Method and system for determining moisture content of podophyllum hexandrum seeds based on spectrum technology

Abstract

The invention relates to the technical field of water content determination, in particular to a podophyllum hexandrum seed water content determination method and system based on a spectrum technology. The method comprises the following steps: obtaining a water content characteristic wave band and a material characteristic wave band in each stage batch according to spectral response overall intensity and water content data of all seeds in different wave bands in each stage batch; obtaining the material similarity of the newly added seeds relative to the seeds in each stage batch according to the similarity between the spectral response overall intensity of the newly added seeds and the spectral response overall intensity of different seeds in each stage batch in the material characteristic wave band; constructing a water content regression equation according to the corresponding spectral response intensity and water content data of different seeds in the water content characteristic wave band in each stage batch; and further obtaining a water content prediction value of the newly-added seeds. According to the method, the similarity of spectral data among the seeds is considered, the characteristic wave band with relatively large water content contribution is selected for regression analysis, and the accuracy of measuring the water content of the seeds is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water content measurement, and specifically relates to a method and system for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology. Background Art

[0002] As seeds of a specific species, Sinopodophyllum hexandrum seeds have important application values in agricultural production and ecological research. With the continuous improvement of the requirements for seed quality and planting efficiency in modern agriculture, the water content of seeds has become an important factor affecting seed storage, germination, and growth and development. Excessive or too low water content of seeds will affect the long-term storage of seeds.

[0003] In the prior art, in the measurement method based on spectral technology, SPA is used to select characteristic bands related to water content changes. However, due to the biological characteristics and water distribution differences of seeds in different growth stages and different regions not being considered, the same characteristic bands are selected for seeds in different periods, resulting in poor accuracy of water content measurement. Summary of the Invention

[0004] In order to solve the technical problem of poor accuracy in water content measurement by analyzing the same characteristic bands without considering the differences of seeds in different growth stages and different regions, the purpose of the present invention is to provide a method and system for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology. The specific technical solutions adopted are as follows: The present invention proposes a method for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology, and the method includes: Obtaining hyperspectral data and water content data of Sinopodophyllum hexandrum seeds in different stage batches, where the hyperspectral data includes spectral response intensities at different positions in different bands; For the seeds in any stage batch, according to the relative distances between all spectral response intensities at different positions, obtaining the position clusters of each seed; according to the spectral response intensity distributions at different bands corresponding to all positions within the position cluster of each seed, obtaining the overall spectral response intensity of each seed in each band; According to the overall spectral response intensities of all seeds in different bands and the water content data in each stage batch, obtaining the water content characteristic bands and substance characteristic bands in each stage batch; according to the similarity of the overall spectral response intensities of new seeds and seeds in each stage batch in the substance characteristic bands, obtaining the substance similarity of new seeds relative to the seeds in each stage batch; Based on the spectral response intensities corresponding to the water content characteristic bands and the water content data of different seeds in each stage batch, a water content regression equation is constructed; based on the material similarity of the new seeds relative to the seeds in different stage batches, the overall spectral response intensity of the water content characteristic bands in the corresponding stage batches, and the water content regression equation, the predicted water content value of the new seeds is obtained.

[0005] Further, the method for obtaining the position clusters of each seed includes: Construct a vector by combining each position and the spectral response intensities at all positions corresponding to that position; obtain the Euclidean distance between the vectors at different positions as the relative distance between different positions; Perform K-means clustering on all positions according to the relative distances between different positions to obtain the position clusters of each seed.

[0006] Further, the method for obtaining the overall spectral response intensity includes: Calculate the mean value of the spectral response intensities corresponding to each identical band at all positions within the position cluster of each seed to obtain the overall spectral response intensity of each seed at each band.

[0007] Further, the method for obtaining the water content characteristic bands includes: Obtain the correlation coefficient between the sequence composed of the overall spectral response intensities of all seeds at each band and the sequence composed of the water content data in each stage batch as the first correlation coefficient; select the band corresponding to the maximum first correlation coefficient as the first water content characteristic band; Obtain the cumulative sum of the correlation coefficients between the sequences composed of the overall spectral response intensities of all seeds at each band and all water content characteristic bands in each stage batch as the second correlation coefficient; obtain the ratio of the first correlation coefficient and the second correlation coefficient corresponding to each band as the water content characteristic possibility of each band; Select the band with the maximum water content characteristic possibility value among all bands, and use the corresponding band as the next water content characteristic band; continuously analyze the water content characteristic possibility of each band to obtain a preset first number of water content characteristic bands.

[0008] Further, the method for obtaining the material characteristic bands includes: Select the band corresponding to the minimum first correlation coefficient as the first material characteristic band; Obtain the sum of the first correlation coefficient and the second correlation coefficient corresponding to each band as the material characteristic non-dominance of each band; Select the band with the minimum material characteristic non-dominance value among all bands, and use the corresponding band as the next material characteristic band; continuously analyze the material characteristic non-dominance of each band to obtain a preset second number of material characteristic bands.

[0009] Furthermore, the method for obtaining the material similarity includes: Obtain the relative distance of the sequence composed of the overall spectral response intensity of the newly added seeds and each seed in each stage batch under the material characteristic wavelength band, and perform negative correlation mapping, which is used as the local similarity between the newly added seeds and each seed in each stage batch; Obtain the cumulative value of the local similarities between the newly added seeds and different seeds in each stage batch, which is used as the material similarity of the newly added seeds relative to the seeds in each stage batch.

[0010] Furthermore, the method for obtaining the water content regression equation includes: Construct a linear regression equation with the overall spectral response intensity corresponding to the water content characteristic wavelength band of the seeds as the independent variable and the water content data of the corresponding seeds as the dependent variable; Substitute the spectral response intensity and water content data corresponding to the water content characteristic wavelength band of different seeds in each stage batch into the linear regression equation, and use the least squares method to obtain the parameters of the linear regression equation, thus forming the water content regression equation.

[0011] Furthermore, the method for obtaining the predicted water content value includes: Normalize the material similarity of the newly added seeds relative to the seeds in each stage batch, which is used as the water content weight of each stage batch; Weighted sum the water content regression equations of the corresponding stage batches according to the water content weights of different stage batches to obtain the water content prediction model of the newly added seeds; Substitute the overall spectral response intensity of the newly added seeds in each water content characteristic wavelength band into the water content prediction model, and use the model result as the predicted water content value of the newly added seeds.

[0012] Furthermore, the method of taking the reciprocal is used for negative correlation mapping.

[0013] The present invention also proposes a system for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of any one of the methods for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology are implemented.

[0014] The present invention has the following beneficial effects: For seeds in any stage batch, according to the relative distances of all spectral response intensities between different positions, the position clusters of each seed are obtained, and the spectral characteristics within the single-seed region are analyzed as a whole; according to the spectral response intensity distributions corresponding to different bands at all positions within the position cluster of each seed, the overall spectral response intensity of each seed in each band is obtained, which more comprehensively reflects the overall intensity of each seed; according to the overall spectral response intensities and water content data of all seeds in each stage batch at different bands, the water content characteristic bands and substance characteristic bands in each stage batch are obtained, and the bands with greater contributions from water and substances are identified, which is more helpful for evaluating the subsequent physiological states; according to the similarity of the overall spectral response intensities of the newly added seeds and the seeds in each stage batch in the substance characteristic bands, the substance similarity of the newly added seeds relative to the seeds in each stage batch is obtained, which reflects the stage batch that the newly added seeds are closest to and is more helpful for reference analysis; according to the spectral response intensity corresponding to the water content characteristic band and the water content data of different seeds in each stage batch, a water content regression equation is constructed, and the water content regression equation of each stage batch is evaluated, which more accurately analyzes the change of water content; according to the substance similarity of the newly added seeds relative to the seeds in different stage batches, the overall spectral response intensity in the water content characteristic band corresponding to the stage batch, and the water content regression equation, the water content prediction value of the newly added seeds is obtained. The present invention considers the similarity of spectral data between seeds, selects the characteristic bands with greater contributions from water content for regression analysis, and improves the accuracy of measuring the water content of seeds. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a flowchart of a method for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology provided by an embodiment of the present invention; Figure 2 It is a flowchart of a method for obtaining water content characteristic bands provided by an embodiment of the present invention; Figure 3 It is a flowchart of a method for obtaining water content prediction values provided by an embodiment of the present invention. Detailed Embodiments

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following combines the accompanying drawings and preferred embodiments to specifically describe a method and system for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology proposed according to the present invention, including its specific implementation manner, structure, characteristics and effects, as detailed below. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a method and system for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology provided by the present invention in combination with the accompanying drawings.

[0020] Please refer to Figure 1 , which shows the method flow chart of a method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology provided by an embodiment of the present invention, specifically including: Step S1: Obtain the hyperspectral data and water content data of seeds in different stage batches, where the hyperspectral data includes the spectral response intensities at different positions in different bands.

[0021] In the embodiment of the present invention, in order to accurately measure the water content of seeds and avoid ignoring the moisture distribution differences in different growth stages, the hyperspectral data and water content data of seeds are processed.

[0022] First, seeds are in different growth states such as the water absorption and swelling stage, the seed germination stage, and germination during different periods of germination. The hyperspectral data of the seeds is obtained using a hyperspectral imaging instrument; the water content of seeds in different stage batches can be measured by drying. The method is as follows: Dry the seeds in a constant temperature drying oven at 130 °C for 1 hour, take them out, cool and weigh, then continue to dry for 30 minutes, cool and weigh until the error between the two drying results is less than or equal to 0.002 g. Take the average value of the two weights as the final weight of the dried seeds, calculate the difference between the weight of the seeds before drying and the weight of the seeds after drying, and divide it by the weight of the seeds before drying to obtain the water content data of each seed. Therefore, the hyperspectral data and water content data of Sinopodophyllum hexandrum seeds in different stage batches are obtained, and the hyperspectral data includes the spectral response intensities at different positions in different bands.

[0023] It should be noted that in an embodiment of the present invention, a batch of seeds is obtained for analysis every 7 days as a stage, and 200 seeds are obtained for analysis in each stage batch; in other embodiments of the present invention, the implementer can set according to specific situations, which will not be limited and elaborated here.

[0024] Step S2: For the seeds of any stage batch, according to the relative distances of all spectral response intensities between different positions, obtain the position clusters of each seed; according to the spectral response intensity distributions corresponding to all positions within the position clusters of each seed at different bands, obtain the overall spectral response intensity of each seed at each band.

[0025] As the band changes, the spectral response degrees in the non-seed area and the area corresponding to the seeds vary differently. The spectral characteristics of the non-seed area, i.e., the background area, are consistent. By analyzing the relative distances of all spectral response intensities between different positions, positions with similar characteristics and close distances can be clustered together, which helps to analyze each seed specifically; for the seeds of any stage batch, according to the relative distances of all spectral response intensities between different positions, obtain the position clusters of each seed.

[0026] Preferably, in an embodiment of the present invention, the method for obtaining the position clusters of each seed includes: Construct a vector for each position and all spectral response intensities at the corresponding position; obtain the Euclidean distance between vectors at different positions as the relative distance between different positions; According to the relative distances between different positions, perform K-means clustering on all positions to obtain the position clusters of each seed.

[0027] It should be noted that the K-means clustering algorithm aggregates multiple clustering objects into the specified K clustering clusters according to the similarity between the clustering objects. Each clustering object belongs to and only belongs to one clustering cluster with the smallest distance to the center of the position cluster; in an embodiment of the present invention, in order to divide each seed into a separate cluster for analysis, the number of clustering clusters is set to the sum of the number of seeds and the number of non-seed areas, cluster the positions of each seed into one cluster, and cluster the positions of the non-seed areas into one cluster; the specific K-means clustering algorithm is a technical means well-known to those skilled in the art and will not be elaborated here.

[0028] The spectral response intensity of a single seed at each wavelength under different positions may be different. In order to perform an overall analysis of the seeds and reflect the overall spectral characteristics of the seeds, according to the spectral response intensity distributions corresponding to all positions within the position clusters of each seed at different bands, obtain the overall spectral response intensity of each seed at each band.

[0029] Preferably, in an embodiment of the present invention, the method for obtaining the overall spectral response intensity includes: Take the mean value of the spectral response intensities corresponding to all positions within the position clusters of each seed at each same band to obtain the overall spectral response intensity of each seed at each band.

[0030] Step S3: According to the overall spectral response intensity and water content data of all seeds in each stage batch at different wavelengths, obtain the water content characteristic wavelength band and substance characteristic wavelength band for each stage batch; according to the similarity of the overall spectral response intensity of the newly added seeds and the seeds in each stage batch at the substance characteristic wavelength band, obtain the substance similarity of the newly added seeds relative to the seeds in each stage batch.

[0031] Since the growth states of seeds in different stages are different, there are multiple wavelength bands of substances unrelated to water content. By analyzing the overall spectral response intensity and water content data of different wavelength bands, select the characteristic wavelength band with a greater contribution to water content, as well as the wavelength band situation of the substances exhibited by the seeds in each stage batch, to reflect the correlation between the overall spectral response intensity of different wavelength bands and the change of water content data, as well as the germination situation of the seeds; according to the overall spectral response intensity and water content data of all seeds in each stage batch at different wavelengths, obtain the water content characteristic wavelength band and substance characteristic wavelength band for each stage batch.

[0032] Preferably, in an embodiment of the present invention, for the method of obtaining the water content characteristic wavelength band, please refer to Figure 2 , which shows a flowchart of a method for obtaining a water content characteristic wavelength band, including: Step S201: Obtain the correlation coefficient between the sequence composed of the overall spectral response intensity of all seeds in each stage batch at each wavelength and the sequence composed of the water content data, as the first correlation coefficient; select the wavelength corresponding to the maximum first correlation coefficient as the first water content characteristic wavelength band.

[0033] For the subsequent determination of water content, by analyzing the correlation coefficient between the sequence composed of the overall spectral response intensity and the sequence composed of the water content data, the greater the correlation coefficient, the more likely it is a characteristic wavelength band with a greater contribution to the water content.

[0034] Take an example, the sequence composed of the overall spectral response intensity at wavelength is , and the sequence composed of the water content data is , where represents the overall spectral response intensity of the th seed at wavelength , and represents the water content data of the th seed.

[0035] It should be noted that in the embodiments of the present invention, the correlation coefficient is the Pearson correlation coefficient. The larger the correlation coefficient, the more correlated the changes in the sequences composed of the overall spectral responses of each band and the sequences composed of water content data, and the greater the influence of the overall spectral response on the water content data. The specific Pearson correlation coefficient is a well-known technical means to those skilled in the art and will not be elaborated here.

[0036] Step S202: Obtain the sum of the correlation coefficients of the sequences composed of the overall spectral responses between each band and all water content characteristic bands for all seeds in each stage batch, as the second correlation coefficient; obtain the ratio of the corresponding first correlation coefficient to the second correlation coefficient for each band, as the water content characteristic possibility of each band.

[0037] To avoid being overly correlated with the spectral response degree under the already obtained water content characteristic bands, and the correlation between the spectral response degree of the band and the water content cannot be too low to ensure that it makes a certain contribution to the change in water content; by analyzing the correlation coefficients of the sequences composed of the overall spectral responses between each band and all water content characteristic bands, the smaller the correlation coefficient, the more inconsistent the performance characteristics of the band and the water content characteristic bands, and the larger the correlation coefficient between the spectral response degree of the band and the water content, the more it can represent the characteristics of the water content, the greater the water content characteristic possibility of the corresponding band, and the more conducive to the determination and analysis of water content.

[0038] Step S203: Select the band with the largest water content characteristic possibility value among all bands, and use the corresponding band as the next water content characteristic band; continuously analyze the water content characteristic possibility of each band to obtain a preset first number of water content characteristic bands.

[0039] In an embodiment of the present invention, the formula for the water content characteristic possibility is expressed as: ; Among them, represents the water content characteristic possibility of band ; represents the correlation coefficient between the sequences composed of the overall spectral responses of band and the sequences composed of water content data, that is, the first correlation coefficient; represents the sequence composed of the overall spectral response of band ; represents the th water content characteristic band and the sequence composed of the overall spectral response degree; represents band and the th water content characteristic band The correlation coefficient of the sequence composed of the overall intensity of spectral responses between them; Indicates the number of characteristic bands of water content.

[0040] In the formula for the possibility of water content characteristics, Adding 0.01 is to avoid the denominator of the formula being 0, making the formula meaningless; Indicates the band and the cumulative sum of the correlation coefficients of the sequences composed of the overall intensity of spectral responses between different characteristic bands of water content, that is, the second correlation coefficient. The smaller the cumulative sum, the smaller the spectral feature correlation between the band and different characteristic bands of water content, and the greater the correlation coefficient between the sequence composed of the overall intensity of the spectral response of the band and different characteristic bands of water content The smaller the spectral feature correlation between them, the greater the influence on the water content for the band The greater the correlation coefficient between the sequence composed of the overall intensity of the spectral response and the sequence composed of water content data, the greater the influence on the water content. The newer the band, the greater the contribution to the change in water content, and the greater the possibility of water content characteristics.

[0041] It should be noted that since there are many bands in Sinopodophyllum hexandrum seeds that are irrelevant to the water content, in order to more accurately reflect the characteristics of water content, in an embodiment of the present invention, the preset first quantity is set to 5, that is, after sequentially selecting each characteristic band of water content, the water content characteristics of each band are analyzed to select the next one until the number of characteristic bands of water content reaches the preset first quantity; in other embodiments of the present invention, the size of the preset first quantity can be specifically set according to specific circumstances, which will not be limited and elaborated here.

[0042] Preferably, in an embodiment of the present invention, the method for obtaining the characteristic bands of substances includes: Select the band corresponding to the minimum first correlation coefficient as the first characteristic band of the substance; Obtain the sum of the first correlation coefficient and the second correlation coefficient corresponding to each band as the non-dominance of the substance characteristics of each band; Select the one with the smallest non-dominance value of substance characteristics among all bands, and use the corresponding band as the next characteristic band of the substance; continuously analyze the non-dominance of the substance characteristics of each band to obtain the preset second quantity of characteristic bands of the substance.

[0043] In an embodiment of the present invention, the formula for the non-dominance of substance characteristics is expressed as: ; Wherein, Indicates the non-dominance of the substance characteristics at the wavelength ; Indicates the correlation coefficient between the sequence composed of the overall intensity of the spectral response of the band and the sequence composed of water content data, that is, the first correlation coefficient; Represents the sequence composed of the overall intensity of the spectral response of the band; Represents the th material characteristic band The sequence composed of the overall degree of the spectral response; Represents the band and the th material characteristic band The correlation coefficient of the sequence composed of the overall intensity of the spectral response between them; Represents the number of material characteristic bands.

[0044] In the formula where the material characteristics are not obvious, Represents the band and the th material characteristic band The sum of the correlation coefficients of the sequences composed of the overall intensity of the spectral response between them, that is, the second correlation coefficient. The smaller the sum, the smaller the spectral feature correlation between the band and the th material characteristic band is, and the more likely it is to show the characteristics of different substances; the smaller the correlation coefficient between the sequence composed of the overall intensity of the spectral response of the band and the sequence composed of the water content data, the smaller the contribution to the correlation of the water content change, the more likely it is the manifestation of other substance components, and the smaller the material characteristic non - dominance.

[0045] It should be noted that in order to analyze the characteristics of seed germination in each stage batch and pay attention to the changes in the remaining substance components in the seeds, the more the substance components are manifested, the more helpful it is to analyze the characteristics of seed germination. In an embodiment of the present invention, the preset second quantity is 9; in other embodiments of the present invention, the preset second quantity can be specifically set according to specific situations and will not be limited and elaborated here.

[0046] The material characteristic band reflects the germination situation of seeds in the stage batch. By analyzing the similarity of the overall intensity of the spectral response of the material characteristic band between the newly added seeds and the seeds in each stage batch, the material similarity of the newly added seeds relative to the seeds in each stage batch is evaluated, which is beneficial to the analysis based on the seed situation of the corresponding stage batch. The material similarity of the newly added seeds relative to the seeds in each stage batch is obtained according to the similarity of the overall intensity of the spectral response of the material characteristic band between the newly added seeds and the seeds in each stage batch.

[0047] Preferably, in an embodiment of the present invention, the method for obtaining the material similarity includes: ​Obtain the relative distance of the sequence composed of the overall spectral response intensity between the newly added seeds and each seed under each stage batch in the material characteristic wavelength band, and perform negative correlation mapping, which is used as the local similarity between the newly added seeds and each seed under each stage batch. Obtain the cumulative value of the local similarities between the newly added seeds and different seeds under each stage batch, which is used as the material similarity of the newly added seeds relative to the seeds under each stage batch.

[0048] It should be noted that the overall spectral response intensity of the newly added seeds in each wavelength band is analyzed according to the same acquisition method steps as the overall spectral response intensity of each seed in each wavelength band in step S2; in the embodiments of the present invention, the relative distance can be calculated by existing distance algorithms such as Euclidean distance or Manhattan distance, and the specific means are well-known technical means to those skilled in the art and will not be elaborated here.

[0049] It should be noted that in an embodiment of the present invention, negative correlation mapping is performed on the relative distance of the spectral response intensity vector by taking the reciprocal. The greater the relative distance, the smaller the material similarity, and the less likely the seeds are in the same germination state. When taking the reciprocal, a threshold is artificially added to avoid the denominator of the formula being 0 and the formula being meaningless; in other embodiments of the present invention, negative correlation mapping can also be performed through and the specific means are well-known technical means to those skilled in the art and will not be elaborated here.

[0050] Step S4: Construct a water content regression equation according to the spectral response intensity and water content data corresponding to different seeds under each stage batch in the water content characteristic wavelength band; obtain the water content prediction value of the newly added seeds according to the material similarity of the newly added seeds relative to the seeds under different stage batches, the overall spectral response intensity in the water content characteristic wavelength band corresponding to the corresponding stage batch, and the water content regression equation.

[0051] Seeds in different stage batches have different water spectral relationships. By establishing a water content regression equation corresponding to the water content characteristic wavelength band, it is more conducive to evaluating the oil content of seeds under each stage batch and reflecting the germination state.

[0052] Construct a water content regression equation according to the spectral response intensity and water content data corresponding to different seeds under each stage batch in the water content characteristic wavelength band.

[0053] Preferably, in an embodiment of the present invention, the acquisition method of the water content regression equation includes: Construct a linear regression equation with the overall spectral response intensity corresponding to the seeds in the water content characteristic wavelength band as the independent variable and the water content data of the corresponding seeds as the dependent variable; where the linear regression equation is , where represents the water content data. represents the intercept term; represents the variable parameter of the water content characteristic band ; represents the overall spectral response intensity of the first water content characteristic band ; represents the variable parameter of the water content characteristic band ; represents the th water content characteristic band ; represents the random component; Substitute the spectral response intensities corresponding to the water content characteristic bands and the water content data of different seeds in each stage batch into the linear regression equation, and use the least squares method to obtain the parameters of the linear regression equation, thus forming the water content regression equation.

[0054] It should be noted that the specific method for solving the parameters by the least squares method is a well-known technical means to those skilled in the art and will not be elaborated here.

[0055] The material similarity reflects the similarity of the growth states between the newly added seeds and different stage batches, which helps to match the most similar stage batch of the newly added seeds. The larger the proportion of the corresponding regression equation, the more comprehensive and accurate the analysis of the water content characteristics of the newly added seeds; according to the material similarity of the newly added seeds relative to the seeds in different stage batches, the overall spectral response intensity of the water content characteristic band in the corresponding stage batch, and the water content regression equation, the water content prediction value of the newly added seeds is obtained.

[0056] Preferably, in an embodiment of the present invention, for the method of obtaining the water content prediction value, please refer to Figure 3 , which shows a flowchart of the method for obtaining the water content prediction value, including: Step S301: Normalize the material similarity of the newly added seeds relative to the seeds in each stage batch as the water content weight of each stage batch.

[0057] The material similarity reflects the degree of similarity of the characteristics of the newly added seeds with the seeds in which stage batch. The greater the degree of similarity, the more helpful it is to analyze the water content of the newly added seeds, and the greater the water content weight.

[0058] Step S302: Perform weighted summation on the water content regression equations of the corresponding stage batches according to the water content weights of different stage batches to obtain the water content prediction model of the newly added seeds.

[0059] In an embodiment of the present invention, the formula of the water content prediction model is expressed as: ; where, Represents the predicted water content value of the newly added seeds; Represents the substance similarity of the batch at the Represents the water content regression equation of the batch at the Represents the normalization function; Represents the number of batches at the stage.

[0060] In the formula of the water content prediction model, Represents the normalization of the substance similarity of the batch at the stage, that is, the water content weight. The greater the substance similarity, the more consistent the germination situation of the newly added seeds with the seeds under this stage batch, which is more conducive to the prediction of the water content.

[0061] Step S303: Substitute the overall spectral response intensity of the newly added seeds in each water content characteristic band into the water content prediction model, and use the model result as the predicted water content value of the newly added seeds.

[0062] Based on this, by assigning a larger weight to the stage batch with a greater substance similarity to the newly added seeds, the prediction deviation caused by the difference in the germination stage is avoided, which is more conducive to accurately analyzing the water content of the newly added seeds and obtaining the moisture information more quickly and non-destructively.

[0063] In summary, the present invention obtains the water content characteristic band and the substance characteristic band under each stage batch according to the overall spectral response intensity and water content data of all seeds under each stage batch at different bands; obtains the substance similarity of the newly added seeds relative to the seeds under each stage batch according to the similarity of the overall spectral response intensity of the newly added seeds and the seeds under each stage batch in the substance characteristic band; constructs a water content regression equation according to the corresponding spectral response intensity and water content data of different seeds under each stage batch in the water content characteristic band; and then obtains the predicted water content value of the newly added seeds. The present invention considers the similarity of spectral data between seeds, selects the characteristic bands with a greater contribution to the moisture content for regression analysis, and improves the accuracy of measuring the moisture content of seeds.

[0064] The present invention also proposes a system for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of any one of the methods for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology are implemented.

[0065] It should be noted that the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology, characterized in that, The method includes: Obtaining the hyperspectral data and water content data of Sinopodophyllum hexandrum seeds in different stage batches, where the hyperspectral data includes the spectral response intensities at different positions in different bands; For the seeds of any stage batch, according to the relative distances of all spectral response intensities between different positions, obtaining the position clusters of each seed; according to the spectral response intensity distributions at different bands corresponding to all positions within the position cluster of each seed, obtaining the overall spectral response intensity of each seed at each band; According to the overall spectral response intensities of all seeds at different bands and the water content data in each stage batch, obtaining the water content characteristic bands and substance characteristic bands in each stage batch; according to the similarity of the overall spectral response intensities of the new seeds and the seeds in each stage batch at the substance characteristic bands, obtaining the substance similarity of the new seeds relative to the seeds in each stage batch; According to the spectral response intensities corresponding to the water content characteristic bands and the water content data of different seeds in each stage batch, constructing a water content regression equation; according to the substance similarity of the new seeds relative to the seeds in different stage batches, the overall spectral response intensity at the water content characteristic bands in the corresponding stage batches, and the water content regression equation, obtaining the water content prediction value of the new seeds.

2. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 1, wherein, The method for obtaining the position cluster of each seed includes: Forming vectors by each position and all spectral response intensities at the corresponding position; obtaining the Euclidean distance between vectors at different positions as the relative distance between different positions; According to the relative distances between different positions, performing K-means clustering on all positions to obtain the position clusters of each seed.

3. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 1, characterized in that, The method for obtaining the overall spectral response intensity includes: Calculating the mean value of the spectral response intensities at each same band corresponding to all positions within the position cluster of each seed to obtain the overall spectral response intensity of each seed at each band.

4. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 1, characterized in that, The method for obtaining the water content characteristic bands includes: Obtaining the correlation coefficient between the sequence composed of the overall spectral response intensities of all seeds at each band and the sequence composed of the water content data in each stage batch as the first correlation coefficient; selecting the band corresponding to the maximum first correlation coefficient as the first water content characteristic band; Obtaining the cumulative sum of the correlation coefficients between the sequence composed of the overall spectral response intensities of all seeds at each band and all water content characteristic bands as the second correlation coefficient; obtaining the ratio of the first correlation coefficient and the second correlation coefficient corresponding to each band as the water content characteristic possibility of each band; Selecting the band with the maximum water content characteristic possibility value among all bands as the next water content characteristic band; continuously analyzing the water content characteristic possibility of each band to obtain a preset first number of water content characteristic bands.

5. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 4, wherein The method for obtaining the substance characteristic bands includes: Selecting the band corresponding to the minimum first correlation coefficient as the first substance characteristic band; Obtaining the sum of the first correlation coefficient and the second correlation coefficient corresponding to each band as the substance characteristic non-dominance of each band; Select the one with the smallest value of the non-dominant material feature in all bands, and use the corresponding band as the next material feature band; continuously analyze the non-dominance of the material feature of each band to obtain a preset second number of material feature bands.

6. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 1, characterized in that, The method for obtaining the material similarity includes: Obtain the relative distance of the sequence composed of the overall spectral response intensity of the new seed and each seed in each stage batch under the material feature band, and perform negative correlation mapping, which is used as the local similarity between the new seed and each seed in each stage batch; Obtain the cumulative value of the local similarities between the new seed and different seeds in each stage batch, which is used as the material similarity of the new seed relative to the seeds in each stage batch.

7. A method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 1, characterized in that The method for obtaining the water content regression equation includes: Construct a linear regression equation with the overall spectral response intensity corresponding to the water content feature band of the seed as the independent variable and the water content data of the corresponding seed as the dependent variable; Substitute the spectral response intensity and water content data corresponding to different seeds in each stage batch under the water content feature band into the linear regression equation, and use the least squares method to obtain the parameters of the linear regression equation, thus forming the water content regression equation.

8. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectroscopy according to claim 1, characterized in that, The method for obtaining the predicted water content value includes: Normalize the material similarity of the new seed relative to the seeds in each stage batch, which is used as the water content weight of each stage batch; Perform weighted summation on the water content regression equations of the corresponding stage batches according to the water content weights of different stage batches to obtain the water content prediction model of the new seed; Substitute the overall spectral response intensity of the new seed in each water content feature band into the water content prediction model, and use the model result as the predicted water content value of the new seed.

9. The method for measuring the moisture content of Sinopodophyllum hexandrum seeds based on spectral technology according to claim 6, wherein Use the method of taking the reciprocal for negative correlation mapping.

10. A determination system for the moisture content of Sinopodophyllum hexandrum seeds based on spectroscopic technology, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for measuring the water content of Sinopodophyllum hexandrum seeds based on spectral technology according to any one of claims 1 to 9.

Images