Detection instrument screening method and system based on water sample analysis

Abstract

The invention is suitable for the technical field of water sample detection, and provides a detection instrument screening method and system based on water sample analysis, and the method comprises the following steps: obtaining characteristic information of a water sample to be detected; the characteristic information comprises smell information, image information and basic physical attribute information; according to the image information of the to-be-detected water sample, spectrum fingerprint characteristics, water surface impurity data and water body real chromaticity of the to-be-detected water sample are determined; according to the smell information, the basic physical attribute information, the spectrum fingerprint characteristics, the water surface impurity data and the real chromaticity of the water sample to be detected, the pollutant types of the water sample to be detected are predicted, predicted values of various pollutants in the water sample to be detected are obtained, and the confidence coefficient of various pollutants contained in the water sample to be detected is determined; and screening corresponding detection instruments according to the confidence coefficient of various pollutants contained in the water sample to be detected. According to the invention, through fusion of multi-source characteristic information such as smell and images, intelligent prediction of water sample pollutant types and detection instrument screening are realized.

Description

Technical Field

[0001] This invention belongs to the field of water sample testing technology, and in particular relates to a method and system for screening testing instruments based on water sample analysis. Background Technology

[0002] Water sample testing is a crucial step in environmental protection, drinking water safety, and industrial production. With technological advancements, various water sample testing instruments targeting different types of pollutants have emerged on the market, such as ammonia nitrogen analyzers, heavy metal analyzers, online automatic CODcr monitors, conventional multi-parameter water analyzers, online oil-water monitors, BOD analyzers, and COD analyzers. However, these instruments differ in their detection principles, accuracy, speed, and applicable scenarios. Given the complexities of water quality testing needs, quickly and accurately selecting the most suitable testing instrument has become a pressing issue.

[0003] Currently, traditional screening methods mainly rely on human experience and subjective perception. When faced with unknown and complex water samples, technicians need to try different testing instruments one by one, which is inefficient, prone to errors, and difficult to standardize. Therefore, there is an urgent need for a system that can intelligently screen testing instruments based on the characteristics of water samples. Summary of the Invention

[0004] The purpose of this invention is to provide a method for screening detection instruments based on water sample analysis, in order to solve the above-mentioned technical problems.

[0005] This invention is implemented as follows: a method for screening detection instruments based on water sample analysis, comprising the following steps: Obtain characteristic information of the water sample to be tested; the characteristic information includes odor information, image information, and basic physical property information; Based on the image information of the water sample to be tested, determine the spectral fingerprint characteristics, surface impurity data, and true color of the water body. Based on the odor information, basic physical property information, spectral fingerprint characteristics, water surface impurity data, and true color of the water sample to be tested, the types of pollutants in the water sample to be tested are predicted, and the predicted values ​​of various pollutants in the water sample to be tested are obtained. Based on the predicted values ​​of various pollutants in the water sample to be tested, determine the confidence level of the presence of various pollutants in the water sample to be tested; Based on a pre-set database of testing instruments, the corresponding testing instruments are selected according to the confidence level of the presence of various pollutants in the water sample to be tested.

[0006] Furthermore, the image information includes hyperspectral imaging data and RGB image data; the steps of determining the spectral fingerprint features, surface impurity data, and true color of the water sample based on the image information of the water sample to be tested specifically include: Extract spectral fingerprint features from hyperspectral imaging data; Identify areas of impurities on the water surface based on RGB image data; Based on the areas of impurities on the water surface, determine the data on water surface impurities and the true color of the water body.

[0007] Furthermore, the step of identifying impurity areas on the water surface based on RGB image data specifically includes: Preprocess RGB image data; Extract color features, texture features, and edge gradient features from the preprocessed RGB image data; Color features, texture features, and edge gradient features are combined into a feature vector, and the impurity region on the water surface is determined based on a pre-defined classification model.

[0008] Furthermore, the steps for determining the surface impurity data and the true color of the water body based on the area of ​​surface impurities include: Based on the area of ​​impurities on the water surface, calculate the proportion of the area of ​​impurities on the water surface to obtain the water surface impurity data; Remove the impurity area on the water surface from the RGB image data to obtain the water body area; Calculate the chromaticity of the water area to obtain the true chromaticity of the water body.

[0009] Furthermore, based on the odor information, basic physical property information, spectral fingerprint characteristics, surface impurity data, and true color of the water sample to be tested, the steps of predicting the types of pollutants in the water sample to be tested and obtaining the predicted values ​​of various pollutants in the water sample to be tested specifically include: First, the baseline values ​​of odor information, basic physical property information, spectral fingerprint features, water surface impurity data, and true color of water body corresponding to various pollutants are preset. Then, the baseline values ​​are compared with the normalized values ​​of odor information, basic physical property information, spectral fingerprint features, water surface impurity data, and true color of water body of the water sample to be tested. Based on the weighted method, the predicted values ​​of various pollutants in the water sample to be tested are calculated.

[0010] Furthermore, the calculation formulas for the predicted values ​​of various pollutants in the water sample to be tested are as follows: ; In the formula, Z i w is the predicted value of pollutant of type i in the water sample to be tested; ij x represents the weight of the j-th characteristic information of the i-th type of pollutant; M represents the number of types of characteristic information in the water sample to be tested; x ij s is the normalized value of the j-th characteristic information of the i-th pollutant; ij This is the baseline value for the j-th characteristic information of the i-th type of pollutant.

[0011] Furthermore, the formula for calculating the confidence level of the presence of various pollutants in the water sample to be tested is as follows: ; In the formula, P i Z represents the confidence level that the water sample to be tested contains pollutant of type i; N represents the number of different types of pollutants; Z represents the confidence level that the water sample contains pollutants of type i. k This represents the predicted value of the kth type of pollutant in the water sample to be tested.

[0012] Furthermore, the basic physical property information includes temperature, electrical conductivity, and light transmittance.

[0013] Another objective of this invention is to provide a water sample analysis-based instrument screening system for implementing the aforementioned instrument screening method, comprising: The characteristic information acquisition module is used to acquire characteristic information of the water sample to be tested; the characteristic information includes odor information, image information and basic physical property information; The image information processing module is used to determine the spectral fingerprint features, surface impurity data, and true color of the water sample based on the image information of the water sample to be tested. The pollutant prediction module is used to predict the types of pollutants in the water sample to be tested based on the odor information, basic physical property information, spectral fingerprint characteristics, water surface impurity data and true color of the water body, and obtain the predicted values ​​of various pollutants in the water sample to be tested. The confidence level determination module is used to determine the confidence level of the presence of various pollutants in the water sample to be tested based on the predicted values ​​of various pollutants in the water sample to be tested. The instrument screening module is used to screen corresponding instruments based on a preset instrument database and the confidence level of the presence of various pollutants in the water sample to be tested.

[0014] Furthermore, the image information processing module specifically includes: The spectral fingerprint extraction unit is used to extract spectral fingerprint features based on hyperspectral imaging data; The impurity region identification unit is used to identify impurity regions on the water surface based on RGB image data. The water sample information determination unit is used to determine the water surface impurity data and the true color of the water body based on the impurity area on the water surface.

[0015] This invention provides a method for screening detection instruments based on water sample analysis. By integrating multi-source characteristic information such as odor and image, it achieves intelligent prediction of the types of pollutants in water samples and screening of detection instruments. This significantly reduces manual intervention and subjective judgment, improves the automation level and reliability of the detection instrument screening process, and overcomes the inefficiency of traditional methods that rely on manual trial and error of instruments one by one. It can complete the water sample characteristic analysis and recommend the optimal detection instrument in a short time, greatly improving detection efficiency. It is suitable for on-site rapid response and batch processing scenarios. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for screening detection instruments based on water sample analysis, provided in an embodiment of the present invention.

[0017] Figure 2 This is a flowchart illustrating step S200 of a water sample analysis-based instrument screening method provided in an embodiment of the present invention.

[0018] Figure 3 This is a flowchart illustrating step S220 of a water sample analysis-based instrument screening method provided in an embodiment of the present invention.

[0019] Figure 4 This is a flowchart illustrating step S230 of a water sample analysis-based instrument screening method provided in an embodiment of the present invention.

[0020] Figure 5 This is a schematic diagram of a screening system for detection instruments based on water sample analysis, provided in an embodiment of the present invention.

[0021] Figure 6 This is a schematic diagram of the image information processing module provided in an embodiment of the present invention.

[0022] Figure 7 This is a schematic diagram of the structure of the impurity region identification unit provided in an embodiment of the present invention.

[0023] Figure 8 This is a schematic diagram of the structure of the water sample information determination unit provided in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] like Figure 1 As shown, in one embodiment of the present invention, a method for screening detection instruments based on water sample analysis is provided, comprising the following steps: S100. Obtain the characteristic information of the water sample to be tested; the characteristic information includes odor information, image information, and basic physical property information; S200. Based on the image information of the water sample to be tested, determine the spectral fingerprint characteristics, surface impurity data, and true color of the water body of the water sample to be tested. S300. Based on the odor information, basic physical property information, spectral fingerprint characteristics, water surface impurity data and true color of the water sample to be tested, predict the types of pollutants in the water sample to be tested and obtain the predicted values ​​of various pollutants in the water sample to be tested. S400. Based on the predicted values ​​of various pollutants in the water sample to be tested, determine the confidence level that the water sample to be tested contains various pollutants. S500: Based on a pre-set database of testing instruments, it selects the corresponding testing instruments according to the confidence level of the presence of various pollutants in the water sample to be tested.

[0026] In this embodiment of the invention, by fusing multi-source characteristic information such as odor and image, intelligent prediction of water sample pollutant types and screening of detection instruments are achieved. This significantly reduces human intervention and subjective judgment, improves the automation level and reliability of the detection instrument screening process, overcomes the inefficiency of traditional methods that rely on manual trial of instruments one by one, and can complete water sample characteristic analysis and recommend the optimal detection instrument in a short time, greatly improving detection efficiency. It is suitable for on-site rapid response and batch processing scenarios.

[0027] In practical applications, odor information can be collected using existing electronic noses; image information includes hyperspectral imaging data and RGB image data. Hyperspectral imaging data can be collected using existing hyperspectral imagers, and RGB image data can be collected using existing color image sensors such as RGB cameras; basic physical property information includes, but is not limited to, temperature, conductivity, and transmittance. Temperature can be measured using thermometers, conductivity can be collected using existing conductivity meters, and transmittance can be collected using existing transmittance meters, etc.

[0028] This invention improves the accuracy of pollutant type prediction by incorporating basic physical properties such as temperature, conductivity, and transmittance. The temperature of the water sample being tested affects the odor and color of pollutants; for example, water samples containing the same type of pollutant will have different colors and odors at different temperatures. Therefore, this invention ensures prediction accuracy by incorporating factors such as temperature.

[0029] like Figure 2 As shown, in a preferred embodiment of the present invention, the step of determining the spectral fingerprint features, surface impurity data, and true color of the water sample based on the image information of the water sample to be tested, i.e., step S200, specifically includes: S210. Extract spectral fingerprint features based on hyperspectral imaging data; S220: Identify impurity areas on the water surface based on RGB image data; S230. Determine the water surface impurity data and the true color of the water body based on the impurity area on the water surface.

[0030] It should be noted that the molecular structure of each chemical substance (pollutant) determines its absorption, reflection, and scattering characteristics to light of different wavelengths; this unique and repeatable spectral response pattern, like a human fingerprint, is therefore referred to as a "spectral fingerprint feature" in this embodiment of the invention. Hyperspectral imaging data can not only acquire the spectral fingerprint features of each pixel but also record the spatial distribution of these spectral fingerprint features; specifically, spectral fingerprint features can be extracted using existing principal component analysis methods and methods such as calculating the first or second derivative of the spectrum.

[0031] In this embodiment of the invention, by using technologies such as hyperspectral imaging and RGB image analysis to extract information such as the spectral fingerprint features of water samples, water surface impurity data, and true color of water bodies, and performing pollutant prediction and confidence calculation, the accuracy of pollutant identification and the reliability of detection instrument matching can be improved.

[0032] like Figure 3 As shown, in a preferred embodiment of the present invention, the step of identifying the water surface impurity region based on RGB image data, i.e., step S220, specifically includes: S221. Preprocess the RGB image data; S222. Extract color features, texture features, and edge gradient features from the preprocessed RGB image data; S223. Combine color features, texture features, and edge gradient features into a feature vector, and determine the impurity area on the water surface based on a preset classification model.

[0033] In practical applications, surface impurities generally include oil layers (oil pollutants such as petroleum and edible oil can form oil layers on the water surface) and suspended solids.

[0034] The training method for the above classification model is as follows: First, preprocess the RGB image data by distortion correction, filtering and denoising; then extract color features, texture features and edge gradient features; then combine the color features, texture features and edge gradient features into a feature vector, and use the labeled image data ("oil layer", "suspended matter", "normal water body" etc.) to train a support vector machine (SVM) or random forest classification model; input the feature vector of the RGB image data of the actual water sample to be tested into the trained classification model to obtain the classification result and determine the impurity area on the water surface.

[0035] The color feature extraction method is as follows: calculate the RGB color histogram and extract color features through the RGB color histogram; it is worth noting that oil layer areas generally exhibit multiple color peaks, while suspended matter areas usually have a more concentrated color distribution. The texture feature extraction method is as follows: calculate the Local Binary Pattern (LBP) features of the image and describe the texture through LBP features; it is worth noting that the LBP response of oil layer areas is weak (smooth texture), while the LBP response of suspended matter and water ripple textures is strong. The edge gradient feature extraction method is as follows: calculate the gradient magnitude using existing operators such as Sobel, and use the gradient magnitude as the edge gradient feature; it is worth noting that oil layer areas suppress small ripples, resulting in a significantly lower gradient magnitude within the region compared to the surrounding water.

[0036] like Figure 4 As shown, in a preferred embodiment of the present invention, the step of determining the water surface impurity data and the true color of the water body based on the water surface impurity region specifically includes: S231. Calculate the proportion of the water surface impurity area based on the water surface impurity area to obtain water surface impurity data; S232. Remove the impurity area on the water surface from the RGB image data to obtain the water body area; S233. Calculate the chromaticity of the water area to obtain the true chromaticity of the water body.

[0037] Specifically, based on the water surface impurity areas identified above, the corresponding pixels are found in the original RGB image, and the ratio of these pixels to the total number of pixels in the original RGB image is calculated to obtain the water surface impurity data. The water surface impurity data is used to characterize the type and content of water surface impurities in the water sample to be tested.

[0038] In addition, based on the water surface impurity areas identified above, the corresponding pixels are found in the original RGB image, and these pixels in the water surface impurity areas are removed, leaving the remaining pixels as the water body area; then, the average R value, average G value, and average B value of each pixel in the water body area are calculated to obtain the true color of the water body.

[0039] In this embodiment of the invention, since the RGB image data is obtained by directly photographing the water sample, the color is the apparent color, which is caused by dissolved substances, suspended solids, oil layers, etc. in the water sample. If only the apparent color is analyzed, it cannot accurately reflect the true color of the water sample, that is, it is difficult to predict the types of dissolved substances in the water sample based on the color. However, this embodiment of the invention first identifies the impurity areas on the water surface, then removes the interference from the impurity areas, and then calculates the chromaticity of the remaining areas. This can more realistically reflect the color produced by dissolved substances in the water sample, thereby improving the accuracy of subsequent pollutant prediction.

[0040] In a preferred embodiment of the present invention, the step of predicting the types of pollutants in the water sample to be tested based on the odor information, basic physical property information, spectral fingerprint characteristics, surface impurity data, and true color of the water sample to be tested, and obtaining the predicted values ​​of various pollutants in the water sample to be tested, specifically includes: First, the baseline values ​​of odor information, basic physical property information, spectral fingerprint features, water surface impurity data, and true color of water body corresponding to various pollutants are preset. Then, the baseline values ​​are compared with the normalized values ​​of odor information, basic physical property information, spectral fingerprint features, water surface impurity data, and true color of water body of the water sample to be tested. Based on the weighted method, the predicted values ​​of various pollutants in the water sample to be tested are calculated.

[0041] It should be noted that the baseline values ​​for odor information, basic physical property information, spectral fingerprint features, surface impurity data, and true water color of various pollutants refer to the normalized values ​​of these parameters under standard environmental conditions (e.g., a temperature of 25°C). These normalized values ​​can be obtained by normalizing the odor information, basic physical property information, spectral fingerprint features, surface impurity data, and true water color. Normalization methods include, but are not limited to, Min-Max normalization and Z-score standardization.

[0042] Specifically, the formulas for calculating the predicted values ​​of various pollutants in the water samples to be tested are as follows: ; In the formula, Z i w is the predicted value of pollutant of type i in the water sample to be tested; ij x represents the weight of the j-th characteristic information of the i-th type of pollutant; M represents the number of types of characteristic information in the water sample to be tested; x ij s is the normalized value of the j-th characteristic information of the i-th pollutant; ij This is the baseline value for the j-th characteristic information of the i-th type of pollutant.

[0043] In practical applications, the aforementioned weight w ij The weights can be determined using a machine learning model. Specifically, the output value of the above formula is used as the input feature of the machine learning model, and the final pollutant type is used as the label. The optimal weights can be determined by training the model using a logistic regression or linear regression model.

[0044] It is worth noting that, in the embodiments of the present invention, the above-mentioned predicted value is used to characterize the degree of deviation between the characteristic information of the water sample to be tested and the benchmark value of the corresponding characteristic information of various pollutants. The larger the predicted value, the smaller the degree of deviation.

[0045] In a preferred embodiment of the present invention, the formula for calculating the confidence level that the water sample to be tested contains various pollutants is as follows: ; In the formula, P i Z represents the confidence level that the water sample to be tested contains pollutant of type i; N represents the number of different types of pollutants; Z represents the confidence level that the water sample contains pollutants of type i. k This represents the predicted value of the kth type of pollutant in the water sample to be tested.

[0046] In practical applications, the database of detection instruments includes, but is not limited to, ammonia nitrogen water quality analyzers, water quality heavy metal analyzers, online automatic CODcr monitoring instruments, conventional multi-parameter water quality analyzers, online oil-in-water monitoring instruments, BOD analyzers, and COD measuring instruments. Each instrument is labeled with the types of pollutants it can detect and their confidence threshold ranges. Based on the confidence levels calculated above, the most suitable detection instrument for the corresponding pollutant can be intelligently selected from the database. For example, the online oil-in-water monitoring instrument detects oil pollutants, and its corresponding confidence threshold range for oil pollutants is ≥85%; the water quality heavy metal analyzer detects heavy metal pollution, and its corresponding confidence threshold range for heavy metal pollutants is ≥80%. If, according to the above method, the confidence level for a water sample containing oil pollutants is determined to be 87%, the confidence level for containing heavy metal pollutants is 3%, and the sum of the confidence levels for other pollutants is 10%, then the online oil-in-water monitoring instrument can be selected to test the water sample.

[0047] like Figure 5 As shown, in another embodiment of the present invention, a detection instrument screening system based on water sample analysis is also provided to implement the above-mentioned detection instrument screening method, comprising: The characteristic information acquisition module 100 is used to acquire characteristic information of the water sample to be tested; the characteristic information includes odor information, image information and basic physical property information; The image information processing module 200 is used to determine the spectral fingerprint features, water surface impurity data and true color of the water sample based on the image information of the water sample to be tested. The pollutant prediction module 300 is used to predict the types of pollutants in the water sample to be tested based on the odor information, basic physical property information, spectral fingerprint characteristics, water surface impurity data and true color of the water body, and to obtain the predicted values ​​of various pollutants in the water sample to be tested. The confidence level determination module 400 is used to determine the confidence level of the presence of various pollutants in the water sample to be tested based on the predicted values ​​of various pollutants in the water sample to be tested. The instrument screening module 500 is used to screen corresponding instruments based on a preset instrument database and the confidence level of various pollutants contained in the water sample to be tested.

[0048] like Figure 6 As shown, in a preferred embodiment of the present invention, the image information processing module 200 specifically includes: The spectral fingerprint extraction unit 210 is used to extract spectral fingerprint features based on hyperspectral imaging data; The impurity region identification unit 220 is used to identify impurity regions on the water surface based on RGB image data; The water sample information determination unit 230 is used to determine the water surface impurity data and the true color of the water body based on the impurity area on the water surface.

[0049] like Figure 7 As shown, in a preferred embodiment of the present invention, the impurity region identification unit 220 specifically includes: Image preprocessing subunit 221 is used to preprocess RGB image data; Image feature extraction subunit 222 is used to extract color features, texture features and edge gradient features from the preprocessed RGB image data; The impurity region determination subunit 223 is used to combine color features, texture features and edge gradient features into a feature vector, and determine the impurity region on the water surface based on a preset classification model.

[0050] like Figure 8 As shown, in a preferred embodiment of the present invention, the water sample information determination unit 230 specifically includes: Impurity data calculation sub-unit 231 is used to calculate the proportion of water surface impurity areas based on the water surface impurity areas to obtain water surface impurity data; The water body region determination sub-unit 232 is used to remove impurity areas on the water surface from the RGB image data to obtain the water body region; Water color calculation sub-unit 233 is used to calculate the color of a water area to obtain the true color of the water body.

[0051] In summary, the water sample analysis-based instrument screening method and system provided by this invention deeply integrates artificial intelligence with water quality analysis, creating an efficient, accurate, and automated intelligent instrument screening solution. Specifically, this invention achieves intelligent prediction and instrument screening of water sample pollutant types through the fusion of multi-source characteristic information such as odor and images. Through accurate prediction and recommendation, it guides users to directly use the most suitable instrument, avoiding problems such as repeated testing, reagent waste, and equipment damage caused by improper selection. This saves time and economic costs, significantly reduces human intervention and subjective judgment, improves the automation level and reliability of the instrument screening process, overcomes the inefficiency of traditional methods that rely on manual instrument trials, and can complete water sample characteristic analysis and recommend the optimal instrument in a short time, greatly improving detection efficiency. It is suitable for rapid on-site response and batch processing scenarios.

[0052] Furthermore, this invention utilizes hyperspectral imaging and RGB image analysis to extract spectral fingerprint features, surface impurity data, and true water color information from water samples. It then performs pollutant prediction and confidence level calculations, improving the accuracy of pollutant identification and the reliability of instrument matching. Specifically, by identifying and removing surface oil layers, suspended solids, and other impurity areas, this invention effectively obtains the true water color produced solely by dissolved substances. This solves the problem of misleading judgments based on the apparent color of traditional water samples, providing crucial evidence for accurately identifying dissolved pollutants in water samples. Moreover, this invention incorporates spectral fingerprint features using hyperspectral imaging technology, enabling more precise identification and differentiation of pollutants with similar chemical compositions. This significantly improves the accuracy of pollutant type prediction, laying a solid foundation for subsequent screening of detection instruments.

[0053] It should be noted that each of the above modules or units can be implemented as a computer program, which can run on a computer device. The computer device's memory can store the computer program that makes up each module, enabling the processor to execute each step of the above method.

[0054] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0055] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory.

[0056] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A method for screening detection instruments based on water sample analysis, characterized in that, The method comprises the following steps: Obtaining characteristic information of a water sample to be detected; the characteristic information comprises odor information, image information and basic physical attribute information; According to the image information of the water sample to be detected, determining spectral fingerprint characteristics, water surface impurity data and real color of the water body of the water sample to be detected; According to the odor information, the basic physical attribute information, the spectral fingerprint characteristics, the water surface impurity data and the real color of the water body of the water sample to be detected, predicting the types of pollutants in the water sample to be detected, and obtaining prediction values of various types of pollutants in the water sample to be detected; According to the prediction values of various types of pollutants in the water sample to be detected, determining the confidence of the water sample to be detected containing various types of pollutants; Based on a preset detector database, according to the confidence of the water sample to be detected containing various types of pollutants, screening corresponding detection instruments.

2. The water sample based detection instrument screening method of claim 1, wherein, The image information comprises hyperspectral imaging data and RGB image data; according to the image information of the water sample to be detected, the spectral fingerprint characteristics, the water surface impurity data and the real color of the water body of the water sample to be detected are determined, and the step specifically comprises: According to the hyperspectral imaging data, extracting spectral fingerprint characteristics; According to the RGB image data, identifying the water surface impurity region; According to the water surface impurity region, determining the water surface impurity data and the real color of the water body.

3. The water sample based detection instrument screening method of claim 2, wherein, According to the RGB image data, the step of identifying the water surface impurity region specifically comprises: Pretreating the RGB image data; Extracting color features, texture features and edge gradient features in the pretreated RGB image data; Combining the color features, the texture features and the edge gradient features into a feature vector, and determining the water surface impurity region based on a preset classification model.

4. The water sample based detection instrument screening method of claim 2, wherein, According to the water surface impurity region, the step of determining the water surface impurity data and the real color of the water body specifically comprises: According to the water surface impurity region, calculating the proportion of the water surface impurity region to obtain the water surface impurity data; Eliminating the water surface impurity region in the RGB image data to obtain a water body region; Calculating the color of the water region to obtain the real color of the water body.

5. The water sample based detection instrument screening method of claim 1, wherein, According to the odor information, the basic physical attribute information, the spectral fingerprint characteristics, the water surface impurity data and the real color of the water body of the water sample to be detected, the step of predicting the types of pollutants in the water sample to be detected, and obtaining prediction values of various types of pollutants in the water sample to be detected, specifically comprises: First, preset the baseline values of the odor information, the basic physical attribute information, the spectral fingerprint characteristics, the water surface impurity data and the real color of the water body corresponding to various types of pollutants, and then compare them with the normalized values of the odor information, the basic physical attribute information, the spectral fingerprint characteristics, the water surface impurity data and the real color of the water body of the water sample to be detected, and calculate the prediction values of various types of pollutants in the water sample to be detected based on a weight weighting method.

6. The water sample based detection instrument screening method of claim 5, wherein, The calculation formula of the prediction values of various types of pollutants in the water sample to be detected is as follows: ； In the formula, Z i is the predicted value of the i-th type of pollutant in the water sample to be detected; w ij is the weight of the j-th characteristic information of the i-th type of pollutant. M is the number of types of characteristic information of the water sample to be detected; x ij is the normalized value of the jth characteristic information of the ith contaminant; s ij is the reference value of the jth characteristic information of the ith contaminant.

7. The water sample based detection instrument screening method of claim 6, wherein, The calculation formula of the confidence of the water sample to be detected containing various types of pollutants is as follows: ； In the formula, P i is the confidence degree that the i-th type of pollutant is contained in the water sample to be detected; N is the number of types of pollutants; Z k is the predicted value of the k-th type of pollutant in the water sample to be detected.

8. The water-based analytical based detection instrument screening method according to claim 1 or 5 or 6, characterized in that, The basic physical attribute information comprises temperature, conductivity and light transmittance.

9. A water sample analysis based detection instrument screening system for implementing the detection instrument screening method of any one of claims 1-8, characterized by, It comprises: A characteristic information acquisition module is configured to acquire characteristic information of a water sample to be detected; the characteristic information comprises odor information, image information and basic physical attribute information; The image information processing module is configured to determine spectral fingerprint features, surface impurity data, and real color of the water sample to be detected according to image information of the water sample to be detected. The pollutant prediction module is configured to predict the types of pollutants in the water sample to be detected according to the odor information, the basic physical attribute information, the spectral fingerprint features, the surface impurity data, and the real color of the water sample to be detected, and obtain prediction values of the types of pollutants in the water sample to be detected. The confidence determination module is configured to determine the confidence of the types of pollutants contained in the water sample to be detected according to the prediction values of the types of pollutants in the water sample to be detected. The detection instrument screening module is configured to screen a corresponding detection instrument according to the confidence of the types of pollutants contained in the water sample to be detected based on a preset detection instrument database.

10. The water sample based detection instrument screening system of claim 9, wherein, The image information processing module specifically includes: The spectral fingerprint extraction unit is configured to extract spectral fingerprint features according to hyperspectral imaging data. The impurity region identification unit is configured to identify a surface impurity region according to RGB image data. The water sample information determination unit is configured to determine surface impurity data and real color of the water sample according to the surface impurity region.

Images