An intelligent screening system and method for peony offspring resistant to humidity and heat based on data analysis

Abstract

The invention discloses an intelligent screening system and method for peony moisture and heat resistance progeny based on data analysis, and relates to the field of data screening and analysis. The invention classifies and evaluates the moisture and heat resistance of peony progeny samples to be screened from three perspectives: peony varieties, peony biological characteristics, and peony genes, so that the classification and screening processes are more comprehensive and accurate. A first final multi-layer perceptron classification model is constructed to provide a classification tool for subsequent classification of characteristic data of peony progeny samples to be screened. Classification center data of peony characterization data sets of each category in a peony characterization data set matrix are obtained and optimized to provide a classification basis for subsequent classification of characterization data of peony progeny samples to be screened. A second final multi-layer perceptron classification model is constructed to provide a classification tool for subsequent classification of genotype sequence data of peony progeny samples to be screened.

Description

Technical Field

[0001] The present invention belongs to the field of data screening and analysis, and in particular, relates to a system and method for intelligent screening of peony moisture-heat-resistant offspring based on data analysis. Background Art

[0002] Most existing screening methods for peony offspring that are resistant to moisture and heat rely on manual observation and data analysis, which is inefficient and requires a lot of time and resources to screen out ideal moisture and heat resistant offspring from a large number of candidate varieties. In addition, affected by human subjectivity and the cognitive limitations of screening personnel, the screening standards are inconsistent and incomplete, which in turn leads to inaccurate screening results. Summary of the invention

[0003] In view of the problems in the related art, the present invention proposes a system and method for intelligent screening of peony offspring resistant to moisture and heat based on data analysis to overcome the above-mentioned technical problems existing in the existing related art.

[0004] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0005] The present invention is a method for intelligently screening peony offspring resistant to humidity and heat based on data analysis, comprising the following steps:

[0006] S1, collecting multiple characteristic type data and moisture and heat resistance grade data that can reflect the peony varieties of multiple peony samples and generating corresponding type labels to obtain a characteristic data set matrix, a moisture and heat resistance grade data matrix and a type label data matrix;

[0007] S2, using the feature data set matrix and the type label data matrix to construct a first final multi-layer perceptron classification model;

[0008] S3, collecting multiple moisture and heat resistance trait characterization type data of multiple peony samples to obtain a peony characterization data set matrix; obtaining and optimizing the classification center data of each category of the peony characterization data set in the peony characterization data set matrix to obtain a final classification center data matrix;

[0009] S4, collecting the moisture and heat resistance genotype sequence data of multiple peony samples and building the second final multi-layer perceptron classification model in conjunction with the moisture and heat resistance grade data matrix;

[0010] S5, collect data of multiple peony offspring samples to be screened, obtain a feature data matrix to be screened, a peony characterization data matrix to be screened, and a genotype sequence data set to be screened; use the first final multi-layer perceptron classification model, the final classification center data matrix, and the second final multi-layer perceptron classification model to classify and compare the feature data matrix to be screened, the peony characterization data matrix to be screened, and the genotype sequence data set to be screened, respectively, to obtain a classification result;

[0011] S6. Screening a plurality of peony offspring samples to be screened according to the classification results to obtain a screened peony offspring sample set;

[0012] Different peony varieties have different resistance to moisture and heat. Therefore, in this scheme, multiple characteristic type data that can reflect the peony varieties are first collected from multiple peony samples, providing data support for the subsequent construction of a classification model that can identify peony varieties; by generating type label data corresponding to each feature data set in the feature data set matrix, the type label data represents the variety data of the corresponding peony, providing classification label data for the subsequent construction of a classification model that can identify peony varieties; by constructing the first and final multi-layer perceptron classification model, a classification tool is provided for the subsequent classification of the characteristic data of the peony offspring samples to be screened; in addition to the great influence of peony varieties on the moisture and heat resistance of peonies, some biological characteristics of peonies themselves also have a great influence on the moisture and heat resistance of peonies; therefore, by collecting multiple moisture and heat resistance trait characterization type data from multiple peony samples, the moisture and heat resistance of peonies can be analyzed from the perspective of the biological characteristics of peonies themselves. The data basis is provided; by obtaining and optimizing the classification center data of each category of the peony characterization data set in the peony characterization data set matrix, a classification basis is provided for the subsequent classification of the characterization data of the peony offspring samples to be screened; in addition, the gene sequence of peony is the fundamental reason affecting the moisture and heat resistance of peony. Therefore, by collecting the moisture and heat resistance genotype sequence data of multiple peony samples, data support is provided for the subsequent classification of the moisture and heat resistance of peony from a genetic perspective; by constructing the second final multi-layer perceptron classification model, a classification tool is provided for the subsequent classification of the genotype sequence data of the peony offspring samples to be screened; finally, the constructed first final multi-layer perceptron classification model, the final classification center data matrix and the second final multi-layer perceptron classification model are used to classify and evaluate the moisture and heat resistance of the peony offspring samples to be screened from three perspectives: peony variety, peony biological characteristics and peony genes, making the classification process more comprehensive and accurate.

[0013] Preferably, the S1 comprises the following steps:

[0014] S11. Set multiple peony types to obtain a peony type set , Indicates the settingi Types of Peonies, Indicates the total number of set peony types; according to the peony type set Collect peony samples and obtain peony sample matrix ;as follows,

[0015] ;

[0016] in, Represents the set of peony types Middle i The first type of peony collected j peony samples; Represents the set of peony types The total number of peony samples collected for each peony type;

[0017] Set the moisture and heat resistance level set , Indicates the setting i Moisture and heat resistance level, Indicates the total number of moisture and heat resistance levels set; Peony Sample Matrix Each peony sample is subjected to a water flooding test for the same time; after the test is completed, the moisture and heat resistance grade data of each peony sample is determined to obtain a moisture and heat resistance grade data matrix ;as follows,

[0018] ;

[0019] in, Represents the set of peony types Middle i The first type of peony collected j The moisture and heat resistance rating data of peony samples;

[0020] Calculate the moisture and heat resistance grade data matrix The average value of each row of data in the , to obtain the average grade of moisture and heat resistance of varieties , Represents the peony type set Middle i Average ratings of moisture and heat resistance of peony types;

[0021] S12, obtaining the peony sample matrix The image data of each peony sample in the image matrix is ​​obtained b ;as follows,

[0022] ;

[0023] in, Represents the set of peony types Middle i The first type of peony collected j Image data of peony samples;

[0024] Set multiple feature types that can reflect peony varieties and obtain a feature type set , Indicates the setting i can reflect the characteristic types of peony varieties. Indicates the total number of characteristic types that can reflect peony varieties;

[0025] According to the feature type set The pre-trained CNN neural network model is used to extract the peony image data matrix b The feature data set matrix is ​​obtained by taking the feature data of each peony image data in ;as follows,

[0026] ;

[0027] in, Indicates for The collected feature datasets, , express No. k Feature data of feature types;

[0028] S13, matching the peony type set Set the feature dataset matrix The type label of each feature data set in the matrix is ​​obtained c ;as follows,

[0029] ;

[0030] in, express The corresponding type tag data;

[0031] The types of peony include Wei Zi, Yao Huang, Zhao Fen, Luoyang Hong, Er Qiao, Jin Pao Hong, Zhu Sha Lei and Bai Xue Ta, etc.; the levels of humidity and heat resistance include, for example, level 1: waterlogging resistant, level 2: relatively waterlogging resistant, level 3: relatively intolerant, level 4: intolerant

[0032] And level 5: extremely intolerant to waterlogging, etc. Through flooding tests, the leaf color, leaf shape, stem and other characteristics of peony varieties under waterlogging stress were evaluated.

[0033] The corresponding moisture and heat resistance grade data were obtained by calculating the degree of change in phenotypic characteristics such as flower color and stem morphology; by calculating the average moisture and heat resistance grade data of each type of peony, the randomness of the data was reduced and the accuracy of the data was improved; the feature type set included flower color, flower shape, number of petals and morphology.

[0034] Preferably, S2 comprises the following steps:

[0035] S21, setting a first training data ratio; and performing a feature data set matrix on the basis of the first training data ratio. and the type label data matrix c Perform data partitioning to obtain a feature training data set matrix, a feature test data set matrix, a type label training data matrix, and a type label test data matrix;

[0036] S22, constructing a first initial multi-layer perceptron classification model and setting a first training error threshold; inputting the feature training data set matrix and the type label training data matrix into the first initial multi-layer perceptron classification model for training; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain a first trained multi-layer perceptron classification model; otherwise, continuing the training;

[0037] S23, setting a first test accuracy threshold; inputting the feature test data set matrix and the type label test data matrix into the first trained multi-layer perceptron classification model for testing; after the test is completed, obtaining a first test accuracy; when the first test accuracy is greater than or equal to the first test accuracy threshold, using the first trained multi-layer perceptron classification model as the first final multi-layer perceptron classification model; otherwise, returning to S22 to continue training until the first test accuracy is greater than or equal to the first test accuracy threshold, obtaining the first final multi-layer perceptron classification model;

[0038] The first initial multilayer perceptron classification model is trained and tested using a feature training data set matrix, a type label training data matrix, a feature test data set matrix, and a type label test data matrix, respectively, so that the obtained first final multilayer perceptron classification model has a good classification ability for the peony sample feature data.

[0039] Preferably, S3 comprises the following steps:

[0040] S31, setting multiple moisture and heat resistance trait characterization types of peony samples, and obtaining a characterization type set , Indicates the setting i Types of moisture and heat resistance characterization, Indicates the total number of moisture and heat resistance trait characterization types set; according to the characterization type set Collect the peony sample matrix The moisture and heat resistance trait characterization data of each peony sample in the peony characterization data set matrix is ​​obtained. ;as follows,

[0041] ;

[0042] in, Represents the set of peony types Middle i The first type of peony collected j Characterization dataset of peony samples; , express The k Characterization data corresponding to the characterization type of moisture and heat resistance;

[0043] S32, according to the moisture and heat resistance grade data matrix Matrix for Peony Characterization Dataset The characterization data in is divided into categories to obtain the characterization data classification matrix set , Indicates the divided i A matrix representing data classification; as follows,

[0044] ;

[0045] in, express Middle j The first k Characterization data corresponding to the characterization type of moisture and heat resistance; express The total number of representation data sets in ;

[0046] S33, obtaining and optimizing the characterization data classification matrix set The classification center data of each characterization data classification matrix in the final classification center data matrix is ​​obtained ;as follows,

[0047] ;

[0048] in, Represents the classification matrix set representing the data Middle i The classification center data set of the data classification matrix j Characterization data corresponding to the characterization type of moisture and heat resistance;

[0049] The characterization types mainly include the depth and distribution of the root system, the stomatal regulation ability of the leaves, the thickness of the wax layer, the permeability of the epidermis, and the content and proportion of biochemical substances such as organic acids, amino acids, and sugars in the peony. These characterization type data have a great influence on the moisture and heat resistance of the peony.

[0050] Preferably, the S33 comprises the following steps:

[0051] S331. Building a chimpanzee population , Indicates the chimpanzee population i Chimpanzee, f represents the size of the chimpanzee population; sets the maximum number of iterations of the chimpanzee population to , the current number of iterations is And the search space dimension is ;

[0052] S332, multiple times from the characterization data classification matrix set A row of data is randomly selected from each data classification matrix in the representation data to form an initial position matrix for each chimpanzee in the chimpanzee population, and an initial position matrix set is obtained. , Indicates the chimpanzee population i The initial position matrix of the chimpanzees is as follows:

[0053] ;

[0054] in, Indicates the chimpanzee population i The first position matrix of the chimpanzee j The classification center data set of the data classification matrix k Characterization data corresponding to the characterization type of moisture and heat resistance;

[0055] S333, according to the initial position matrix set And the matrix set representing the data classification Constructing a fitness function for a chimpanzee population ;as follows,

[0056] ;

[0057] In the formula, express Middle The first k Characterization data corresponding to the characterization type of moisture and heat resistance; Indicates the chimpanzee population The first position matrix of the chimpanzee i The classification center data set of the data classification matrix k Characterization data corresponding to the characterization type of moisture and heat resistance;

[0058] S334, start iteration, before iteration, the current number of iterations is Set to 1; the fitness function of the chimpanzee population is used in each iteration Calculate the fitness value of each chimpanzee; and as the iteration proceeds, it is divided into a search phase and an attack phase;

[0059] During the search phase, the chimpanzee population i The position update formula of each chimpanzee is as follows:

[0060] ;

[0061] In the formula, , Respectively represent the chimpanzee population i the position of each chimpanzee in the next iteration and in the current iteration; Indicates the location of the prey during the current round of iteration; , Respectively represent the chimpanzee population i Two random numbers between 0 and 1 generated by the chimpanzees during the current round of iterations; represents the chaos coefficient;

[0062] During the attack phase, the chimpanzees in the population i The position update formula of each chimpanzee is as follows:

[0063] ;

[0064] In the formula, , , , Respectively represent the positions of the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population during the current round of iterations; , , , Respectively represent the first position update parameters corresponding to the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population; , , , respectively represent the second position update parameters corresponding to the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population; , , , They respectively represent the chaos coefficients corresponding to the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population;

[0065] When the positions of all chimpanzees in the chimpanzee population are updated, the current number of iterations is Add one and enter the next iteration;

[0066] S335, when When , the iteration is stopped and the final global optimal position is output; the final global optimal position is used as the final classification center data matrix ; Otherwise, continue iterating;

[0067] The chimpanzee optimization algorithm can effectively avoid falling into the local optimal solution, has a strong global search capability, helps to find the global optimal solution or at least a satisfactory approximate solution, and maintains a balance between exploring new solution space and developing known solution space, which helps to jump out of the local optimal trap and continue to search for the global optimal solution; therefore, in this scheme, the chimpanzee optimization algorithm is used to perform multiple iterative optimization operations on the classification center data of each representation data classification matrix in the representation data classification matrix set, and the overall discrete degree of the data in the representation data classification matrix set is used as the fitness function, so that with continuous iterations, the overall discrete degree of the data in the representation data classification matrix set is continuously reduced, that is, the obtained classification center data can better represent the classification of each representation data classification matrix in the representation data classification matrix set.

[0068] Preferably, S4 comprises the following steps:

[0069] S41, collecting the peony sample matrix The moisture and heat resistance genotype sequence data corresponding to each peony sample in the genotype sequence data matrix is ​​obtained. ;as follows,

[0070] ;

[0071] in, Represents the set of peony types Middle i The first type of peony collected j The genotype sequence data of the moisture and heat tolerance of peony samples;

[0072] For the genotype sequence data matrix Each genotype sequence data is binary encoded to obtain the genotype sequence encoding data matrix l ;as follows,

[0073] ;

[0074] in, express The corresponding coded data;

[0075] S42, constructing a second initial multi-layer perceptron classification model; using the genotype sequence encoding data matrix l And moisture and heat resistance grade data matrix Performing training and testing operations on the second initial multi-layer perceptron classification model to obtain a second final multi-layer perceptron classification model;

[0076] By binary encoding the genotype sequence data of peony, it is convenient to classify the genotype sequence data later.

[0077] Preferably, the S42 comprises the following steps:

[0078] S421, setting a second training data ratio; encoding the genotype sequence data matrix according to the second training data ratio l And moisture and heat resistance grade data matrix Perform data division to obtain a genotype sequence coding training data matrix, a genotype sequence coding test data matrix, a moisture and heat resistance grade training data matrix, and a moisture and heat resistance grade test data matrix;

[0079] S422, constructing a second initial multi-layer perceptron classification model and setting a second training error threshold; inputting the genotype sequence encoding training data matrix and the moisture and heat resistance level training data matrix into the second initial multi-layer perceptron classification model for training operation; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a second trained multi-layer perceptron classification model; otherwise, continuing the training;

[0080] S423, set a second test accuracy threshold; input the genotype sequence coding test data matrix and the moisture and heat resistance level training data matrix into the second trained multi-layer perceptron classification model for testing; after the test is completed, obtain the second test accuracy; when the second test accuracy is greater than or equal to the second test accuracy threshold, use the second trained multi-layer perceptron classification model as the second final multi-layer perceptron classification model; otherwise, return to S422 to continue training until the second test accuracy is greater than or equal to the second test accuracy threshold, and obtain the second final multi-layer perceptron classification model;

[0081] The second initial multilayer perceptron classification model was trained and tested using the genotype sequence coding training data matrix, the moisture and heat resistance grade training data matrix, the genotype sequence coding test data matrix, and the moisture and heat resistance grade test data matrix, respectively, so that the obtained second final multilayer perceptron classification model has a better classification ability for the genotype sequence coding data of peony samples.

[0082] Preferably, the S5 comprises the following steps:

[0083] S51. Setting the sample set of peony offspring to be screened , Indicates the setting i Peony progeny samples to be screened, Indicates the total number of peony progeny samples to be screened;

[0084] S52, obtaining the sample set of peony offspring to be screened The image data of each peony offspring sample to be screened is obtained to obtain a peony image dataset to be screened;

[0085] According to the feature type set A pre-trained CNN neural network model is used to extract feature data of each peony image data to be screened in the peony image data matrix to obtain a feature data matrix to be screened;

[0086] According to the characterization type set Collect the sample set of peony offspring to be screened The moisture and heat resistance trait characterization data of each peony offspring sample to be screened are obtained to obtain the peony characterization data matrix to be screened;

[0087] Collect the sample set of peony offspring to be screened The moisture and heat resistance genotype sequence data of each peony offspring sample to be screened are obtained to obtain a genotype sequence data set to be screened;

[0088] S53, inputting the feature data matrix to be screened into the first final multi-layer perceptron classification model for classification, obtaining a type label data set to be screened; and combining the average grade data set of moisture and heat resistance of the variety with the data set Obtain the moisture and heat resistance grade data corresponding to each type of label data to be filtered in the type of label data to be filtered, and obtain the first moisture and heat resistance grade data set , Indicates the first moisture and heat resistance grade in the first dataset i The moisture and heat resistance grade data corresponding to the peony offspring samples to be screened;

[0089] S54, respectively calculating each row of data in the peony characterization data matrix to be screened to the final classification center data matrix The Euclidean distance data between each row of data in the matrix to be screened is obtained. ;as follows,

[0090] ;

[0091] in, Indicates the first i Row data to final classification center data matrix Middle j Euclidean distance data between rows of data;

[0092] The Euclidean distance data matrix to be screened The final classification center data matrix corresponding to the smallest data in each row of data The moisture and heat resistance grade data corresponding to the data in the second moisture and heat resistance grade data are recorded as the second moisture and heat resistance grade data; the second moisture and heat resistance grade data is used as the moisture and heat resistance grade data of the corresponding peony offspring sample to be screened, and the second moisture and heat resistance grade data set is obtained. , Indicates the first i The moisture and heat resistance grade data corresponding to the peony offspring samples to be screened;

[0093] S55, binary-encode each genotype sequence data to be screened in the genotype sequence data set to be screened to obtain a genotype sequence encoding data set to be screened; input the genotype sequence encoding data set to be screened into the second final multi-layer perceptron classification model for classification to obtain a third moisture and heat resistance grade data set , Indicates the third moisture and heat resistance grade data set i The moisture and heat resistance grade data corresponding to the peony offspring samples to be screened;

[0094] The first final multi-layer perceptron classification model, the final classification center data matrix and the second final multi-layer perceptron classification model were used to classify the feature data matrix to be screened, the peony characterization data matrix to be screened and the genotype sequence data set to be screened respectively. The moisture and heat resistance grade data of the peony offspring samples to be screened were obtained from three perspectives, making the obtained moisture and heat resistance grade data more comprehensive, providing an analytical basis for subsequent comprehensive analysis.

[0095] Preferably, the S6 includes the following process:

[0096] S61, the first moisture and heat resistance level data set , the second moisture and heat resistance level data set And the third moisture and heat resistance level data set The moisture and heat resistance grade data at the corresponding positions are compared in numerical value to obtain the final moisture and heat resistance grade data set to be screened. , Indicates i The final moisture and heat resistance grade data corresponding to the peony offspring samples to be screened; the calculation formula is as follows:

[0097] ;

[0098] S62, setting a moisture and heat resistance grade data threshold; and Extracting peony offspring samples corresponding to the final moisture and heat resistance grade data that is greater than or equal to the moisture and heat resistance grade data threshold value to obtain a screened peony offspring sample set;

[0099] By comparing the first moisture and heat resistance grade data set, the second moisture and heat resistance grade data set, and the third moisture and heat resistance grade data set, the final judgment process is quantified to obtain the final moisture and heat resistance grade data of the peony offspring samples; and then according to the final moisture and heat resistance grade data, the peony offspring samples to be screened whose moisture and heat resistance grade data meet the requirements in the peony offspring sample set to be screened are extracted, making the extraction process more scientific and accurate.

[0100] A data analysis-based intelligent screening system for peony offspring with humidity and heat resistance, comprising a first peony characteristic data collection module, a humidity and heat resistance grade data acquisition module, a type label generation module, a first classification model construction module, a second peony characteristic data collection module, a characterization classification center data optimization module, a genotype sequence data collection module, a second classification model construction module, a peony offspring data collection module to be screened, a classification module and a final screening module;

[0101] The first peony characteristic data collection module is used to collect a plurality of characteristic type data reflecting peony varieties from a plurality of peony samples to obtain a characteristic data set matrix;

[0102] The moisture and heat resistance grade data acquisition module is used to acquire moisture and heat resistance grade data of multiple peony samples to obtain a moisture and heat resistance grade data matrix;

[0103] The type label generation module is used to generate a type label corresponding to each feature data set in the feature data set matrix to obtain a type label data matrix;

[0104] The first classification model construction module is used to construct a first final multi-layer perceptron classification model using the feature data set matrix and the type label data matrix;

[0105] The second peony characteristic data collection module is used to collect multiple moisture and heat resistance trait characterization type data of multiple peony samples to obtain a peony characterization data set matrix;

[0106] The characterization classification center data optimization module is used to obtain and optimize the classification center data of each category of the peony characterization data set in the peony characterization data set matrix to obtain a final classification center data matrix;

[0107] The genotype sequence data acquisition module is used to collect the moisture and heat resistance genotype sequence data of multiple peony samples to obtain a genotype sequence data matrix;

[0108] The second classification model construction module is used to construct a second final multi-layer perceptron classification model using the genotype sequence data matrix and the moisture and heat resistance grade data matrix;

[0109] The peony offspring data collection module to be screened is used to collect data of multiple peony offspring samples to be screened, and obtain a feature data matrix to be screened, a peony characterization data matrix to be screened, and a genotype sequence data set to be screened;

[0110] The classification module is used to classify and compare the feature data matrix to be screened, the peony characterization data matrix to be screened, and the genotype sequence data set to be screened by using the first final multi-layer perceptron classification model, the final classification center data matrix, and the second final multi-layer perceptron classification model, respectively, to obtain a classification result;

[0111] The final screening module is used to screen a plurality of peony offspring samples to be screened according to the classification results to obtain a screened peony offspring sample set.

[0112] The present invention has the following beneficial effects:

[0113] 1. In the present invention, the moisture and heat resistance of the peony offspring samples to be screened is classified and evaluated from three perspectives: peony varieties, peony biological characteristics, and peony genes, so that the classification and screening process is more comprehensive and accurate; wherein, by constructing a first final multi-layer perceptron classification model, a classification tool is provided for the subsequent classification of the characteristic data of the peony offspring samples to be screened; by obtaining and optimizing the classification center data of the peony characterization data set of each category in the peony characterization data set matrix, a classification basis is provided for the subsequent classification of the characterization data of the peony offspring samples to be screened; by constructing a second final multi-layer perceptron classification model, a classification tool is provided for the subsequent classification of the genotype sequence data of the peony offspring samples to be screened.

[0114] 2. In the present invention, the chimpanzee optimization algorithm is used to perform multiple iterative optimization operations on the classification center data of each characterization data classification matrix in the characterization data classification matrix set, and the overall discrete degree of the data in the characterization data classification matrix set is used as the fitness function. As a result, with continuous iterations, the overall discrete degree of the data in the characterization data classification matrix set becomes smaller and smaller, that is, the obtained classification center data can better represent the classification of each characterization data classification matrix in the characterization data classification matrix set.

[0115] 3. In the present invention, by calculating the average grade data of moisture and heat resistance of each type of peony, the randomness of the data is reduced and the accuracy of the data is improved; the feature type set includes flower color, flower shape, number of petals and morphology.

[0116] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0118] Figure 1 The present invention is a schematic diagram of the process of screening peony moisture and heat resistant offspring by a peony moisture and heat resistant offspring intelligent screening system based on data analysis. DETAILED DESCRIPTION

[0119] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0120] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0121] Embodiment 1

[0122] This embodiment is a method for intelligently screening peony offspring with humidity and heat resistance based on data analysis, comprising the following steps:

[0123] S1, collecting multiple characteristic type data and moisture and heat resistance grade data that can reflect the peony varieties of multiple peony samples and generating corresponding type labels to obtain a characteristic data set matrix, a moisture and heat resistance grade data matrix and a type label data matrix;

[0124] The S1 comprises the following steps:

[0125] S11. Set multiple peony types to obtain a peony type set , Indicates the setting i Types of Peonies, Indicates the total number of set peony types; according to the peony type set Collect peony samples and obtain peony sample matrix ;as follows,

[0126] ;

[0127] in, Represents the set of peony types Middle i The first type of peony collected j peony samples; Represents the set of peony types The total number of peony samples collected for each peony type;

[0128] Set the moisture and heat resistance level set , Indicates the setting i Moisture and heat resistance level, Indicates the total number of moisture and heat resistance levels set; Peony Sample Matrix Each peony sample is subjected to a water flooding test for the same time; after the test is completed, the moisture and heat resistance grade data of each peony sample is determined to obtain a moisture and heat resistance grade data matrix ;as follows,

[0129] ;

[0130] in, Represents the set of peony types Middle i The first type of peony collected j The moisture and heat resistance rating data of peony samples;

[0131] Calculate the moisture and heat resistance grade data matrix The average value of each row of data in the , to obtain the average grade of moisture and heat resistance of varieties , Represents the peony type set Middle i Average ratings of moisture and heat resistance of peony types;

[0132] S12, obtaining the peony sample matrix The image data of each peony sample in the image matrix is ​​obtained b ;as follows,

[0133] ;

[0134] in, Represents the set of peony types Middle i The first type of peony collected j Image data of peony samples;

[0135] Set multiple feature types that can reflect peony varieties and obtain a feature type set , Indicates the setting i can reflect the characteristic types of peony varieties. Indicates the total number of characteristic types that can reflect peony varieties;

[0136] According to the feature type set The pre-trained CNN neural network model is used to extract the peony image data matrix b The feature data set matrix is ​​obtained by taking the feature data of each peony image data in ;as follows,

[0137] ;

[0138] in, Indicates for The collected feature datasets, , express No. k Feature data of feature types;

[0139] S13, matching the peony type set Set the feature dataset matrix The type label of each feature data set in the matrix is ​​obtained c ;as follows,

[0140] ;

[0141] in, express The corresponding type tag data;

[0142] S2, using the feature data set matrix and the type label data matrix to construct a first final multi-layer perceptron classification model;

[0143] The S2 comprises the following steps:

[0144] S21, setting a first training data ratio; and performing a feature data set matrix on the basis of the first training data ratio. and the type label data matrix c Perform data partitioning to obtain a feature training data set matrix, a feature test data set matrix, a type label training data matrix, and a type label test data matrix;

[0145] S22, constructing a first initial multi-layer perceptron classification model and setting a first training error threshold; inputting the feature training data set matrix and the type label training data matrix into the first initial multi-layer perceptron classification model for training; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain a first trained multi-layer perceptron classification model; otherwise, continuing the training;

[0146] S23, setting a first test accuracy threshold; inputting the feature test data set matrix and the type label test data matrix into the first trained multi-layer perceptron classification model for testing; after the test is completed, obtaining a first test accuracy; when the first test accuracy is greater than or equal to the first test accuracy threshold, using the first trained multi-layer perceptron classification model as the first final multi-layer perceptron classification model; otherwise, returning to S22 to continue training until the first test accuracy is greater than or equal to the first test accuracy threshold, obtaining the first final multi-layer perceptron classification model;

[0147] S3, collecting multiple moisture and heat resistance trait characterization type data of multiple peony samples to obtain a peony characterization data set matrix; obtaining and optimizing the classification center data of each category of the peony characterization data set in the peony characterization data set matrix to obtain a final classification center data matrix;

[0148] The S3 comprises the following steps:

[0149] S31, setting multiple moisture and heat resistance trait characterization types of peony samples, and obtaining a characterization type set , Indicates the setting i Types of moisture and heat resistance characterization, Indicates the total number of moisture and heat resistance trait characterization types set; according to the characterization type set Collect the peony sample matrix The moisture and heat resistance trait characterization data of each peony sample in the peony characterization data set matrix is ​​obtained. ;as follows,

[0150] ;

[0151] in, Represents the set of peony types Middle i The first type of peony collected j Characterization dataset of peony samples; , express The k Characterization data corresponding to the characterization type of moisture and heat resistance;

[0152] S32, according to the moisture and heat resistance grade data matrix Matrix for Peony Characterization Dataset The characterization data in is divided into categories to obtain the characterization data classification matrix set , Indicates the divided i A matrix representing data classification; as follows,

[0153] ;

[0154] in, express Middle j The first k Characterization data corresponding to the characterization type of moisture and heat resistance; express The total number of representation data sets in ;

[0155] S33, obtaining and optimizing the characterization data classification matrix set The classification center data of each characterization data classification matrix in the final classification center data matrix is ​​obtained ;as follows,

[0156] ;

[0157] in, Represents the classification matrix set representing the data Middle i The classification center data set of the data classification matrix j Characterization data corresponding to the characterization type of moisture and heat resistance;

[0158] The S33 comprises the following steps:

[0159] S331. Building a chimpanzee population , Indicates the chimpanzee population i Chimpanzee,f represents the size of the chimpanzee population; sets the maximum number of iterations of the chimpanzee population to , the current number of iterations is And the search space dimension is ;

[0160] S332, multiple times from the characterization data classification matrix set A row of data is randomly selected from each data classification matrix in the representation data to form an initial position matrix for each chimpanzee in the chimpanzee population, and an initial position matrix set is obtained. , Indicates the chimpanzee population i The initial position matrix of the chimpanzees is as follows:

[0161] ;

[0162] in, Indicates the chimpanzee population i The first position matrix of the chimpanzee j The classification center data set of the data classification matrix k Characterization data corresponding to the characterization type of moisture and heat resistance;

[0163] S333, according to the initial position matrix set And the matrix set representing the data classification Constructing a fitness function for a chimpanzee population ;as follows,

[0164] ;

[0165] In the formula, express Middle The first k Characterization data corresponding to the characterization type of moisture and heat resistance; Indicates the chimpanzee population The first position matrix of the chimpanzee i The classification center data set of the data classification matrix k Characterization data corresponding to the characterization type of moisture and heat resistance;

[0166] S334, start iteration, before iteration, the current number of iterations is Set to 1; the fitness function of the chimpanzee population is used in each iteration Calculate the fitness value of each chimpanzee; and as the iteration proceeds, it is divided into a search phase and an attack phase;

[0167] During the search phase, the chimpanzee population i The position update formula of each chimpanzee is as follows:

[0168] ;

[0169] In the formula, , Respectively represent the chimpanzee population i the position of each chimpanzee in the next iteration and in the current iteration; Indicates the location of the prey during the current round of iteration; , Respectively represent the chimpanzee population i Two random numbers between 0 and 1 generated by the chimpanzees during the current round of iterations; represents the chaos coefficient;

[0170] During the attack phase, the chimpanzees in the population i The position update formula of each chimpanzee is as follows:

[0171] ;

[0172] In the formula, , , , Respectively represent the positions of the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population during the current round of iterations; , , , Respectively represent the first position update parameters corresponding to the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population; , , , respectively represent the second position update parameters corresponding to the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population; , , , They respectively represent the chaos coefficients corresponding to the attacking chimpanzees, the blocking chimpanzees, the chasing chimpanzees and the driving chimpanzees in the chimpanzee population;

[0173] When the positions of all chimpanzees in the chimpanzee population are updated, the current number of iterations is Add one and enter the next iteration;

[0174] S335, when When , the iteration is stopped and the final global optimal position is output; the final global optimal position is used as the final classification center data matrix ; Otherwise, continue iterating;

[0175] S4, collecting the moisture and heat resistance genotype sequence data of multiple peony samples and building the second final multi-layer perceptron classification model in conjunction with the moisture and heat resistance grade data matrix;

[0176] The S4 comprises the following steps:

[0177] S41, collecting the peony sample matrix The moisture and heat resistance genotype sequence data corresponding to each peony sample in the genotype sequence data matrix is ​​obtained. ;as follows,

[0178] ;

[0179] in, Represents the set of peony types Middle i The first type of peony collected j The genotype sequence data of the moisture and heat tolerance of peony samples;

[0180] For the genotype sequence data matrix Each genotype sequence data is binary encoded to obtain the genotype sequence encoding data matrix l ;as follows,

[0181] ;

[0182] in, express The corresponding coded data;

[0183] S42, constructing a second initial multi-layer perceptron classification model; using the genotype sequence encoding data matrix l And moisture and heat resistance grade data matrix Performing training and testing operations on the second initial multi-layer perceptron classification model to obtain a second final multi-layer perceptron classification model;

[0184] S5, collect data of multiple peony offspring samples to be screened, obtain a feature data matrix to be screened, a peony characterization data matrix to be screened, and a genotype sequence data set to be screened; use the first final multi-layer perceptron classification model, the final classification center data matrix, and the second final multi-layer perceptron classification model to classify and compare the feature data matrix to be screened, the peony characterization data matrix to be screened, and the genotype sequence data set to be screened, respectively, to obtain a classification result;

[0185] The S5 comprises the following steps:

[0186] S51. Setting the sample set of peony offspring to be screened , Indicates the setting i samples of peony offspring to be screened, Indicates the total number of peony progeny samples to be screened;

[0187] S52, obtaining the peony offspring sample set to be screened The image data of each peony offspring sample to be screened is obtained to obtain a peony image dataset to be screened;

[0188] According to the feature type set A pre-trained CNN neural network model is used to extract feature data of each peony image data to be screened in the peony image data matrix to obtain a feature data matrix to be screened;

[0189] According to the characterization type set Collect the sample set of peony offspring to be screened The moisture and heat resistance trait characterization data of each peony offspring sample to be screened are obtained to obtain the peony characterization data matrix to be screened;

[0190] Collect the sample set of peony offspring to be screened The moisture and heat resistance genotype sequence data of each peony offspring sample to be screened are obtained to obtain a genotype sequence data set to be screened;

[0191] S53, inputting the feature data matrix to be screened into the first final multi-layer perceptron classification model for classification, obtaining a type label data set to be screened; and combining the average grade data set of moisture and heat resistance of the variety with the data set Obtain the moisture and heat resistance grade data corresponding to each type of label data to be filtered in the type of label data to be filtered, and obtain the first moisture and heat resistance grade data set , Indicates the first moisture and heat resistance grade in the first dataset i The moisture and heat resistance grade data corresponding to the peony offspring samples to be screened;

[0192] S54, respectively calculating each row of data in the peony characterization data matrix to be screened to the final classification center data matrix The Euclidean distance data between each row of data in the matrix to be screened is obtained. ;as follows,

[0193] ;

[0194] in, Indicates the first i Row data to final classification center data matrix Middle j Euclidean distance data between rows of data;

[0195] The Euclidean distance data matrix to be screened The final classification center data matrix corresponding to the smallest data in each row of data The moisture and heat resistance grade data corresponding to the data in the second moisture and heat resistance grade data are recorded as the second moisture and heat resistance grade data; the second moisture and heat resistance grade data is used as the moisture and heat resistance grade data of the corresponding peony offspring sample to be screened, and the second moisture and heat resistance grade data set is obtained. , Indicates the first i The moisture and heat resistance grade data corresponding to the peony offspring samples to be screened;

[0196] S55, binary-encode each genotype sequence data to be screened in the genotype sequence data set to be screened to obtain a genotype sequence encoding data set to be screened; input the genotype sequence encoding data set to be screened into the second final multi-layer perceptron classification model for classification to obtain a third moisture and heat resistance grade data set , Indicates the third moisture and heat resistance grade data set i The moisture and heat resistance grade data corresponding to the peony offspring samples to be screened;

[0197] S6. Screening a plurality of peony offspring samples to be screened according to the classification results to obtain a screened peony offspring sample set;

[0198] The S6 includes the following process:

[0199] S61, the first moisture and heat resistance level data set , the second moisture and heat resistance level data set And the third moisture and heat resistance level data set The moisture and heat resistance grade data at the corresponding positions are compared in numerical value to obtain the final moisture and heat resistance grade data set to be screened. , Indicates i The final moisture and heat resistance grade data corresponding to the peony offspring samples to be screened; the calculation formula is as follows:

[0200] ;

[0201] S62, setting a moisture and heat resistance grade data threshold; and The peony offspring samples corresponding to the final moisture and heat resistance grade data that is greater than or equal to the moisture and heat resistance grade data threshold are extracted to obtain the screened peony offspring sample set.

[0202] Embodiment 2

[0203] The present embodiment discloses an intelligent screening system for peony offspring resistant to humidity and heat based on data analysis, and the system can implement the method of the above embodiment, including a first peony characteristic data collection module, a humidity and heat resistance grade data acquisition module, a type label generation module, a first classification model construction module, a second peony characteristic data collection module, a characterization classification center data optimization module, a genotype sequence data collection module, a second classification model construction module, a peony offspring data collection module to be screened, a classification module and a final screening module;

[0204] The first peony characteristic data collection module is used to collect a plurality of characteristic type data reflecting peony varieties from a plurality of peony samples to obtain a characteristic data set matrix;

[0205] The moisture and heat resistance grade data acquisition module is used to acquire moisture and heat resistance grade data of multiple peony samples to obtain a moisture and heat resistance grade data matrix;

[0206] The type label generation module is used to generate a type label corresponding to each feature data set in the feature data set matrix to obtain a type label data matrix;

[0207] The first classification model construction module is used to construct a first final multi-layer perceptron classification model using the feature data set matrix and the type label data matrix;

[0208] The second peony characteristic data collection module is used to collect multiple moisture and heat resistance trait characterization type data of multiple peony samples to obtain a peony characterization data set matrix;

[0209] The characterization classification center data optimization module is used to obtain and optimize the classification center data of each category of the peony characterization data set in the peony characterization data set matrix to obtain a final classification center data matrix;

[0210] The genotype sequence data acquisition module is used to collect the moisture and heat resistance genotype sequence data of multiple peony samples to obtain a genotype sequence data matrix;

[0211] The second classification model construction module is used to construct a second final multi-layer perceptron classification model using the genotype sequence data matrix and the moisture and heat resistance grade data matrix;

[0212] The peony offspring data collection module to be screened is used to collect data of multiple peony offspring samples to be screened, and obtain a feature data matrix to be screened, a peony characterization data matrix to be screened, and a genotype sequence data set to be screened;

[0213] The classification module is used to classify and compare the feature data matrix to be screened, the peony characterization data matrix to be screened, and the genotype sequence data set to be screened by using the first final multi-layer perceptron classification model, the final classification center data matrix, and the second final multi-layer perceptron classification model, respectively, to obtain a classification result;

[0214] The final screening module is used to screen a plurality of peony offspring samples to be screened according to the classification results to obtain a screened peony offspring sample set.

[0215] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0216] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A method for intelligent screening of peony offspring resistant to humidity and heat based on data analysis, characterized in that: The following steps are involved: S1, collecting multiple characteristic type data and moisture and heat resistance grade data that can reflect the peony varieties of multiple peony samples and generating corresponding type labels to obtain a characteristic data set matrix, a moisture and heat resistance grade data matrix and a type label data matrix; S2, using the feature data set matrix and the type label data matrix to construct a first final multi-layer perceptron classification model; The S2 comprises the following steps: S21, partitioning the feature data set matrix and the type label data matrix to obtain a feature training data set matrix, a feature test data set matrix, a type label training data matrix, and a type label test data matrix; S22, constructing a first initial multi-layer perceptron classification model and setting a first training error threshold; inputting the feature training data set matrix and the type label training data matrix into the first initial multi-layer perceptron classification model for training; during the training process, when the training error is less than the first training error threshold, stopping the training to obtain a first trained multi-layer perceptron classification model; otherwise, continuing the training; S23, setting a first test accuracy threshold; inputting the feature test data set matrix and the type label test data matrix into the first trained multi-layer perceptron classification model for testing; after the test is completed, obtaining a first test accuracy; when the first test accuracy is greater than or equal to the first test accuracy threshold, using the first trained multi-layer perceptron classification model as the first final multi-layer perceptron classification model; otherwise, returning to S22 to continue training until the first test accuracy is greater than or equal to the first test accuracy threshold, obtaining the first final multi-layer perceptron classification model; S3, collecting multiple moisture and heat resistance trait characterization type data of multiple peony samples to obtain a peony characterization data set matrix; obtaining and optimizing the classification center data of each category of the peony characterization data set in the peony characterization data set matrix to obtain a final classification center data matrix; S4, collecting the moisture and heat resistance genotype sequence data of multiple peony samples and building the second final multi-layer perceptron classification model in conjunction with the moisture and heat resistance grade data matrix; The S4 comprises the following steps: S41, collecting the moisture and heat resistance genotype sequence data corresponding to each peony sample in the peony sample matrix to obtain a genotype sequence data matrix; performing binary encoding on each genotype sequence data in the genotype sequence data matrix to obtain a genotype sequence encoding data matrix; S42, constructing a second initial multilayer perceptron classification model; using the genotype sequence encoding data matrix and the moisture and heat resistance grade data matrix to train and test the second initial multilayer perceptron classification model to obtain a second final multilayer perceptron classification model; The S42 comprises the following steps: S421, setting a second training data ratio; encoding the genotype sequence data matrix according to the second training data ratio l And moisture and heat resistance grade data matrix Perform data division to obtain a genotype sequence coding training data matrix, a genotype sequence coding test data matrix, a moisture and heat resistance grade training data matrix, and a moisture and heat resistance grade test data matrix; S422, constructing a second initial multi-layer perceptron classification model and setting a second training error threshold; inputting the genotype sequence encoding training data matrix and the moisture and heat resistance level training data matrix into the second initial multi-layer perceptron classification model for training operation; during the training process, when the training error is less than the second training error threshold, stopping the training to obtain a second trained multi-layer perceptron classification model; otherwise, continuing the training; S423, set a second test accuracy threshold; input the genotype sequence coding test data matrix and the moisture and heat resistance level training data matrix into the second trained multi-layer perceptron classification model for testing; after the test is completed, obtain the second test accuracy; when the second test accuracy is greater than or equal to the second test accuracy threshold, use the second trained multi-layer perceptron classification model as the second final multi-layer perceptron classification model; otherwise, return to S422 to continue training until the second test accuracy is greater than or equal to the second test accuracy threshold, and obtain the second final multi-layer perceptron classification model; S5, collect data of multiple peony offspring samples to be screened, obtain a feature data matrix to be screened, a peony characterization data matrix to be screened, and a genotype sequence data set to be screened; use the first final multi-layer perceptron classification model, the final classification center data matrix, and the second final multi-layer perceptron classification model to classify and compare the feature data matrix to be screened, the peony characterization data matrix to be screened, and the genotype sequence data set to be screened, respectively, to obtain a classification result; S6. Screening a plurality of peony offspring samples to be screened according to the classification results to obtain a screened peony offspring sample set.

2. The method for intelligent screening of peony offspring resistant to humidity and heat based on data analysis according to claim 1, characterized in that: The S1 comprises the following steps: S11, set multiple peony types to obtain a peony type set; collect peony samples according to the peony type set to obtain a peony sample matrix; set a moisture and heat resistance grade set; perform a water flooding test on each peony sample in the peony sample matrix for the same time in conjunction with the moisture and heat resistance grade set; after the test is completed, determine the moisture and heat resistance grade data of each peony sample to obtain a moisture and heat resistance grade data matrix; calculate the average value of each row of data in the moisture and heat resistance grade data matrix to obtain a variety moisture and heat resistance average grade data set; S12, obtaining the image data of each peony sample in the peony sample matrix to obtain a peony image data matrix; setting a plurality of feature types that can reflect the peony variety to obtain a feature type set; extracting the feature data of each peony image data in the peony image data matrix according to the feature type set to obtain a feature data set matrix; S13, setting the type label of each feature data set in the feature data set matrix in accordance with the peony type set to obtain a type label data matrix.

3. The method for intelligent screening of peony offspring resistant to humidity and heat based on data analysis according to claim 2, characterized in that: The S3 comprises the following steps: S31, setting a plurality of moisture and heat resistance trait characterization types of peony samples to obtain a characterization type set; collecting moisture and heat resistance trait characterization data of each peony sample in the peony sample matrix according to the characterization type set to obtain a peony characterization data set matrix; S32, classifying the characterization data in the peony characterization data set matrix according to the moisture and heat resistance grade data matrix to obtain a characterization data classification matrix set; S33, obtaining and optimizing the classification center data of each characterization data classification matrix in the characterization data classification matrix set to obtain a final classification center data matrix.

4. The method for intelligently screening peony offspring resistant to humidity and heat based on data analysis according to claim 3, characterized in that: The S33 comprises the following steps: S331, constructing a chimpanzee population; setting the maximum number of iterations of the chimpanzee population to , the current number of iterations is ; S332, randomly selecting a row of data from each of the characterization data classification matrix sets for multiple times and combining them to form an initial position matrix for each chimpanzee in the chimpanzee population, thereby obtaining an initial position matrix set; S333, constructing a fitness function of the chimpanzee population according to the initial position matrix set and the characterization data classification matrix set; S334, start iteration; in each round of iteration, the fitness function of the chimpanzee population is used to calculate the fitness value of each chimpanzee, and as the iteration proceeds, it is divided into a search phase and an attack phase; S335, when When , stop the iteration and output the final global optimal position; use the final global optimal position as the final classification center data matrix; otherwise, continue to iterate.

5. The method for intelligent screening of peony offspring resistant to humidity and heat based on data analysis according to claim 4, characterized in that: The S5 comprises the following steps: S51, setting a sample set of peony offspring to be screened; S52, acquiring the image data of each peony offspring sample to be screened in the peony offspring sample set to be screened, and obtaining the peony image data set to be screened; extracting the feature data of each peony image data to be screened in the peony image data matrix to be screened according to the feature type set, and obtaining the feature data matrix to be screened; collecting the moisture and heat resistance trait characterization data of each peony offspring sample to be screened in the peony offspring sample set to be screened according to the characterization type set, and obtaining the peony characterization data matrix to be screened; collecting the moisture and heat resistance genotype sequence data of each peony offspring sample to be screened in the peony offspring sample set to be screened, and obtaining the genotype sequence data set to be screened; S53, inputting the feature data matrix to be screened into the first final multi-layer perceptron classification model for classification, and obtaining a type label data set to be screened; obtaining the moisture and heat resistance grade data corresponding to each type label data to be screened in the type label data set to be screened in conjunction with the average moisture and heat resistance grade data set of the varieties, and obtaining a first moisture and heat resistance grade data set; S54, respectively calculating the Euclidean distance data between each row of data in the peony characterization data matrix to be screened and each row of data in the final classification center data matrix, to obtain the Euclidean distance data matrix to be screened; recording the moisture and heat resistance grade data corresponding to the data in the final classification center data matrix corresponding to the smallest data in each row of data in the Euclidean distance data matrix to be screened as the second moisture and heat resistance grade data; using the second moisture and heat resistance grade data as the moisture and heat resistance grade data of the corresponding peony offspring sample to be screened, to obtain the second moisture and heat resistance grade data set; S55, binary-encode each genotype sequence data to be screened in the genotype sequence data set to be screened to obtain a genotype sequence encoding data set to be screened; input the genotype sequence encoding data set to be screened into the second final multi-layer perceptron classification model for classification to obtain a third moisture and heat resistance grade data set.

6. The method for intelligent screening of peony offspring resistant to humidity and heat based on data analysis according to claim 5, characterized in that: The S6 includes the following process: S61, comparing the values ​​of the moisture and heat resistance grade data at corresponding positions among the first moisture and heat resistance grade data set, the second moisture and heat resistance grade data set, and the third moisture and heat resistance grade data set, to obtain a final moisture and heat resistance grade data set to be screened; S62, setting a moisture and heat resistance grade data threshold; extracting peony progeny samples corresponding to final moisture and heat resistance grade data greater than or equal to the moisture and heat resistance grade data threshold in the final moisture and heat resistance grade data to be screened, to obtain a screened peony progeny sample set.

7. A system for implementing the method for intelligent screening of peony offspring resistant to moisture and heat based on data analysis as described in any one of claims 1 to 6.

Images