Hyperspectral camera-based organic solid wastewater component and element detection system and method

Abstract

The invention relates to the technical field of waste incineration power generation, in particular to an organic solid waste water component and element detection system and method based on a hyperspectral camera, all parts of the organic solid waste water component and element detection system based on the hyperspectral camera are started, self-inspection is completed, an operator remotely controls a camera walking platform, and the camera walking platform is controlled by the operator. The hyperspectral camera is moved to the expected position of the edge of the garbage pit, the hyperspectral camera is started, camera parameters are set, an image starts to be shot, software in the computer is started, the software in the computer can cut the hyperspectral image into a plurality of rectangles with the same area, and the rectangles are stored in the computer; then, a pre-trained machine learning model is called to predict the moisture content and carbon-hydrogen-oxygen (CHO) element composition of each rectangular spectrum curve, and then detection work is completed, through real-time monitoring and quick response, the system and method can improve the organic solid waste treatment efficiency and accuracy, and optimize the organic solid waste incineration power generation process.

Description

Technical Field

[0001] The present invention relates to the technical field of waste incineration power generation, and in particular to a system and method for detecting moisture and elements in organic solid waste based on a hyperspectral camera. Background Art

[0002] In the process of environmental protection, organic solid waste treatment is a crucial link. With the acceleration of urbanization and the improvement of residents' living standards, the output of organic solid waste has increased year by year, bringing huge pressure to environmental protection. Among the current organic solid waste treatment technologies, incineration power generation is a treatment method with significant advantages.

[0003] In the process of incineration and power generation of organic solid waste, it is of great significance to detect the moisture content and carbon, hydrogen and oxygen (CHO) elements of organic solid waste. The moisture content of organic solid waste directly affects the efficiency of incineration and power generation. Too high or too low moisture content will have an adverse effect on incineration. The CHO element composition determines the calorific value and combustion stability of organic solid waste. By detecting these elements, organic solid waste can be classified and optimized, and the efficiency and quality of incineration power generation can be improved.

[0004] However, the existing method for detecting moisture and CHO elements in organic solid waste mainly relies on special instruments in the laboratory to complete offline analysis. This method involves complex operations such as sampling and drying. It is time-consuming and difficult to achieve efficient online monitoring, which limits the use of analysis results to improve the operation of waste incineration power generation. Summary of the invention

[0005] The purpose of the present invention is to provide a system and method for detecting moisture and elements in organic solid waste based on a hyperspectral camera, aiming to solve the problem that the existing method for detecting moisture and CHO elements in organic solid waste mainly relies on special instruments in the laboratory to complete offline analysis. This method involves complex operations such as sampling and drying, which is time-consuming and difficult to achieve efficient online monitoring, limiting the technical problem of using the analysis results to improve the operation of waste incineration power generation.

[0006] To achieve the above-mentioned objectives, the present invention adopts an organic solid waste moisture and element detection system based on a hyperspectral camera, including a camera walking platform, a supporting turntable, a computer, an electric slide rail and a hyperspectral camera. The supporting turntable and the computer are both arranged on the camera walking platform, the electric slide rail is installed on the supporting turntable, and the hyperspectral camera is installed on the electric slide rail.

[0007] Among them, the organic solid waste moisture and element detection system based on a hyperspectral camera also includes a halogen light source, and the halogen light source is arranged on the electric slide rail.

[0008] The present invention also provides a method for detecting moisture and elements in organic solid waste using a hyperspectral camera, using the organic solid waste moisture and element detection system based on a hyperspectral camera as described above.

[0009] The steps include:

[0010] Step 1: construct a machine learning model of the moisture content and CHO element composition of organic solid waste, and deploy it in the computer;

[0011] Step 2: Start each component of the organic solid waste moisture and element detection system based on the hyperspectral camera and complete self-test;

[0012] Step 3: The operator remotely controls the camera walking platform to move it to a desired position at the edge of the garbage pit;

[0013] Step 4: Start the hyperspectral camera, set the camera parameters, and start taking images;

[0014] Step 5: Start the software in the computer to call the pre-trained machine learning model to predict the moisture and CHO element composition of the photographed organic solid waste to complete the detection work.

[0015] Among them, the specific method of constructing the machine learning model of moisture content and CHO element composition of organic solid waste is as follows:

[0016] Collect organic solid waste samples, crush and mix them until no 2mm particles are visible, and then obtain the moisture content and CHO element composition of the organic solid waste through experimental tests based on instrumental analysis;

[0017] A hyperspectral camera is used to collect hyperspectral images of organic solid waste;

[0018] The obtained moisture content and CHO element composition of organic solid waste are related to the hyperspectral image to construct a machine learning model.

[0019] The specific method of collecting organic solid waste samples, crushing and mixing them so that no 2 mm particles are visible, and then obtaining the moisture content and CHO element composition of the organic solid waste through experimental testing based on instrumental analysis is as follows:

[0020] Collect organic solid waste samples, crush and mix them until no particles smaller than 2 mm are visible;

[0021] Steps to obtain moisture content:

[0022] Weigh the organic solid waste sample and place it in a constant weight crucible with a crucible mass of m0. Record the mass of the crucible and sample before drying m1. Place the crucible with the sample in a 105°C oven and dry it to constant weight, then cool it to room temperature. Weigh the total mass of the crucible and sample after drying m2.

[0023]

[0024] Steps to obtain the composition of CHO elements:

[0025] The samples were placed in an oven at 105°C for drying, and then the CHO element composition (mass fraction) of the samples was determined using an organic element analyzer.

[0026] Use a hyperspectral camera to collect hyperspectral images of organic solid waste:

[0027] For this specific implementation, a standard whiteboard and blackboard are used to obtain the wavelength curve of the light source. In this process, the hyperspectral images of the whiteboard and blackboard of the halogen light source in the laboratory are taken, and the hyperspectral images of the whiteboard and blackboard of the halogen light source installed on the camera walking platform are taken;

[0028] A hyperspectral camera was used to collect hyperspectral images of organic solid waste, with a wavelength range of 920-1700nm and a step length of 10nm. 79 hyperspectral images in different bands were obtained for each sample, and the images were corrected by the software on the computer. The collected original near-infrared hyperspectral image was corrected for black and white according to the following formula to obtain the corrected image R:

[0029]

[0030] Among them, R is the corrected hyperspectral image, I s is the original hyperspectral image of the sample collected; D is the standard blackboard calibration image collected by the near-infrared hyperspectral system; W is the standard whiteboard calibration image collected by the near-infrared hyperspectral system;

[0031] Select each region of interest in the image, extract the spectral curve of the sample, and calculate the average spectral value.

[0032] The obtained moisture content and CHO element composition of organic solid waste are related to the hyperspectral image to build a machine learning model.

[0033] According to this specific implementation method, firstly, the water content and CHO element composition of organic solid waste are subjected to data preprocessing with the hyperspectral image;

[0034] The characteristic wavelength selection method is used to select the characteristic wavelengths with significant differences under different treatments;

[0035] Divide the preprocessed data into a test data set and a training data set according to a predetermined ratio;

[0036] Select one or more of the support vector regression, random forest, and partial least squares regression modeling algorithms as candidate prediction models;

[0037] Grid search and cross validation are used to select the optimal parameters for the hyperparameters involved in the relevant prediction model to obtain the model configuration;

[0038] The machine learning model can be obtained by performing model evaluation on the model configuration.

[0039] Among them, the data preprocessing methods for the moisture content and CHO element composition of organic solid waste and hyperspectral images include standardization, standard normal variable transformation and multivariate scattering correction;

[0040] Among them, standardization is used to make data of different dimensions have the same distribution scale. The formula is as follows:

[0041]

[0042] Among them, multivariate scattering correction is used to eliminate the spectral differences caused by different scattering levels, thereby enhancing the correlation between the spectrum and the data. The calculation process is as follows

[0043] For the i-th band of all samples, calculate the average value as the benchmark:

[0044]

[0045] Establish a univariate linear regression based on the benchmark data and determine the regression coefficient and bias of each sample:

[0046]

[0047] Correct the original sample:

[0048]

[0049] Among them, the standard normal variable transformation SNV is mainly used to remove the systematic offset and scaling effect in the spectral data; the SNV algorithm processes a spectrum, the formula is:

[0050]

[0051] In the formula, m is the number of wavelength points, k=1,2,…,m

[0052] Among them, the determination coefficient R is used to evaluate the model configuration. 2 and root mean square error MSE;

[0053] The specific formula is as follows:

[0054] Coefficient of determination

[0055] Among them, y i represents the actual observed value, represents the mean of the true observations, Indicates the predicted value. The maximum value of the coefficient of determination is 1 and the minimum value is 0. The closer its value is to 1, the better the model is; the closer its value is to 0, the worse the model is.

[0056] Mean square error

[0057] in The mean square error is a convenient way to measure the "average error" and can evaluate the degree of data variation. The smaller the value, the better the accuracy of the prediction model in describing the experimental data.

[0058] The present invention discloses a system and method for detecting moisture and elements of organic solid waste based on a hyperspectral camera. The system starts various components of the system for detecting moisture and elements of organic solid waste based on a hyperspectral camera and completes self-inspection. An operator remotely controls the camera walking platform to move it to a desired position at the edge of a garbage pit, starts the hyperspectral camera, sets camera parameters, starts taking images, and starts the software in the computer. The software in the computer cuts the hyperspectral image into a number of rectangles of equal area, and then calls a pre-trained machine learning model to predict the moisture content and CHO element of the spectral curve of each rectangle, thereby completing the detection work. In this way, the prior art method for detecting moisture and CHO element of organic solid waste is solved, which mainly relies on special instruments in the laboratory to complete offline analysis. This method involves complex operations such as sampling and drying. Because it is time-consuming, it is difficult to achieve efficient online monitoring, which limits the technical problem of using the analysis results to improve the operation of waste incineration power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 It is a structural schematic diagram of the organic solid waste moisture and element detection system based on a hyperspectral camera of the present invention.

[0061] 1-Camera walking platform, 2-Hyperspectral camera, 3-Halogen light source, 4-Computer, 5-Support turntable, 6-Electric slide rail. DETAILED DESCRIPTION

[0062] Embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0063] See also Figure 1 , Figure 1 It is a structural schematic diagram of the organic solid waste moisture and element detection system based on a hyperspectral camera of the present invention.

[0064] The present invention provides an organic solid waste moisture and element detection system based on a hyperspectral camera, comprising a camera walking platform 1, a supporting turntable 5, a computer 4, an electric slide rail 6 and a hyperspectral camera 2, wherein the supporting turntable 5 and the computer 4 are both arranged on the camera walking platform 1, the electric slide rail 6 is installed on the supporting turntable 5, and the hyperspectral camera 2 is installed on the electric slide rail 6;

[0065] The organic solid waste moisture and element detection system based on a hyperspectral camera also includes a halogen light source 3 , which is disposed on the electric slide rail 6 .

[0066] The camera walking platform 1 is modified from a wire-controlled chassis, and all other equipment are installed on the platform. The camera walking platform 1 can be remotely controlled to walk around the garbage pit to capture hyperspectral images of organic solid waste at different locations; the hyperspectral camera 2 is installed on the electric slide rail 6 to adjust the focus of the hyperspectral camera 2; the electric slide rail 6 is installed on the supporting turntable 5 to rotate the direction of the hyperspectral camera 2. The hyperspectral camera 2 has a photographic wavelength range of 920 to 1700 nm, and can capture hyperspectral images in steps of 5, 10, 15 nm, etc.; the computer 4 receives remote instructions and controls the actions of the camera walking platform 1, and is also responsible for controlling the hyperspectral camera 2, the halogen light source 3, and the electric slide rail 6. After the shooting is completed, the moisture content and CHO element composition are estimated, and the image, moisture content and CHO element composition are sent to the server or user. The above components are connected by signal cables.

[0067] The present invention also provides a method for detecting moisture and elements in organic solid waste using a hyperspectral camera 2, using the organic solid waste moisture and element detection system based on a hyperspectral camera as described above,

[0068] The steps include:

[0069] S101, constructing a machine learning model of the moisture content and CHO element composition of organic solid waste, and deploying it in the computer 4;

[0070] S102, starting each component of the organic solid waste moisture and element detection system based on the hyperspectral camera and completing self-inspection;

[0071] S103, the operator remotely controls the camera walking platform 1 to move it to a desired position at the edge of the garbage pit;

[0072] S104, start the hyperspectral camera 2, set camera parameters, and start taking images;

[0073] S105, starting the software in the computer 4 to call the pre-trained machine learning model to predict the moisture and CHO element composition of the photographed organic solid waste, and completing the detection work.

[0074] Among them, the specific method of constructing the machine learning model of moisture content and CHO element composition of organic solid waste is as follows:

[0075] Collect organic solid waste samples, crush and mix them until no 2mm particles are visible, and then obtain the moisture content and CHO element composition of the organic solid waste through experimental tests based on instrumental analysis;

[0076] According to this specific implementation method, firstly, an organic solid waste sample is collected, crushed and mixed so that no particles of 2 mm are visible;

[0077] Steps to obtain moisture content:

[0078] Weigh the organic solid waste sample and place it in a constant weight crucible with a crucible mass of m0. Record the mass of the crucible and sample before drying m1. Place the crucible with the sample in a 105°C oven and dry it to constant weight, then cool it to room temperature. Weigh the total mass of the crucible and sample after drying m2.

[0079]

[0080] Steps to obtain the CHO elemental composition:

[0081] The samples were placed in an oven at 105°C for drying, and then the CHO element composition of the samples was determined using an organic element analyzer.

[0082] A hyperspectral camera 2 is used to collect hyperspectral images of organic solid waste;

[0083] The obtained moisture content and CHO element composition of organic solid waste are related to the hyperspectral image to construct a machine learning model.

[0084] Among them, a standard whiteboard and a blackboard are used to obtain the wavelength curve of the light source;

[0085] Hyperspectral camera 2 is used to collect hyperspectral images of organic solid waste, with a wavelength range of 920-1700nm and a step length of 10nm. 79 hyperspectral images in different bands are obtained for each sample, and the images are corrected by software on computer 4. The collected original near-infrared hyperspectral image is corrected for black and white according to the following formula to obtain the corrected image R:

[0086]

[0087] Among them, R is the corrected hyperspectral image, I s is the original hyperspectral image of the sample collected; D is the standard blackboard calibration image collected by the near-infrared hyperspectral system; W is the standard whiteboard calibration image collected by the near-infrared hyperspectral system;

[0088] Select each region of interest in the image, extract the spectral curve of the sample, and calculate the average spectral value.

[0089] The specific method of obtaining the wavelength curve of the light source using a standard whiteboard and a blackboard is as follows:

[0090] Take hyperspectral images of whiteboard and blackboard of halogen light source 3 in the laboratory;

[0091] The whiteboard and blackboard hyperspectral images of the halogen light source 3 installed on the camera walking platform 1 are captured.

[0092] The specific steps for building a machine learning model by relating the moisture content and CHO element composition of the obtained organic solid waste to the hyperspectral image are as follows:

[0093] Data preprocessing of moisture content and CHO element composition of organic solid waste and hyperspectral images;

[0094] The characteristic wavelength selection method is used to select the characteristic wavelengths with significant differences under different treatments. In this method, the characteristic screening method is used to select the principal component analysis PCA. The implementation steps are: subtract the mean value (centering) → calculate the covariance matrix → calculate the eigenvalues ​​and eigenvectors of the covariance matrix → sort the eigenvalues ​​→ retain the eigenvectors corresponding to the first N largest eigenvalues ​​→ convert the original features to the new space constructed by the above N eigenvectors;

[0095] Divide the preprocessed data into a test data set and a training data set according to a predetermined ratio;

[0096] Select one or more of the support vector regression, random forest, and partial least squares regression modeling algorithms as candidate prediction models;

[0097] Grid search and cross validation are used to select the optimal parameters for the hyperparameters involved in the relevant prediction model to obtain the model configuration;

[0098] The machine learning model can be obtained by performing model evaluation on the model configuration.

[0099] Among them, the data preprocessing methods for the moisture content and CHO element composition of organic solid waste and hyperspectral images include standardization, standard normal variable transformation and multivariate scattering correction;

[0100] Among them, standardization is used to make data of different dimensions have the same distribution scale. The formula is as follows:

[0101]

[0102] Among them, multivariate scattering correction is used to eliminate the spectral differences caused by different scattering levels, thereby enhancing the correlation between the spectrum and the data. The calculation process is as follows

[0103] For the i-th band of all samples, calculate the average value as the benchmark:

[0104]

[0105] Establish a univariate linear regression based on the benchmark data and determine the regression coefficient and bias of each sample:

[0106]

[0107] Correct the original sample:

[0108]

[0109] Among them, the standard normal variable transformation SNV is mainly used to remove the systematic offset and scaling effect in the spectral data; the SNV algorithm processes a spectrum, the formula is:

[0110]

[0111] In the formula, m is the number of wavelength points, k=1,2,…,m

[0112] Among them, the determination coefficient R is used to evaluate the model configuration. 2 and root mean square error MSE;

[0113] The specific formula is as follows:

[0114] Coefficient of determination

[0115] Among them, y i represents the actual observed value, represents the mean of the true observations, Indicates the predicted value. The maximum value of the coefficient of determination is 1 and the minimum value is 0. The closer its value is to 1, the better the model is; the closer its value is to 0, the worse the model is.

[0116] Mean square error

[0117] in The mean square error is a convenient way to measure the "average error" and can evaluate the degree of data variation. The smaller the value, the better the accuracy of the prediction model in describing the experimental data.

[0118] Using a system and method for detecting moisture and elements of organic solid waste based on a hyperspectral camera of the present embodiment, various components of the system for detecting moisture and elements of organic solid waste based on a hyperspectral camera are started, and self-inspection is completed. The operator remotely controls the camera walking platform 1 to move it to the desired position at the edge of the garbage pit, starts the hyperspectral camera 2, sets camera parameters, starts taking images, and starts the software in the computer 4. The software in the computer 4 cuts the hyperspectral image into a number of rectangles of equal area, and then calls a pre-trained machine learning model to predict the moisture content and CHO element of the spectral curve of each rectangle, thereby completing the detection work. In this way, the problem of the organic solid waste moisture and CHO element detection method in the prior art, which mainly relies on special instruments in the laboratory to complete offline analysis, and the method involves complex operations such as sampling and drying, is time-consuming and difficult to achieve efficient online monitoring, which limits the technical problem of using the analysis results to improve the operation of waste incineration power generation.

[0119] What is disclosed above is only a preferred embodiment of the present invention, and it certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. An organic solid waste moisture and element detection system based on a hyperspectral camera, characterized in that: The invention comprises a camera walking platform, a supporting turntable, a computer, an electric slide rail and a hyperspectral camera. The supporting turntable and the computer are both arranged on the camera walking platform, the electric slide rail is installed on the supporting turntable, and the hyperspectral camera is installed on the electric slide rail.

2. The organic solid waste moisture and element detection system based on a hyperspectral camera as claimed in claim 1, characterized in that: The organic solid waste moisture and element detection system based on a hyperspectral camera also includes a halogen light source, which is arranged on the electric slide rail.

3. A method for detecting moisture and elements of organic solid waste based on a hyperspectral camera, using the organic solid waste moisture and element detection system based on a hyperspectral camera as claimed in claim 2, characterized in that: The steps include: Step 1: construct a machine learning model of the moisture content and CHO element composition of organic solid waste, and deploy it in the computer; Step 2: Start each component of the organic solid waste moisture and element detection system based on the hyperspectral camera and complete self-test; Step 3: The operator remotely controls the camera walking platform to move it to a desired position at the edge of the garbage pit; Step 4: Start the hyperspectral camera, set the camera parameters, and start taking images; Step 5: Start the software in the computer to call the pre-trained machine learning model to predict the moisture and CHO element composition of the photographed organic solid waste to complete the detection work.

4. The organic solid waste moisture and element detection system based on a hyperspectral camera as claimed in claim 3 is characterized in that: The specific method of constructing a machine learning model for the moisture content and CHO element composition of organic solid waste is as follows: Collect organic solid waste samples, crush and mix them until no 2mm particles are visible, and then obtain the moisture content and CHO element composition of the organic solid waste through experimental tests based on instrumental analysis; A hyperspectral camera is used to collect hyperspectral images of organic solid waste; The obtained moisture content and CHO element composition of organic solid waste are related to the hyperspectral image to construct a machine learning model.

5. The method for detecting moisture and elements in organic solid waste based on a hyperspectral camera as claimed in claim 4, characterized in that: Collect organic solid waste samples, crush and mix them until no 2 mm particles are visible, and then conduct experimental tests based on instrumental analysis to obtain the moisture content and CHO element composition of organic solid waste. The specific method is as follows: Collect organic solid waste samples, crush and mix them until no particles smaller than 2 mm are visible; Steps to obtain moisture content: Weigh the organic solid waste sample and place it in a constant weight crucible with a crucible mass of m0. Record the mass of the crucible and sample before drying m1. Place the crucible with the sample in a 105°C oven and dry it to constant weight, then cool it to room temperature. Weigh the total mass of the crucible and sample after drying m2. Steps to obtain the composition of CHO elements: The samples were dried in an oven at 105°C, and then the CHO element composition of the samples was determined using an organic element analyzer.

6. The method for detecting moisture and elements in organic solid waste based on a hyperspectral camera as claimed in claim 5, characterized in that: Use standard whiteboard and blackboard to obtain the wavelength curve of light source; A hyperspectral camera was used to collect hyperspectral images of organic solid waste, with a wavelength range of 920-1700nm and a step length of 10nm. 79 hyperspectral images in different bands were obtained for each sample, and the images were corrected by the software on the computer. The collected original near-infrared hyperspectral image was corrected for black and white according to the following formula to obtain the corrected image R: Among them, R is the corrected hyperspectral image, I s is the original hyperspectral image of the sample collected; D is the standard blackboard calibration image collected by the near-infrared hyperspectral system; W is the standard whiteboard calibration image collected by the near-infrared hyperspectral system; Select each region of interest in the image, extract the spectral curve of the sample, and calculate the average spectral value.

7. The method for detecting moisture and elements in organic solid waste based on a hyperspectral camera according to claim 6, characterized in that: The specific method of obtaining the wavelength curve of the light source using a standard whiteboard and a blackboard is: Take hyperspectral images of whiteboards and blackboards under halogen light sources in the laboratory; The hyperspectral images of the whiteboard and the blackboard of the halogen light source installed on the camera walking platform are taken.

8. The method for detecting moisture and elements in organic solid waste based on a hyperspectral camera as claimed in claim 6, characterized in that: The specific steps for building a machine learning model by correlating the moisture content and CHO element composition of the obtained organic solid waste with the hyperspectral image are as follows: Data preprocessing of moisture content and CHO element composition of organic solid waste and hyperspectral images; The characteristic wavelength selection method is used to select the characteristic wavelengths with significant differences under different treatments; Divide the preprocessed data into a test data set and a training data set according to a predetermined ratio; Select one or more of the support vector regression, random forest, and partial least squares regression modeling algorithms as candidate prediction models; Grid search and cross validation are used to select the optimal parameters for the hyperparameters involved in the relevant prediction model to obtain the model configuration; The machine learning model can be obtained by performing model evaluation on the model configuration.

9. The method for detecting moisture and elements in organic solid waste based on a hyperspectral camera according to claim 6, characterized in that: The data preprocessing methods for the moisture content and CHO element composition of organic solid waste and hyperspectral images include standardization, standard normal variable transformation and multivariate scatter correction; Among them, standardization is used to make data of different dimensions have the same distribution scale. The formula is as follows: Among them, multivariate scattering correction is used to eliminate the spectral differences caused by different scattering levels, thereby enhancing the correlation between the spectrum and the data. The calculation process is as follows For the i-th band of all samples, calculate the average value as the benchmark: Establish a univariate linear regression based on the benchmark data and determine the regression coefficient and bias of each sample: Correct the original sample: Among them, the standard normal variable transformation SNV is mainly used to remove the systematic offset and scaling effect in the spectral data; the SNV algorithm processes a spectrum, the formula is: In the formula, m is the number of wavelength points, k = 1, 2,…, m.

10. The method for detecting moisture and elements in organic solid waste based on a hyperspectral camera according to claim 6, characterized in that: The determination coefficient R is used to evaluate the model configuration. 2 and root mean square error MSE; The specific formula is as follows: Among them, y i represents the actual observed value, represents the mean of the true observations, Indicates the predicted value; the maximum value of the determination coefficient is 1 and the minimum is 0. The closer its value is to 1, the better the model is; the closer its value is to 0, the worse the model is; in The mean square error is a convenient way to measure the "average error" and evaluate the degree of data variation. The smaller the value, the better the accuracy of the prediction model in describing the experimental data.

Images