A method, system, and detection method for constructing a model for detecting crude protein content in chicken manure.

Abstract

A method, system, and detection method for constructing a model for detecting crude protein content in chicken manure, relating to the field of bioinformatics, solves the problems of high pollution during the detection process and severe information redundancy, high cost, and poor model stability in existing chicken manure crude protein content detection technologies. The method involves obtaining the visible-near-infrared reflectance spectrum and crude protein content of chicken manure from a sample set; obtaining three spectral indices corresponding to the visible-near-infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum; selecting the optimal wavelength positions and corresponding spectral indices; constructing a data set by combining the three spectral indices with the crude protein content; fitting the data using various methods; evaluating the various fitting equations; and selecting the best fitting equation, along with the corresponding preprocessing method and wavelength position, as the chicken manure crude protein content detection model. The method described in this invention is applicable to the detection of crude protein content in chicken manure for manure resource utilization.

Description

Technical Field

[0001] This invention relates to the field of bioinformatics technology, specifically to the field of crude protein content detection technology in chicken manure. Background Technology

[0002] Chicken manure, a high-yield and nutrient-rich byproduct of chicken farming, plays a vital role in the sustainable development of agriculture through its resource utilization. Crude protein, a key indicator of chicken manure quality, not only evaluates the feed conversion efficiency of chickens but also functions crucially in the resource utilization of chicken manure. Therefore, accurately and efficiently determining the crude protein content in chicken manure is of significant practical importance for promoting the high-value utilization of agricultural waste, achieving integrated crop-livestock farming, and developing green agriculture.

[0003] Current methods for detecting crude protein content in chicken manure, such as the Kjeldahl method, while highly accurate, are cumbersome, time-consuming, and require large amounts of reagents, resulting in high costs and the risk of chemical contamination. These methods also demand a high level of expertise from the testing personnel and are insufficient for the practical needs of large-scale, rapid testing, especially in field applications such as farms.

[0004] In recent years, spectroscopic analysis technology has been widely used in agricultural testing due to its advantages such as speed, non-destructive nature, and environmental friendliness. However, there is a lack of research on its direct application to chicken manure. The principle of spectroscopic analysis technology is based on the reflection and absorption characteristics of substances to different wavelengths of light. By establishing a mathematical model between the spectrum and target physicochemical indicators, quantitative or qualitative analysis of sample components can be achieved. However, if this technology is directly applied to the detection of crude protein content in chicken manure, the full-spectrum modeling method suffers from high data dimensionality and severe information redundancy, posing significant challenges to model stability and operational efficiency, and often relying on expensive spectroscopic equipment.

[0005] In summary, existing technologies for detecting crude protein content in chicken manure suffer from several problems, including significant pollution during the detection process, severe information redundancy when directly applying spectral analysis techniques to detect crude protein in chicken manure, high costs, and poor model stability. Summary of the Invention

[0006] This invention solves the problems of high pollution during the detection process of existing chicken manure crude protein content detection technologies, and severe information redundancy, high cost, and poor model stability when spectral analysis technology is directly applied to chicken manure crude protein detection. This invention provides the following solution:

[0007] Option 1: A method for constructing a model for detecting crude protein content in chicken manure, comprising the following steps:

[0008] Step A1: Obtain a sample set, which includes the visible-near infrared reflectance spectrum and crude protein content of chicken manure;

[0009] Step A2: The visible-near infrared reflectance spectrum is preprocessed using the MSC preprocessing method and the SNV preprocessing method respectively to obtain the corresponding MSC preprocessed spectrum and SNV preprocessed spectrum;

[0010] Step A3: Within the full-spectrum wavelength range of 400nm to 900nm, obtain the corresponding three spectral indices based on all wavelength positions in the visible-near-infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, respectively.

[0011] The three spectral indices mentioned include the ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI.

[0012] Step A4: Evaluate the correlation between the spectral index and the corresponding crude protein content of chicken manure, and screen the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum respectively.

[0013] Step A5: Based on the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, obtain the corresponding three spectral indices, and construct data sets DA1, DA2, and DA3 with the crude protein content, respectively. Fit the above data sets DA1, DA2, and DA3 with various fitting methods to obtain multiple fitting equations.

[0014] Step A6: The various fitting equations are evaluated using the coefficient of determination and root mean square error, and the best fitting equation, along with the corresponding preprocessing method and wavelength position, is selected as the detection model for crude protein content in chicken manure.

[0015] Furthermore, in one embodiment of the present invention, the various fitting methods mentioned in step A5 specifically refer to linear fitting, quadratic fitting, cubic fitting, logarithmic fitting, or exponential fitting.

[0016] Option 2: A system for constructing a model for detecting crude protein content in chicken manure, comprising the following modules:

[0017] Module A1 is used to obtain a sample set, which includes the visible-near-infrared reflectance spectrum and crude protein content of chicken manure;

[0018] Module A2 is used to preprocess the visible-near infrared reflectance spectrum using the MSC preprocessing method and the SNV preprocessing method respectively, to obtain the corresponding MSC preprocessed spectrum and SNV preprocessed spectrum;

[0019] Module A3 is used to obtain three corresponding spectral indices based on all wavelength positions in the visible-near infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum within the full spectrum wavelength range of 400nm to 900nm.

[0020] The three spectral indices mentioned include the ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI.

[0021] Module A4 is used to evaluate the correlation between the spectral index and the corresponding crude protein content of chicken manure, and to screen the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum respectively.

[0022] Module A5 is used to obtain three corresponding spectral indices based on the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum, and construct data sets DA1, DA2 and DA3 with the crude protein content, respectively. Multiple fitting methods are used to fit the above data sets DA1, DA2 and DA3 to obtain multiple fitting equations.

[0023] Module A6 is used to evaluate the various fitting equations using the coefficient of determination and root mean square error, and to select the best fitting equation, along with the corresponding preprocessing method and wavelength position, as a model for detecting crude protein content in chicken manure.

[0024] Option 3: A method for detecting the crude protein content in chicken manure, comprising the following steps:

[0025] Step B1: Collect and preprocess the chicken manure to be tested to obtain preprocessed chicken manure;

[0026] Step B2: Use a chicken manure spectral data acquisition device to acquire spectral data of the pretreated chicken manure and obtain the visible-near infrared reflectance spectrum of the pretreated chicken manure;

[0027] Step B3: Input the visible-near-infrared reflectance spectrum into the chicken manure crude protein content detection model for detection to obtain the chicken manure crude protein content;

[0028] The crude protein content detection model for chicken manure is the crude protein content detection model for chicken manure obtained by the method described in Scheme 1.

[0029] Option 4: An electronic device according to the present invention includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus.

[0030] Memory, used to store computer programs;

[0031] When the processor executes the program stored in the memory, it implements any of the above-described methods for constructing a model for detecting crude protein content in chicken manure.

[0032] Option 5: A computer-readable storage medium according to the present invention, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements any of the above-described methods for constructing a chicken manure crude protein content detection model.

[0033] This invention provides a model construction method, system, and detection method for detecting crude protein content in chicken manure. It solves the problems of high pollution during the detection process and severe information redundancy, high cost, and poor model stability when spectral analysis technology is directly applied to chicken manure crude protein detection. It achieves rapid, non-destructive, and low-cost quantitative detection of crude protein in chicken manure, improving the intelligent level of chicken manure resource utilization. Specific beneficial effects include:

[0034] 1. The model for detecting crude protein content in chicken manure obtained by the model construction method described in this invention can be applied to the crude protein detection technology of chicken manure, providing reliable technical support for fields such as chicken feed conversion efficiency, chicken manure fertilizer quality evaluation, and precision fertilization management.

[0035] Although methods for constructing spectral indices have been applied in plant nutrition and fruit and vegetable quality, given the differences between plants, fruits and vegetables, and livestock and poultry manure, these methods cannot be used to detect the protein content in livestock and poultry manure. Furthermore, existing technologies have not applied these methods to the rapid detection of nutrient components in complex biomass materials such as livestock and poultry manure. This invention, based on the biological characteristics of livestock and poultry manure, preprocesses the visible-near-infrared reflectance spectra using MSC and SNV preprocessing methods. After calculating the spectral indices for all wavelength positions of the above spectra, the optimal wavelength positions are selected using the correlation matrix method. Based on the optimal wavelength positions and spectral indices, a high-performance regression equation is constructed to achieve rapid, non-destructive, and low-cost quantitative detection of crude protein in chicken manure. This method is expected to provide reliable technical support for fields such as chicken feed conversion efficiency, chicken manure fertilizer quality evaluation, and precision fertilization management.

[0036] 2. The MSC preprocessing method and SNV preprocessing method used in this invention are selected considering the spectral characteristics and complexity of chicken manure samples. This invention constructs the subsequent fitting equation based on these two preprocessing methods. On the one hand, the process of constructing the fitting equation is optimized. On the other hand, these two preprocessing methods are applicable to the visible-near infrared reflectance spectral characteristics of chicken manure samples, improving the accuracy of the fitting equation and thus improving the detection accuracy of the detection model.

[0037] 3. The optimal fitting equation described in this invention is selected by calculating the spectral indices of characteristic bands (including ratio spectral index RI, difference spectral index DI, and normalized spectral index NDVI) and further combining them with the crude protein content of chicken manure for correlation evaluation. This effectively extracts spectral information that is highly correlated with the crude protein content of chicken manure, which can significantly reduce the data dimensionality and improve the model accuracy and robustness.

[0038] The method described in this invention is applicable to the detection of crude protein content in chicken manure using manure resources. Furthermore, this strategy has better interpretability and portability, making it suitable for integration into low-cost multispectral equipment and showing promising prospects for widespread application. Attached Figure Description

[0039] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0040] Figure 1 This is a flowchart of the method for constructing the detection equation for crude protein content in chicken manure as described in Implementation Method 1.

[0041] Figure 2 This is a schematic diagram of the spectral data acquisition device described in Embodiment 1, wherein: 1. Computer, 2. White reference plate, 3. Sample, 4. Blackboard, 5. Reflection probe, 6. Light source, 7. Spectrometer.

[0042] Figure 3 This is a graph showing the maximum value of the correlation coefficient between the original spectrum and the spectral index of the optimal pretreatment method described in Implementation Method 1 and the crude protein content of chicken manure, as well as the wavelength position.

[0043] Figure 4 The graph shows the results of fitting equations and evaluation results of the original spectrum and the spectral index of the optimal pretreatment method with the crude protein content of chicken manure as described in Implementation Method 1.

[0044] Figure 5 This is a graph showing the fitting effect between the spectral index of the visible-near-infrared reflectance spectrum described in Implementation Method 1 and the crude protein content of chicken manure. The horizontal axis represents the predicted value of the crude protein content of chicken manure, and the vertical axis represents the measured value of the crude protein content of chicken manure. Among them, (a) is the ratio spectral index; (b) is the difference spectral index; and (c) is the normalized spectral index.

[0045] Figure 6 This is a graph showing the fitting effect between the spectral index of the visible-near-infrared reflectance spectrum of the MSC pretreatment method described in Embodiment 1 and the crude protein content of chicken manure. The horizontal axis represents the predicted value of the crude protein content of chicken manure, and the vertical axis represents the measured value of the crude protein content of chicken manure. Among them, (a) is the ratio spectral index of the MSC pretreatment method; (b) is the difference spectral index of the MSC pretreatment method; and (c) is the normalized spectral index of the MSC pretreatment method.

[0046] Figure 7 This is a graph showing the fitting effect between the spectral index of the visible-near-infrared reflectance spectrum and the crude protein content of chicken manure in the SNV pretreatment method described in Implementation Method 1. The horizontal axis represents the predicted value of the crude protein content of chicken manure, and the vertical axis represents the measured value of the crude protein content of chicken manure. Among them, (a) is the ratio spectral index of the SNV pretreatment method; (b) is the difference spectral index of the SNV pretreatment method; and (c) is the normalized spectral index of the SNV pretreatment method. Detailed Implementation

[0047] Various embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. The embodiments described with reference to the drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0048] Implementation Method 1: The method for constructing a detection model for crude protein content in chicken manure described in this implementation method, such as... Figure 1 As shown, it includes the following steps:

[0049] Step A1: Obtain a sample set, which includes the visible-near infrared reflectance spectrum and crude protein content of chicken manure;

[0050] Step A2: The visible-near infrared reflectance spectrum is preprocessed using the MSC preprocessing method and the SNV preprocessing method respectively to obtain the corresponding MSC preprocessed spectrum and SNV preprocessed spectrum;

[0051] Step A3: Within the full-spectrum wavelength range of 400nm to 900nm, obtain the corresponding three spectral indices based on all wavelength positions in the visible-near-infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, respectively.

[0052] The three spectral indices mentioned include the ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI.

[0053] Step A4: Evaluate the correlation between the spectral index and the corresponding crude protein content of chicken manure, and screen the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum respectively.

[0054] Step A5: Based on the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, obtain the corresponding three spectral indices, and construct data sets DA1, DA2, and DA3 with the crude protein content, respectively. Fit the above data sets DA1, DA2, and DA3 with various fitting methods to obtain multiple fitting equations.

[0055] Step A6: The various fitting equations are evaluated using the coefficient of determination and root mean square error, and the best fitting equation, along with the corresponding preprocessing method and wavelength position, is selected as the detection model for crude protein content in chicken manure.

[0056] In this embodiment, the sample set is obtained in step A1 as follows:

[0057] Step A11: Collect chicken manure and pre-treat the collected chicken manure to obtain pre-treated chicken manure;

[0058] Step A12: Collect the visible-near infrared reflectance spectrum of the pretreated chicken manure;

[0059] In step A11, the collected chicken manure is pretreated, preferably by dividing the collected chicken manure samples into portions and sealing them in plastic bags within each group. 10 mL of 10% H2SO4 is added to each 100 g of manure sample for nitrogen fixation. After 4 days of treatment, each replicate chicken manure sample is thoroughly mixed and dried at 65°C for 72 hours, followed by rehydration for 24 hours. The dried manure sample is then pulverized to 40 mesh to ensure homogenization, yielding the chicken manure sample.

[0060] The process of collecting the visible-near-infrared reflectance spectrum of pretreated chicken manure in step A12 is as follows:

[0061] When acquiring visible-near-infrared reflectance spectra, the following methods are used: Figure 2 The apparatus shown is implemented by first connecting the optical fiber of the reflective probe 5 to the spectrometer 7 and the light source 6, and then fixing the reflective probe 5 with a reflective probe holder to ensure that the angle between the probe 5 and the surface of the sample 3 is 90°. After the light source is preheated for 8 minutes, whiteboard calibration is performed using the blackboard 4 and the white reference board 2, and the measurement results are displayed on the computer 1. During the experiment, whiteboard calibration is performed every 30 minutes. 50g of dried chicken manure from each group is taken as a sample and placed in a 90mm diameter petri dish. Spectral data is collected for each sample, and this spectral data and the corresponding crude protein content of the sample are used as sample data in the sample set.

[0062] In this embodiment, spectral data acquisition can be achieved using existing spectral data acquisition software tools, such as using AvaSoft 8 software (Aventes, Netherlands) in conjunction with a spectrometer to acquire spectral data and export the acquired spectral data. Preprocessing is preferably performed using The Unscrambler X 10.4 software (CAMO, Norway). The detection model for crude protein content in chicken manure is preferably constructed using Matlab 2023b software (MathWorks, USA). Plotting is preferably performed using Origin2021 software (OriginLab, USA).

[0063] In this embodiment, the crude protein content of chicken manure in the sample set can be obtained by detecting each sample using existing technology, for example, the Kjeldahl nitrogen determination method can be used, based on:

[0064]

[0065] Obtain crude protein content , in %, of which , The volume of standard hydrochloric acid solution consumed for titrating the blank, in mL; The volume of standard hydrochloric acid solution consumed for titrating the sample, in mL; The concentration of the hydrochloric acid standard titration solution is expressed in mol / L. The mass of the sample is expressed in grams. This represents the total volume of the sample decomposition solution, in mL. 14 represents the volume of sample decomposition solution used for distillation, in mL; 14 represents the molar mass of nitrogen, in g / mol; 6.25 represents the average coefficient for converting nitrogen to crude protein.

[0066] In this embodiment, the correlation evaluation described in step A4 can be achieved using the correlation matrix method, and the optimal wavelength position is selected by taking the wavelength position with the maximum correlation coefficient as the optimal wavelength position.

[0067] In this embodiment, step A3 is based on

[0068]

[0069]

[0070]

[0071] The ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI were obtained, among which... , Let be the spectral reflectance at any spectral position.

[0072] This embodiment describes a method for constructing an equation for detecting crude protein content in chicken manure. This method, through screening pretreatment methods, optimal wavelength indices, and fitting equations, obtains the best fitting equation, achieving rapid, non-destructive, and low-cost quantitative detection of crude protein in chicken manure. This method is expected to provide reliable technical support for fields such as chicken feed conversion efficiency, chicken manure fertilizer quality evaluation, and precision fertilization management.

[0073] In this embodiment, MSC and SNV preprocessing were performed on the visible-near infrared reflectance spectra, primarily considering the spectral characteristics and complexity of chicken manure samples. The spectral signal of chicken manure is easily affected by factors such as particle size, morphology, and compositional differences, especially fiber and mineral components, which can induce scattering effects and spectral biases. Furthermore, although drying reduces moisture content, differences in residual moisture between different samples or different parts of the same sample can still affect the degree of scattering. These physical factors may mask characteristic information related to crude protein content, reducing modeling accuracy. Therefore, employing appropriate preprocessing methods to eliminate these interferences is crucial.

[0074] MSCs eliminate scattering noise caused by factors such as particle size and density by correcting spectral data to a common reference spectrum, making data from different samples more consistent and reducing the interference of physical properties on the extraction of chemical information. Therefore, MSCs help reduce the impact of particle differences in chicken manure and improve the accuracy of quantitative analysis.

[0075] SNV, by independently centering and standardizing each spectral sample, removes the scattering effects caused by differences in optical path length and particle inhomogeneity, reducing the impact of intraspectral heterogeneity on modeling and improving data comparability. Therefore, SNV helps improve the comparability of spectra of chicken manure samples of different forms and the robustness of the model.

[0076] Therefore, this embodiment uses MSC and SNV to preprocess chicken manure spectral data to improve the accuracy and reliability of the chicken manure crude protein content detection model.

[0077] This implementation defines specific spectral indices, including the ratio spectral index (RI), the difference spectral index (DI), and the normalized spectral index (NDVI). The ratio spectral index (RI), by dividing the reflectance of two bands, effectively reduces the influence of environmental factors in the spectral data, enhancing sensitivity to target features. The difference spectral index (DI), by calculating the difference between two bands, highlights the magnitude of spectral reflectance changes, is sensitive to variations, is suitable for rapid feature extraction, and facilitates the establishment of quantitative models. The normalized spectral index (NDVI), through normalization processing, eliminates the influence of illumination and background interference, while limiting the threshold to the [-1, 1] range, enhancing the comparability between different samples and improving data stability.

[0078] This implementation method provides an example, with the following specific steps:

[0079] Step S1: Obtain a sample set, which includes the visible-near infrared reflectance spectrum and crude protein content of chicken manure;

[0080] The sample set is obtained in the following way:

[0081] Step S11: Pre-treat the collected chicken manure to obtain pre-treated chicken manure;

[0082] Specifically, the collected chicken manure samples were aliquoted into individual plastic bags within each group. 10 mL of 10% H₂SO₄ was added to each 100 g of manure sample for nitrogen fixation treatment. After 4 days of treatment, each replicate of chicken manure samples was thoroughly mixed and dried at 65°C for 72 hours, followed by rehydration for 24 hours. The dried manure samples were then pulverized to 40 mesh to ensure homogenization, thus obtaining the chicken manure samples.

[0083] Step S12: Collect the visible-near infrared reflectance spectrum of the pretreated chicken manure;

[0084] Specifically, when acquiring visible-near-infrared reflectance spectra, the following methods are used: Figure 2 The apparatus shown is implemented by first connecting the optical fiber of the reflective probe 5 to the spectrometer 7 and the light source 6, and then fixing the reflective probe 5 with a reflective probe bracket to ensure that the angle between the probe 5 and the surface of the sample 3 is 90°. After the light source is preheated for 8 minutes, whiteboard calibration is performed using the blackboard 4 and the white reference board 2, and the measurement is displayed on the computer 1. During the experiment, whiteboard calibration is performed every 30 minutes. 50g of dried chicken manure from each group is taken as a sample and placed in a 90mm diameter petri dish. Spectral data corresponding to each sample are collected, and this spectral data and the corresponding crude protein content of the sample are used as sample data in the sample set. Eight spectral data points are collected for each sample, resulting in a total of 120 spectral data points. The spectral information has 873 data points, forming a high-dimensional matrix of 873×120.

[0085] The crude protein content of chicken manure was determined using the Kjeldahl method, based on:

[0086]

[0087] Obtain crude protein content , in %, of which , The volume of standard hydrochloric acid solution consumed for titrating the blank, in mL; The volume of standard hydrochloric acid solution consumed for titrating the sample, in mL; The concentration of the hydrochloric acid standard titration solution is expressed in mol / L. The mass of the sample is expressed in grams. This represents the total volume of the sample decomposition solution, in mL. 14 represents the volume of sample decomposition solution used for distillation, in mL; 14 represents the molar mass of nitrogen, in g / mol; 6.25 represents the average coefficient for converting nitrogen to crude protein.

[0088] Step S2: The visible-near infrared reflectance spectrum is preprocessed using the MSC preprocessing method and the SNV preprocessing method respectively to obtain the corresponding MSC preprocessed spectrum and SNV preprocessed spectrum.

[0089] Step S3: Within the full-spectrum wavelength range of 400nm to 900nm, obtain the corresponding three spectral indices based on all wavelength positions in the visible-near-infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, respectively.

[0090] The three spectral indices mentioned above specifically include the ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI.

[0091] Step S4: Evaluate the correlation between the spectral index and the corresponding crude protein content of chicken manure, and screen the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum respectively.

[0092] In this embodiment, the correlation evaluation described in step S4 can be achieved using the correlation matrix method, and the optimal wavelength position is selected based on the wavelength position containing the maximum correlation coefficient. For example... Figure 3 The figure shows the maximum correlation coefficients and wavelength positions of the visible-near-infrared reflectance spectrum (denoted by R), the MSC preprocessed spectrum, and the SNV preprocessed spectrum with the crude protein content of chicken manure.

[0093] Step S5: Based on the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum, obtain the corresponding three spectral indices, and construct data groups A1, A2 and A3 with the crude protein content, respectively. Fit the above data groups A1, A2 and A3 with various fitting methods to obtain multiple fitting equations.

[0094] The various fitting methods mentioned above specifically refer to linear fitting, quadratic fitting, cubic fitting, logarithmic fitting, or exponential fitting.

[0095] Step S6: The multiple fitting equations are evaluated using the coefficient of determination and root mean square error, and the best fitting equation, along with the corresponding preprocessing method and wavelength position, is selected as the detection model for crude protein content in chicken manure.

[0096] In this embodiment, 80% of the spectral sample set is used as the training set to construct the fitting equation, and the remaining 20% ​​of the sample set is used as the model for detecting the crude protein content of chicken manure to verify the fitting equation.

[0097] The training set was used to construct the fitting equation, and the result is as follows: Figure 4The figure shows the fitting equations and evaluation results of the visible-near-infrared reflectance spectrum (denoted by R), the MSC preprocessed spectrum, and the SNV preprocessed spectrum with the crude protein content of chicken manure. The F-value is the statistic of the F-test, used to test the null hypothesis that "all independent variables have no combined effect on the dependent variable." A larger F-value indicates that the variance explained by the model is much greater than the random error, meaning the model is more significant overall. In curve estimation, the F-value helps in selecting the optimal model. It can be found that the ratio spectral index constructed by the visible-near-infrared reflectance spectrum has the best fitting effect on the crude protein content of chicken manure, with a coefficient of determination R0. 2 The coefficient of determination (R0.784) was 0.784, and the F-value was 111.233. Under both MSC and SNV pretreatment, the difference spectral index showed the best fitting effect against the crude protein content in chicken manure, with a coefficient of determination (R0.784). 2 Both are 0.769. The F values ​​are 101.953 and 101.974, respectively.

[0098] The test set was used to validate the model for detecting crude protein content in chicken manure, and the results are as follows: Figure 5 , Figure 6 and Figure 7 The figure shows the fitting effect of the visible-near-infrared reflectance spectrum, the MSC preprocessed spectrum, and the SNV preprocessed spectrum with the crude protein content of chicken manure. The fitting result of the ratio spectral index of visible-near-infrared reflectance spectrum with the crude protein content of chicken manure is R. 2 =0.8128, RMSE=0.4540; the fitting results of the difference spectral index-crude protein content of chicken manure under MSC and SNV pretreatment were respectively R 2 =0.7204 and 0.7098, RMSE=1.4835 and 0.5541.

[0099] Implementation Method 2: This implementation method further defines the method for constructing the detection equation for crude protein content in chicken manure described in Implementation Method 1. In this implementation method, the various fitting methods mentioned in step A5 specifically refer to linear fitting, quadratic fitting, cubic fitting, logarithmic fitting, or exponential fitting.

[0100] This embodiment further defines step A5, illustrating various fitting methods used in step A5. This embodiment employs linear fitting, quadratic fitting, cubic fitting, logarithmic fitting, or exponential fitting to fit the data, effectively capturing different patterns of data change: linear fitting is suitable for data exhibiting stable growth or decline trends, resulting in a simple and easily interpretable model; quadratic and cubic fitting can describe nonlinear changes and inflection point characteristics, suitable for simulating complex trends; logarithmic fitting excels at handling relationships where the growth rate slows down as the variable increases, while exponential fitting is suitable for data exhibiting rapid growth or decline. By comparing the goodness of fit of different fitting equations, the optimal fitting equation can be selected, ensuring both model accuracy and avoiding over-complexity, thus making predictions and interpretations more reliable.

Claims

1. A method for constructing a model for detecting crude protein content in chicken manure, characterized in that, Includes the following steps: Step A1: Obtain a sample set, which includes the visible-near infrared reflectance spectrum and crude protein content of chicken manure; Step A2: The visible-near infrared reflectance spectrum is preprocessed using the MSC preprocessing method and the SNV preprocessing method respectively to obtain the corresponding MSC preprocessed spectrum and SNV preprocessed spectrum; Step A3: Within the full-spectrum wavelength range of 400nm to 900nm, obtain the corresponding three spectral indices based on all wavelength positions in the visible-near-infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, respectively. The three spectral indices mentioned include the ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI. Step A4: Evaluate the correlation between the spectral index and the corresponding crude protein content of chicken manure, and screen the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum respectively. Step A5: Based on the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum, obtain the corresponding three spectral indices, and construct data sets DA1, DA2, and DA3 with the crude protein content, respectively. Fit the above data sets DA1, DA2, and DA3 with various fitting methods to obtain multiple fitting equations. Step A6: The various fitting equations are evaluated using the coefficient of determination and root mean square error, and the best fitting equation, along with the corresponding preprocessing method and wavelength position, is selected as the detection model for crude protein content in chicken manure.

2. The method for constructing a detection model for crude protein content in chicken manure according to claim 1, characterized in that, The various fitting methods mentioned in step A5 specifically refer to linear fitting, quadratic fitting, cubic fitting, logarithmic fitting, or exponential fitting.

3. A system for constructing a model for detecting crude protein content in chicken manure, characterized in that, Includes the following modules: Module A1 is used to obtain a sample set, which includes the visible-near-infrared reflectance spectrum and crude protein content of chicken manure; Module A2 is used to preprocess the visible-near infrared reflectance spectrum using the MSC preprocessing method and the SNV preprocessing method respectively, to obtain the corresponding MSC preprocessed spectrum and SNV preprocessed spectrum; Module A3 is used to obtain three corresponding spectral indices based on all wavelength positions in the visible-near infrared reflectance spectrum, MSC preprocessed spectrum, and SNV preprocessed spectrum within the full spectrum wavelength range of 400nm to 900nm. The three spectral indices mentioned include the ratio spectral index RI, the difference spectral index DI, and the normalized spectral index NDVI. Module A4 is used to evaluate the correlation between the spectral index and the corresponding crude protein content of chicken manure, and to screen the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum respectively. Module A5 is used to obtain three corresponding spectral indices based on the optimal wavelength positions of the visible-near infrared reflectance spectrum, MSC preprocessed spectrum and SNV preprocessed spectrum, and construct data sets DA1, DA2 and DA3 with the crude protein content, respectively. Multiple fitting methods are used to fit the above data sets DA1, DA2 and DA3 to obtain multiple fitting equations. Module A6 is used to evaluate the various fitting equations using the coefficient of determination and root mean square error, and to select the best fitting equation, along with the corresponding preprocessing method and wavelength position, as a model for detecting crude protein content in chicken manure.

4. A method for detecting the crude protein content in chicken manure, characterized in that, Includes the following steps: Step B1: Collect and preprocess the chicken manure to be tested to obtain preprocessed chicken manure; Step B2: Use a chicken manure spectral data acquisition device to acquire spectral data of the pretreated chicken manure and obtain the visible-near infrared reflectance spectrum of the pretreated chicken manure; Step B3: Input the visible-near-infrared reflectance spectrum into the chicken manure crude protein content detection model for detection to obtain the chicken manure crude protein content; The crude protein content detection model for chicken manure is the crude protein content detection model for chicken manure obtained by the method described in claim 1 or 2.

5. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor, when executing a program stored in memory, implements the method for constructing a chicken manure crude protein content detection model as described in claim 1 or 2.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for constructing a detection model for crude protein content in chicken manure as described in claim 1 or 2.

Images