios and cannot be used in all scenarios, and cannot achieve the measurement of water sediment content in the full time and space domain.

[0005] Therefore, there is an urgent need for a generalized method to measure the sediment content in water bodies in the entire time and space domain. Summary of the invention

[0006] In view of this, the present disclosure proposes a sediment index measurement solution.

[0007] According to one aspect of the present disclosure, a measurement method is provided for measuring sediment index, comprising: obtaining a spectral signal of a sample and a true value of a sediment index, wherein the spectral signal comprises: a scattering signal, and / or a transmission signal, and / or a reflection signal; determining at least one target factor affecting the sediment index for the sample; constructing a spectral feature characterizing each target factor using the spectral signal, wherein a single sample corresponds to multiple spectral features; constructing first data using the spectral feature and the true value of the sediment index; determining a sediment index calculation model based on the first data, wherein the sediment index calculation model is used to calculate the sediment index of the sample to be measured; wherein constructing the spectral feature characterizing each target factor using the spectral signal comprises:

[0008] Determine at least two first wavelengths of a spectrum that responds to a target factor, and first spectrum signals corresponding to the first wavelengths, wherein a difference between the first spectrum signals corresponding to the at least two first wavelengths satisfies a preset intensity difference condition;

[0009] Processing the first spectral signal to eliminate background signals or enhance the sensitivity of the signal to the target factor, and obtaining at least one candidate feature;

[0010] The spectral features of the target factor are screened out from the candidate features.

[0011] In a possible implementation, the spectral signal also includes: a fluorescence signal; and / or the sediment index includes: at least one of sediment content, average sediment particle size, and sediment composition. In a possible implementation, determining at least two first wavelengths of the spectrum that responds to the target factor includes: determining the first wavelength according to the response intensity of the spectrum to the target factor; processing the first spectral signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor, and obtaining at least one candidate feature includes: taking the ratio of the response intensity corresponding to the two first wavelengths as the candidate feature.

[0012] In a possible implementation, the response intensity includes signal intensity, the target factor includes target particle size, and the determining of the first wavelength according to the response intensity of the spectrum to the target factor includes: determining multiple target particle sizes based on the sediment particle size distribution in the sample; among the wavelengths of the spectrum that responds to a single target particle size, determining a wavelength pair whose signal intensity difference is greater than a first intensity threshold as the first wavelength; and / or, the response intensity includes signal intensity, the target factor includes target component, and the determining of the first wavelength according to the response intensity of the spectrum to the target factor includes: obtaining water quality information of the sample, The water quality information includes the target component; in the spectrum that responds to the single target component, a target band is determined; a wavelength pair within the target band whose signal intensity difference is greater than a second intensity threshold is determined as a first wavelength; and / or the response intensity includes signal intensity, the target factor includes the target component ratio, and the first wavelength is determined based on the response intensity of the spectrum to the target factor, comprising: obtaining water quality information of the sample, the water quality information includes the target component ratio; in the wavelength of the spectrum that responds to the single target component ratio, a wavelength pair whose signal intensity difference is greater than a third intensity threshold is determined as the first wavelength.

[0013] In a possible implementation, screening out the spectral feature of the target factor from the candidate features includes: determining the correlation between each of the candidate features and the true value of the sediment index; and determining the spectral feature of the target factor based on the correlation.

[0014] In a possible implementation, the first data includes: training data and test data, and determining the sediment index calculation model based on the first data includes: obtaining multiple initial models; determining multiple first models and first calculated sediment indexes corresponding to each of the training data based on the multiple training data and the initial models, wherein a single training data corresponds to a true value of a sediment index and a first calculated sediment index; screening the multiple first models based on the true values ​​of the sediment indexes corresponding to the multiple training data and the first calculated sediment index to obtain at least two second models; and using the test data to screen the at least two second models to obtain the sediment index calculation model.

[0015] In a possible implementation, the method of using the test data to screen the at least two second models to obtain the sediment index calculation model includes: inputting the test data into the second model to obtain a second sediment calculation index corresponding to the test data, wherein a single test data corresponds to a true value of a sediment index and a second calculated sediment index; determining at least one model evaluation index based on the difference between the true value of the sediment index of the test data and the second calculated sediment index; and determining the sediment index calculation model from the second model based on the model evaluation index.

[0016] According to another aspect of the present disclosure, a measuring device is provided for measuring sediment indicators, comprising:

[0017] A sample spectral signal and sediment index acquisition unit is used to acquire a spectral signal of the sample and a true value of the sediment index, wherein the spectral signal includes: a scattering signal, and / or a transmission signal, and / or a reflection signal;

[0018] A target factor determination unit, used to determine at least one target factor affecting sediment indicators for the sample;

[0019] A spectral feature determination unit, used to construct spectral features characterizing each target factor using the spectral signal, wherein a single sample corresponds to multiple spectral features;

[0020] A first data construction unit, configured to construct first data using the spectral feature and the true value of the sediment index;

[0021] A sediment index calculation model determination unit is used to determine a sediment index calculation model based on the first data, wherein the sediment index calculation model is used to calculate the sediment index of the sample to be tested; wherein the spectral feature determination unit includes:

[0022] a first wavelength and first spectrum signal determination unit, used to determine at least two first wavelengths of a spectrum that responds to a target factor, and first spectrum signals corresponding to the first wavelengths, wherein a difference between the first spectrum signals corresponding to the at least two first wavelengths satisfies a preset intensity difference condition;

[0023] a candidate feature determination unit, configured to process the first spectral signal to eliminate background signals or enhance the sensitivity of the signal to the target factor, and obtain at least one candidate feature;

[0024] The spectral feature determination subunit is used to screen out the spectral feature of the target factor from the candidate features.

[0025] In a possible implementation, the spectral signal further includes: a fluorescence signal; and / or,

[0026] The sediment index includes at least one of sediment content, average sediment particle size, and sediment composition.

[0027] In a possible implementation manner, the first wavelength and first spectrum signal determining unit is further used to:

[0028] determining the first wavelength according to the response intensity of the spectrum to the target factor;

[0029] The processing of the first spectral signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor and obtain at least one candidate feature includes:

[0030] The ratio of the response intensities corresponding to the two first wavelengths is used as the candidate feature.

[0031] In a possible implementation, the response intensity includes signal intensity, the target factor includes target particle size, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes:

[0032] Determining a plurality of target particle sizes based on the particle size distribution of sediment in the sample;

[0033] Among the wavelengths of the spectrum that responds to a single target particle size, a wavelength pair whose signal intensity difference is greater than a first intensity threshold is determined as the first wavelength; and / or,

[0034] The response intensity includes signal intensity, the target factor includes a target component, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes:

[0035] Acquiring water quality information of the sample, wherein the water quality information includes the target component;

[0036] determining a target wavelength band in the spectrum that responds to a single target component;

[0037] Determine a wavelength pair within the target wavelength band whose signal strength difference is greater than a second strength threshold as a first wavelength; and / or,

[0038] The response intensity includes signal intensity, the target factor includes a target component ratio, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes:

[0039] Obtaining water quality information of the sample, wherein the water quality information includes the proportion of the target component;

[0040] Among the wavelengths of the response spectrum for a single target component ratio, a wavelength pair whose signal intensity difference is greater than a third intensity threshold is determined as the first wavelength.

[0041] In a possible implementation, the spectral feature determination subunit is further used to:

[0042] Determining the correlation between each of the candidate features and the true value of the sediment index;

[0043] According to the correlation, the spectral characteristics of the target factor are determined.

[0044] In a possible implementation, the first data includes: training data and test data, and the sediment index calculation model determination unit is further used to:

[0045] Obtain multiple initial models;

[0046] Determine a plurality of first models and a first calculated sediment index corresponding to each of the training data according to the plurality of training data and the initial model, wherein a single training data corresponds to a true value of a sediment index and a first calculated sediment index;

[0047] Based on the true values ​​of the sediment indicators corresponding to the plurality of training data and the first calculated sediment indicator, the plurality of first models are screened to obtain at least two second models;

[0048] The test data is used to screen the at least two second models to obtain the sediment index calculation model.

[0049] In a possible implementation, the using the test data to screen the at least two second models to obtain the sediment index calculation model includes:

[0050] Inputting the test data into the second model to obtain a second sediment calculation index corresponding to the test data, wherein a single test data corresponds to a true value of a sediment index and a second calculated sediment index;

[0051] Determining at least one model evaluation index based on a difference between a true value of a sediment index of the test data and a second calculated sediment index;

[0052] Based on the model evaluation index, the sediment index calculation model is determined from the second model.

[0053] According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0054] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions implement the above method when executed by a processor.

[0055] According to another aspect of the present disclosure, a computer program product is provided, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.

[0056] In the embodiment of the present disclosure, the first data can be used to construct (verify), train, and test the model, and the model that meets the pre-set conditions can be used as the sediment index calculation model. The conditions here can include conditions of multiple dimensions, such as: accuracy dimension, stability dimension, generalization ability and other dimensions. The first data contains one or more spectral features that characterize the target factor. Therefore, the process of determining the sediment index calculation model is actually the process of finding the mapping relationship between the spectral features and the sediment index. In this way, the sediment index calculation model in the embodiment of the present disclosure can only rely on the spectral signal (spectral feature) to accurately determine the sediment index without considering other factors, thereby improving the accuracy and efficiency of the determined sediment index. In addition, the method of the embodiment of the present disclosure is not limited by the scene and the time and space domain, and can be used as a generalized method for determining the sediment index in the sample in the full time and space domain, thereby improving the applicability of the method of the present disclosure.

[0057] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0059] Figure 1A flow chart of a measurement method provided in an embodiment of the present disclosure.

[0060] Figure 2 A schematic diagram of the calculation effect of the second model provided in an embodiment of the present disclosure on the training data.

[0061] Figure 3 A schematic diagram of the calculation effect of the second model provided in an embodiment of the present disclosure on test data.

[0062] Figure 4 A schematic diagram of the structure of a measuring device provided in an embodiment of the present disclosure.

[0063] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0064] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0065] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0066] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0067] Monitoring of water sediment indicators is of great significance to infrastructure construction, water source protection, soil and water loss control and many other aspects.

[0068] Usually, the method of measuring sediment indicators is only applicable to certain scenarios and cannot be used in all scenarios. For example: for sediment content, water bodies with high sediment content, especially tens to hundreds of kilograms, the interaction between solutes, absorbance is no longer proportional to concentration, Beer-Lambert law is no longer applicable, and the commonly used sediment content measurement method is no longer applicable. Another example: the commonly used sediment content measurement method cannot be applied to water bodies with large fluctuations in water quality at all times, or cannot be applied to water bodies with large fluctuations in water quality throughout the entire basin. In particular, it is even more difficult to measure the sediment content of water bodies with irregular changes in water quality. Therefore, there is an urgent need for a generalized method for measuring sediment indicators in water bodies in the full time and space domain.

[0069] Figure 1 The following is a flow chart of the measurement method provided by the embodiment of the present disclosure. Figure 1 As shown, the method includes:

[0070] S11, obtaining a spectral signal of the sample and a true value of a sediment index, wherein the spectral signal includes: a scattering signal, and / or a transmission signal, and / or a reflection signal.

[0071] In the embodiments of the present disclosure, the sample may be a sample containing sediment obtained from a natural water body, or may be an artificially configured sample. Furthermore, the true value of the sediment index of the sample may be a relatively accurate sediment index measured by relevant technologies, such as a drying method, a capacitance method, a laser method, etc. The sediment index includes at least one of sediment content, average sediment particle size, and sediment composition.

[0072] When using the measuring device, the spectral signal may include one or more types of signals, such as a scattered signal, a transmission signal, and a reflection signal. In addition, the spectral signal may also include a fluorescence signal.

[0073] The measuring device includes: a light source and a spectrum sensor, wherein the light source is a wide spectrum light source emitting light with a wavelength ranging from visible light to near infrared light, the spectrum sensor is used to receive light emitted by the light source and after passing through a sample containing sediment, and output spectrum information, the spectrum sensor includes: a quantum dot filter and a detector, the quantum dot filter is located on the receiving side of the detector. The measuring device can be an in-situ contact measuring device or a remote sensing measuring device.

[0074] Optionally, the light source can emit light of any wavelength band within the range of 390nm-1000nm, for example, the light source can emit light of 400-800nm, 410-900nm, 450-1000nm, etc., which will not be described in detail here.

[0075] S12, determining at least one target factor affecting sediment indicators for the sample.

[0076] The target factor may be a core factor affecting the sediment index in the sample. The target factor may include: particle size, composition, component ratio, etc. This is only an exemplary description of the target factor, and factors that affect the sediment index that are not listed also belong to the target factors disclosed in this disclosure. For a single sample, the target factor may be one or more.

[0077] For example, when the composition of sample A is stable and the type is clear, the core factor affecting the sediment content of sample A may be the particle size. Therefore, when measuring the sediment content of sample A, the particle size can be used as a target factor.

[0078] For another example, when the composition of sample B is unknown, the core factors affecting the sediment content of sample A may be particle size, composition, and composition ratio. Therefore, when measuring the sediment content of sample B, particle size, composition, and composition ratio may all be used as target factors.

[0079] S13, using the spectral signal, constructing spectral features characterizing each target factor, and a single sample corresponds to multiple spectral features.

[0080] In the embodiments of the present disclosure, the same type of spectral signals can be used to construct spectral features that characterize target factors; different types of spectral signals can also be used to construct spectral features that characterize target factors. The spectrum that characterizes each target factor can be determined based on the spectral signal, and features can be extracted from each spectrum to obtain spectral features. In the case of a single target factor, features can also be directly extracted from the spectral signal to obtain spectral features that characterize the target factor. Spectral features can characterize the significant characteristics of the spectrum under the influence of a single target factor from the perspective of the single target factor.

[0081] S14, constructing first data using the spectral feature and the true value of the sediment index.

[0082] In the disclosed embodiment, at least one spectral feature and at least one true value of a sediment index corresponding to a single sample can be used as a sample data. The true value of the sediment index can be used as the true value of the sample data. The sample data can be divided according to a preset quantity ratio to construct training data, test data, and verification data. Alternatively, only training data and test data can be constructed.

[0083] S15, determining a sediment index calculation model based on the first data, wherein the sediment index calculation model is used to calculate the sediment index of the sample to be tested.

[0084] In the embodiment of the present disclosure, the first data can be used to construct (verify), train, and test the model, and the model that meets the pre-set conditions can be used as the sediment index calculation model. The conditions here can include conditions of multiple dimensions, such as: accuracy dimension, stability dimension, generalization ability and other dimensions. The first data contains one or more spectral features that characterize the target factor. Therefore, the process of determining the sediment index calculation model is actually the process of finding the mapping relationship between the spectral features and the sediment index. In this way, the sediment index calculation model in the embodiment of the present disclosure can only rely on the spectral signal (spectral feature) to accurately determine the sediment index without considering other factors, thereby improving the accuracy and efficiency of the determined sediment index. In addition, the method of the embodiment of the present disclosure is not limited by the scene and the time and space domain, and can be used as a generalized method for determining the sediment index in the sample in the full time and space domain, thereby improving the applicability of the method of the present disclosure.

[0085] In one possible implementation, the use of the spectral signal to construct a spectral feature characterizing each target factor includes: determining at least two first wavelengths of the spectrum that responds to the target factor, and a first spectral signal corresponding to the first wavelength, wherein the difference between the first spectral signals corresponding to the at least two first wavelengths satisfies a preset intensity difference condition; processing the first spectral signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor, and obtaining at least one candidate feature; and screening out the spectral feature of the target factor from the candidate features.

[0086] The first wavelength in the embodiment of the present disclosure may be a specific wavelength or waveband in the same spectrum that responds to the target factor. And the intensities of the first spectrum signals generated by the at least two first wavelengths in response to the target factor are significantly different. In the embodiment of the present disclosure, an intensity threshold may be set in advance. The preset intensity difference condition may include: the absolute value of the difference in signal intensity of the two first spectrum signals is not less than the intensity threshold. For another example, an intensity interval may be set in advance, and the preset intensity difference condition may include: the ratio of the signal intensity of the two first spectrum signals does not fall within the intensity interval. The spectrum that responds to the target factor may include: a scattered spectrum, a transmission spectrum, and a reflection spectrum.

[0087] In the disclosed embodiment, different channel combinations (the quantum dot filter has multiple filter channels), Fourier transform method, background subtraction method, spectral line fitting and analysis method, etc. can be used to process the first spectral signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor, and the processing result can be used as a candidate feature. The disclosed embodiment does not limit the specific processing method.

[0088] In the disclosed embodiment, a single first spectral signal can be used as a processing object to eliminate the background signal or enhance the sensitivity of the signal to the target factor, and obtain a single candidate feature; two or more first spectral signals can also be used as processing objects together to eliminate the background signal or enhance the sensitivity of the signal to the target factor, and obtain a single candidate feature. Then, the features with strong explanatory power for the change of sediment indicators are selected from the candidate features as the spectral features of the target factor.

[0089] In the embodiment of the present disclosure, the first spectral signal is selected from the spectrum that responds to the target factor, indicating that the first spectral signal also responds to the target factor. Moreover, the first spectral signal is processed to eliminate the background signal or enhance the sensitivity of the signal to the target factor, which can highlight the information related to the target factor carried in the first spectral signal and reduce the information unrelated to the target factor. And the processed spectral signal is used as a candidate feature. The spectral features of the target factor are further screened out from the candidate features. In this way, the spectral features of the target factor can be constructed from the spectrum that responds to the target factor. And the spectral feature can highlight the target factor and weaken other factors (non-target factors), so that the accuracy of the study on the correlation between the target factor and the sediment index is improved. Moreover, the spectral feature is a feature with strong explanatory power for the change in the amount of sediment index, that is, the spectral feature can accurately characterize the feature of the relationship between the target factor and the change in the sediment index. Therefore, the features of the embodiment of the present disclosure can improve the accuracy of the calculation of the sediment index.

[0090] In one possible implementation, determining at least two first wavelengths of a spectrum that responds to a target factor includes: determining the first wavelength based on the response intensity of the spectrum to the target factor; processing the first spectrum signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor, and obtaining at least one candidate feature, includes: taking the ratio of the response intensities corresponding to the two first wavelengths as the candidate feature.

[0091] The response intensity may be the absorbance, scattering rate, transmittance, or reflectivity of the target factor to light. The response intensity may be the spectral signal intensity. In the disclosed embodiment, two bands or wavelengths with large response intensity differences may be selected as the first wavelength. In the disclosed embodiment, two bands or wavelengths may be selected as the first wavelength in at least one or more specific bands.

[0092] For example, there are n bands that respond to the target factor, where n is a positive integer. The response strength increases as the target factor changes. The order of response strength increase is: band 1, band 2, ... band n. Band 1 and band n, the two bands with the largest difference in response strength, can be used as the first wavelength.

[0093] For another example, there are m bands that respond to the target factor, where m is a positive integer. However, due to the existence of interference factors, not all of the m bands respond to the target factor, and some of the bands respond to the interference factors. In order to reduce the negative impact of the interference factors on the determination of spectral characteristics, the band a that responds to the interference factors can be determined according to the response intensity corresponding to the interference factors. Band a is removed from the m bands to obtain the target band, and the first wavelength is determined in the target band.

[0094] The number of the first wavelengths may be at least two. The ratio of the response strength of the two first wavelengths to the target factor may be used as a candidate feature. For example, the selected first wavelengths may be combined in pairs, and the ratio of the response strength of the two first wavelengths to the target factor in each combination may be used as a candidate feature.

[0095] The spectrum can reflect the wavelength and the response intensity of each wavelength to the target factor, so the first wavelength can be accurately and conveniently determined by the response intensity. Using the response intensities corresponding to the two first wavelengths to calculate the ratio can effectively eliminate the background signal, and can eliminate non-target factors as background signals, so that the candidate features can prominently characterize the target factors. In this way, the efficiency and accuracy of determining the candidate features can be improved.

[0096] In one possible implementation, the response intensity includes signal intensity, the target factor includes target particle size, and determining the first wavelength based on the response intensity of the spectrum to the target factor includes: determining multiple target particle sizes based on the sediment particle size distribution in the sample; and determining, among the wavelengths of the spectrum that responds to a single target particle size, a wavelength pair whose signal intensity difference is greater than a first intensity threshold as the first wavelength.

[0097] And / or, the response intensity includes signal intensity, the target factor includes target component, and determining the first wavelength based on the response intensity of the spectrum to the target factor includes: obtaining water quality information of the sample, the water quality information including the target component; determining a target band in the spectrum that responds to a single target component; and determining a wavelength pair within the target band whose signal intensity difference is greater than a second intensity threshold as the first wavelength.

[0098] And / or, the response intensity includes signal intensity, the target factor includes the proportion of target components, and determining the first wavelength based on the response intensity of the spectrum to the target factor includes: obtaining water quality information of the sample, the water quality information including the proportion of the target component; among the wavelengths of the response spectrum for a single proportion of the target component, determining a wavelength pair whose signal intensity difference is greater than a third intensity threshold as the first wavelength.

[0099] The signal intensity may be a spectral signal intensity, that is, it may be a scattered spectrum signal intensity, or a transmitted spectrum signal intensity, or a reflected spectrum signal intensity.

[0100] Optionally, the above-mentioned spectrum signal may also include a fluorescence signal. The signal intensity may also be the fluorescence spectrum signal intensity.

[0101] A single target particle size can be a particle size value. As the particle size increases, the signal intensity becomes less and less sensitive to changes in wavelength (band). Therefore, the signal intensity corresponding to the wavelength (band) that responds to the target particle size is closer.

[0102] Thus, the particle size distribution of the sediment in the sample can be obtained, and the particle size greater than the content threshold is determined as the target particle size. In this way, multiple target particle sizes can be obtained.

[0103] In the spectrum that responds to a single target particle size, two first wavelengths whose signal intensity difference is greater than a first intensity threshold are determined, and the two first wavelengths are regarded as a wavelength pair. A single target particle size may correspond to multiple wavelength pairs.

[0104] In this way, with the particle size as a target factor, the accuracy of determining the first wavelength can be improved and the number of first wavelengths can be enriched.

[0105] In addition, the signal intensity may be a spectral signal intensity, that is, it may be a scattered spectrum signal intensity, or a transmitted spectrum signal intensity, or a reflected spectrum signal intensity.

[0106] The target component may include one or more components. However, not all components of the sample are related to the sediment index. For example, substance A contained in the sample is caused by upstream discharge, not by the sediment in the sample. Another example is that the algae in the sample are not related to the determination of the sediment index (such as sediment content).

[0107] Thus, the water quality information of the sample can be obtained, and the water quality information can include various components in the sample, including the target component. The target band can be determined in the spectrum that responds to the single target component.

[0108] For example, the band b responding to the target component may include the band c responding to the algae, and the band c can be removed from the band b to obtain the target band, and the first wavelengths whose signal intensity difference is greater than the second intensity threshold are determined in the target band, and the two first wavelengths are regarded as a wavelength pair. A single target component can correspond to multiple wavelength pairs.

[0109] In this way, when the component is taken as a target factor, the accuracy of determining the first wavelength can be improved and the number of first wavelengths can be enriched.

[0110] In addition, the signal intensity may be a spectral signal intensity, that is, it may be a scattered spectrum signal intensity, or a transmitted spectrum signal intensity, or a reflected spectrum signal intensity.

[0111] The target component ratio can be the content ratio of multiple target components. For example, the higher the sediment content, the higher the content of some components in the sample. For example, total suspended solids, phosphorus-containing salts, nitrogen-containing salts, etc. Conversely, the lower the sediment content, the lower the content of these components in the sample.

[0112] Therefore, when the content ratio of components related to sediment indicators in the sample is lower than the first content threshold or higher than the second content threshold, the water quality information of the sample can be obtained. The water quality information may include the content of various components in the sample, or the content ratio of various components, including the proportion of target components.

[0113] In the spectrum that responds to a single target component ratio, two first wavelengths whose signal intensity difference is greater than a first intensity threshold are determined, and the two first wavelengths are regarded as a wavelength pair. A single target component ratio may correspond to multiple wavelength pairs.

[0114] In this way, when the component ratio is taken as a target factor, the accuracy of determining the first wavelength can be improved and the number of first wavelengths can be enriched.

[0115] In a possible implementation, screening out the spectral feature of the target factor from the candidate features includes: determining the correlation between each of the candidate features and the true value of the sediment index; and determining the spectral feature of the target factor based on the correlation.

[0116] A single sample may correspond to multiple candidate features. In the disclosed embodiment, the correlation between the candidate features corresponding to a single sample and the true value of the sediment index of the single sample may be determined.

[0117] Exemplarily, the candidate features with the highest correlation may be selected as spectral features according to a preset first quantity.

[0118] For example, a model with high reliability, high accuracy and strong stability, such as a linear model, can be preset, and the candidate features with high correlation are added to the linear model one by one by the incremental method. When the accuracy of the model is no longer improved, or the improvement amplitude does not reach the amplitude threshold, the candidate features that have been added to the model are used as spectral features.

[0119] For example, the spectral features may be determined from the candidate features using methods such as feature importance evaluation or recursive feature elimination.

[0120] To facilitate understanding, formula (1) is used to explain the process of determining the correlation between each candidate feature and the true value of the sediment index of the sample.

[0121] (1)

[0122] in, is the pth sample The correlation coefficient between the candidate features and the true value of the sediment index of the pth sample; is the kth candidate feature of the pth sample, is the mean of the candidate features of the pth sample; is the true value of the sediment index of the pth sample, is the mean of the true values ​​of the sediment indicators of multiple samples including the pth sample.

[0123] In the disclosed embodiment, the correlation between the candidate features and the true value of the sediment index in the sample is used as an indicator for determining the spectral features, which is targeted for calculating the sediment index and can improve the accuracy and generalization ability of the sediment index calculation model.

[0124] In a possible implementation, the first data includes: training data and test data, and determining the sediment index calculation model based on the first data includes: obtaining multiple initial models; determining multiple first models and first calculated sediment indexes corresponding to each of the training data based on the multiple training data and the initial models, wherein a single training data corresponds to a true value of a sediment index and a first calculated sediment index; screening the multiple first models based on the true values ​​of the sediment indexes corresponding to the multiple training data and the first calculated sediment index to obtain at least two second models; and using the test data to screen the at least two second models to obtain the sediment index calculation model.

[0125] In the disclosed embodiment, multiple initial models may be set, and the initial models may include multiple unknown first parameters. The spectral features in the training data may be substituted into each initial model to determine the first parameters of each initial model, and the first calculated sediment index determined according to the spectral features using the initial model. For ease of description, the initial model in which the first parameter is determined may be named the first model. For a single first model, each training data will correspond to a first calculated sediment index and a true value of the sediment index.

[0126] Based on the principle of minimizing the numerical difference between the true value of the sediment index of the training data and the corresponding first calculated sediment index, the numerical difference between the true value of the sediment index and the first calculated sediment index can be determined for each first model, so as to select the first model with the smallest difference according to the preset second number as the second model. The value of the second number can be no less than 2.

[0127] Taking the sediment content as an example, Figure 2Schematic diagram of the calculation effect of the second model provided in the embodiment of the present disclosure on the training data. Two second models are determined using the training data. The second models are model 1 and model 2. As shown in the figure, the x-axis represents the true value of the sediment content of the sample (the measured sediment content), and the y-axis represents the first calculated sediment content of the sample determined using the second model (model prediction). We hope that the first sediment content calculated by the second model is as close as possible to the true value of the sediment content. Figure 2 In this way, Model 1 and Model 2 are selected and used as the second model.

[0128] The second model can be an initial model that has accurate calculation capabilities for training data in the scenario of calculating sediment content. However, it does not mean that the second model will also obtain accurate calculation results for other data. Then, it is necessary to use test data to further screen the second model. The test data and the training data can be completely different. The screening of the second model using test data can not only screen the model accuracy, but also the model stability and other performance. The second model with the best accuracy and stability among the second models is used as the sediment index calculation model.

[0129] In the disclosed embodiment, the training data is used to determine the parameters of the model, and a second model is selected that calculates the sediment content close to the true value of the sediment content. The test data is then used to further screen the second model for accuracy and performance to obtain a sediment index calculation model, thereby reducing the dependence of the sediment index calculation model on data and scenarios and improving accuracy and generalization ability.

[0130] In a possible implementation, the method of using the test data to screen the at least two second models to obtain the sediment index calculation model includes: inputting the test data into the second model to obtain a second sediment calculation index corresponding to the test data, wherein a single test data corresponds to a true value of a sediment index and a second calculated sediment index; determining at least one model evaluation index based on the difference between the true value of the sediment index of the test data and the second calculated sediment index; and determining the sediment index calculation model from the second model based on the model evaluation index.

[0131] In the embodiments of the present disclosure, the model evaluation index may be an index reflecting the accuracy of the model, such as accuracy, precision, F1 score, recall rate, mean square error, etc.; the model evaluation index may be an index reflecting the inherent defects of the model, such as systematic error, etc. The model evaluation index may be an index reflecting the model fitting situation, such as Chi Chi information criterion, etc. In a possible implementation, the evaluation index includes at least one of the following: determination coefficient, systematic error, random uncertainty.

[0132] For ease of understanding, formulas (2) and (3) are used to represent the process of determining systematic error and random uncertainty, respectively.

[0133] (2)

[0134] S e =2 [ 1 n-2 ∑ i=1 n ( Y Ai - Y Gi Y Gi ) 2 ] 1 / 2 (3)

[0135] in, represents the system error, is the random uncertainty; For samples The calculated value of sediment index is For samples The true value of the sediment index, n means there are n samples in total.

[0136] In one example, the second model includes model 1 and model 2. The evaluation indicators include: determination coefficient, systematic error, and random uncertainty. Among them, the smaller the value of the systematic error, the more accurate the model; the smaller the value of the random uncertainty, the more accurate the model; the larger the value of the determination coefficient, the better and more accurate the model fit. Moreover, the importance of random uncertainty and determination coefficient is higher than that of systematic error. For example, higher weights can be assigned to random uncertainty and determination coefficient. The performance comparison of model 1 and model 2 is shown in Table 1:

[0137] Table 1 Performance comparison of the second model

[0138]

[0139] As shown in Table 1, in terms of the systematic error dimension, model 1 is better than model 2. However, in terms of random uncertainty and determination coefficient, model 2 is better than model 1, and the two evaluation indicators of random uncertainty and determination coefficient are more important than the systematic error. In summary, model 2 can be determined as the sediment index calculation model.

[0140] In one example, the second model includes model 1 and model 2. The evaluation index includes: average relative deviation. Figure 3 A schematic diagram of the calculation effect of the second model provided in the embodiment of the present disclosure on the test data. As shown in the figure, the average relative deviation of Model 1 and Model 2 does not exceed 3%, which performs well. Among them, the average relative deviation of Model 2 is lower, so Model 2 can be determined as the sediment index calculation model.

[0141] In the disclosed embodiment, the second model is comprehensively evaluated using test data, and the model itself (fitting condition, inherent defects of the model) is further screened to improve the accuracy of the sediment index calculation model. Moreover, the second model is evaluated in dimensions other than accuracy, which improves the stability and universality of the sediment index calculation model.

[0142] Figure 4 This is a schematic diagram of the structure of the measuring device provided in the embodiment of the present disclosure. Figure 4 As shown, the device 20 comprises:

[0143] The sample spectrum signal and sediment index acquisition unit 21 is used to acquire the spectrum signal of the sample and the true value of the sediment index, wherein the spectrum signal includes: a scattering signal, and / or a transmission signal, and / or a reflection signal;

[0144] A target factor determination unit 22, used to determine at least one target factor affecting sediment indicators for the sample;

[0145] A spectral feature determination unit 23 is used to construct a spectral feature characterizing each target factor using the spectral signal, and a single sample corresponds to multiple spectral features;

[0146] A first data construction unit 24, configured to construct first data using the spectral feature and the true value of the sediment index;

[0147] The sediment index calculation model determination unit 25 is used to determine a sediment index calculation model based on the first data, wherein the sediment index calculation model is used to calculate the sediment index of the sample to be tested.

[0148] The spectral feature determination unit 23 comprises:

[0149] a first wavelength and first spectrum signal determination unit, used to determine at least two first wavelengths of a spectrum that responds to a target factor, and first spectrum signals corresponding to the first wavelengths, wherein a difference between the first spectrum signals corresponding to the at least two first wavelengths satisfies a preset intensity difference condition;

[0150] a candidate feature determination unit, configured to process the first spectral signal to eliminate background signals or enhance the sensitivity of the signal to the target factor, and obtain at least one candidate feature;

[0151] The spectral feature determination subunit is used to screen out the spectral feature of the target factor from the candidate features.

[0152] In a possible implementation, the spectral signal further includes: a fluorescence signal; and / or,

[0153] The sediment index includes at least one of sediment content, average sediment particle size, and sediment composition.

[0154] In a possible implementation manner, the first wavelength and first spectrum signal determining unit is further used to:

[0155] determining the first wavelength according to the response intensity of the spectrum to the target factor;

[0156] The processing of the first spectral signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor and obtain at least one candidate feature includes:

[0157] The ratio of the response intensities corresponding to the two first wavelengths is used as the candidate feature.

[0158] In a possible implementation, the response intensity includes signal intensity, the target factor includes target particle size, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes:

[0159] Determining a plurality of target particle sizes based on the particle size distribution of sediment in the sample;

[0160] Among the wavelengths of the spectrum that responds to a single target particle size, a wavelength pair whose signal intensity difference is greater than a first intensity threshold is determined as the first wavelength; and / or,

[0161] The response intensity includes signal intensity, the target factor includes a target component, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes:

[0162] Acquiring water quality information of the sample, wherein the water quality information includes the target component;

[0163] determining a target wavelength band in the spectrum that responds to a single target component;

[0164] Determine a wavelength pair within the target wavelength band whose signal strength difference is greater than a second strength threshold as a first wavelength; and / or,

[0165] The response intensity includes signal intensity, the target factor includes a target component ratio, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes:

[0166] Obtaining water quality information of the sample, wherein the water quality information includes the proportion of the target component;

[0167] Among the wavelengths of the response spectrum for a single target component ratio, a wavelength pair whose signal intensity difference is greater than a third intensity threshold is determined as the first wavelength.

[0168] In a possible implementation, the spectral feature determination subunit is further used to:

[0169] Determining the correlation between each of the candidate features and the true value of the sediment index;

[0170] According to the correlation, the spectral characteristics of the target factor are determined.

[0171] In a possible implementation, the first data includes: training data and test data, and the sediment index calculation model determination unit 25 is further used to:

[0172] Obtain multiple initial models;

[0173] Determine a plurality of first models and a first calculated sediment index corresponding to each of the training data according to the plurality of training data and the initial model, wherein a single training data corresponds to a true value of a sediment index and a first calculated sediment index;

[0174] Based on the true values ​​of the sediment indicators corresponding to the plurality of training data and the first calculated sediment indicator, the plurality of first models are screened to obtain at least two second models;

[0175] The test data is used to screen the at least two second models to obtain the sediment index calculation model.

[0176] In a possible implementation, the using the test data to screen the at least two second models to obtain the sediment index calculation model includes:

[0177] Inputting the test data into the second model to obtain a second sediment calculation index corresponding to the test data, wherein a single test data corresponds to a true value of a sediment index and a second calculated sediment index;

[0178] Determining at least one model evaluation index based on a difference between a true value of a sediment index of the test data and a second calculated sediment index;

[0179] Based on the model evaluation index, the sediment index calculation model is determined from the second model.

[0180] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0181] The embodiment of the present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and the computer program instructions implement the above method when executed by a processor. The computer-readable storage medium can be a volatile or non-volatile computer-readable storage medium.

[0182] An embodiment of the present disclosure further proposes an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0183] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.

[0184] Figure 5 The electronic device 1900 is a schematic diagram of the structure of the electronic device provided in the embodiment of the present disclosure. For example, the electronic device 1900 can be provided as a server or a terminal device. Figure 5 , the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0185] The electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2000. TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM or similar.

[0186] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to perform the above method.

[0187] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0188] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0189] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0190] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer-readable program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0191] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0192] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0193] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0194] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0195] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A measurement method for measuring sediment indicators, characterized in that: include: Acquire a spectral signal and a true value of a sediment index of the sample, wherein the spectral signal includes: a scattering signal, and / or a transmission signal, and / or a reflection signal, and the sediment index includes: at least one of: sediment content, average sediment particle size, and sediment composition; For the sample, determining at least one target factor affecting sediment indicators; Using the spectral signal, constructing spectral features characterizing each target factor, a single sample corresponds to multiple spectral features; constructing first data using the spectral feature and the true value of the sediment index; Based on the first data, a sediment index calculation model is determined, and the sediment index calculation model is used to calculate the sediment index of the sample to be tested; wherein the spectral signal is used to construct a spectral feature characterizing each target factor, including: Determine at least two first wavelengths of a spectrum that responds to a target factor, and first spectrum signals corresponding to the first wavelengths, wherein a difference between the first spectrum signals corresponding to the at least two first wavelengths satisfies a preset intensity difference condition; Processing the first spectral signal to eliminate background signals or enhance the sensitivity of the signal to the target factor, and obtaining at least one candidate feature; The spectral features of the target factor are screened out from the candidate features.

2. The method according to claim 1, characterized in that The spectral signal also includes: a fluorescence signal.

3. The method according to claim 1, characterized in that The determining of at least two first wavelengths of a spectrum that responds to a target factor comprises: determining the first wavelength according to the response intensity of the spectrum to the target factor; The processing of the first spectral signal to eliminate the background signal or enhance the sensitivity of the signal to the target factor and obtain at least one candidate feature includes: The ratio of the response intensities corresponding to the two first wavelengths is used as the candidate feature.

4. The method according to claim 3, characterized in that The response intensity includes signal intensity, the target factor includes target particle size, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes: Determining a plurality of target particle sizes based on the particle size distribution of sediment in the sample; Among the wavelengths of the spectrum that responds to a single target particle size, a wavelength pair whose signal intensity difference is greater than a first intensity threshold is determined as the first wavelength; and / or, The response intensity includes signal intensity, the target factor includes a target component, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes: Acquiring water quality information of the sample, wherein the water quality information includes the target component; determining a target wavelength band in the spectrum that responds to a single target component; Determine a wavelength pair within the target wavelength band whose signal strength difference is greater than a second strength threshold as a first wavelength; and / or, The response intensity includes signal intensity, the target factor includes a target component ratio, and determining the first wavelength according to the response intensity of the spectrum to the target factor includes: Obtaining water quality information of the sample, wherein the water quality information includes the proportion of the target component; Among the wavelengths of the response spectrum for a single target component ratio, a wavelength pair whose signal intensity difference is greater than a third intensity threshold is determined as the first wavelength.

5. The method according to claim 2, characterized in that: The step of selecting the spectral feature of the target factor from the candidate features includes: Determining the correlation between each of the candidate features and the true value of the sediment index; According to the correlation, the spectral characteristics of the target factor are determined.

6. The method according to any one of claims 1 to 5, characterized in that The first data includes: training data and test data, and the method of determining a sediment index calculation model based on the first data includes: Obtain multiple initial models; Determine a plurality of first models and a first calculated sediment index corresponding to each of the training data according to the plurality of training data and the initial model, wherein a single training data corresponds to a true value of a sediment index and a first calculated sediment index; Based on the true values ​​of the sediment indicators corresponding to the plurality of training data and the first calculated sediment indicator, the plurality of first models are screened to obtain at least two second models; The test data is used to screen the at least two second models to obtain the sediment index calculation model.

7. The method according to claim 6, characterized in that The method of using the test data to screen the at least two second models to obtain the sediment index calculation model includes: Inputting the test data into the second model to obtain a second sediment calculation index corresponding to the test data, wherein a single test data corresponds to a true value of a sediment index and a second calculated sediment index; Determining at least one model evaluation index based on a difference between a true value of a sediment index of the test data and a second calculated sediment index; Based on the model evaluation index, the sediment index calculation model is determined from the second model.

8. A measuring device for measuring sediment index, characterized in that: include: A sample spectral signal and sediment index acquisition unit, used to acquire a spectral signal of the sample and a true value of a sediment index, wherein the spectral signal includes: a scattering signal, and / or a transmission signal, and / or a reflection signal, and the sediment index includes: at least one of sediment content, average sediment particle size, and sediment composition; A target factor determination unit, used to determine at least one target factor affecting sediment indicators for the sample; A spectral feature determination unit, used to construct spectral features characterizing each target factor using the spectral signal, wherein a single sample corresponds to multiple spectral features; A first data construction unit, configured to construct first data using the spectral feature and the true value of the sediment index; A sediment index calculation model determination unit is used to determine a sediment index calculation model based on the first data, and the sediment index calculation model is used to calculate the sediment index of the sample to be tested; wherein the spectral feature determination unit includes: a first wavelength and first spectral signal determination unit, which is used to determine at least two first wavelengths of the spectrum that responds to the target factor, and a first spectral signal corresponding to the first wavelength, and the difference between the first spectral signals corresponding to the at least two first wavelengths meets a preset intensity difference condition; a candidate feature determination unit, configured to process the first spectral signal to eliminate background signals or enhance the sensitivity of the signal to the target factor, and obtain at least one candidate feature; The spectral feature determination subunit is used to screen out the spectral feature of the target factor from the candidate features.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the method described in any one of claims 1 to 7 when executing the instructions stored in the memory.

10. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images