Intelligent analysis system and method for muskmelon breeding through grey decision and symmetry simulation analysis

Abstract

The invention discloses an intelligent analysis system and method for muskmelon breeding through gray decision and symmetry simulation analysis. The system comprises an acquisition module, an extraction module, a construction module and an analysis module, the acquisition module is used for acquiring image data of watermelons and melons; the extraction module is used for extracting image features in the image data; the construction module is used for constructing an analysis model based on the image features; and the analysis module is used for completing analysis of watermelon breeding by using the analysis model. According to the method, crop information simulating morphological characteristics and growth rules of real plants is collected, labor is saved, data collection is accurate, decision analysis is carried out on the collected information, an optimal scheme is given, excellent variety resources are reserved, and the breeding process is accelerated. The use is convenient and effective.

Description

Technical Field

[0001] The present invention relates to the field of modern agricultural information and intelligent equipment, and in particular to a grey decision-making and symmetry simulation analysis system and method for melon breeding intelligent analysis. Background Art

[0002] With the development of modern agricultural technology, melon breeding technology has become a key link in improving melon yield and quality. Traditional melon breeding methods mainly rely on manual observation and empirical judgment. This method is not only inefficient, but also greatly affected by human factors, making it difficult to achieve accurate and systematic breeding decisions. In addition, traditional breeding methods cannot comprehensively and quickly collect and analyze the morphological characteristics and growth patterns of melons, resulting in a slow breeding process and low efficiency in screening excellent varieties.

[0003] In the field of modern agricultural information and intelligent equipment, how to use modern information technology and intelligent equipment to improve the efficiency and accuracy of melon breeding has become a technical problem that needs to be solved urgently. Some existing automation and intelligent technologies attempt to assist breeding decisions through image recognition and data analysis, but these technologies often lack systematic analysis models and precise decision-making methods, and are difficult to meet the needs of complex morphological characteristics and growth law analysis in melon breeding.

[0004] In addition, the morphological characteristics and growth patterns of melons are affected by both genetic and environmental factors. Traditional analysis methods are difficult to comprehensively consider these factors, resulting in limited accuracy and reliability of breeding decisions. Therefore, developing an intelligent analysis system and method for melon breeding that can simulate the morphological characteristics and growth patterns of real plants and provide accurate decision-making analysis is of great significance for accelerating the breeding process and improving breeding efficiency and quality. Summary of the invention

[0005] In order to solve the technical problems in the above background, the present invention aims to collect image data of melons, extract image features, construct an analysis model, and use the model to complete accurate analysis of melon breeding, thereby realizing rapid screening and retention of excellent variety resources.

[0006] To achieve the above-mentioned purpose, the present invention provides a grey decision-making and symmetry simulation analysis system for melon breeding intelligence, comprising: a collection module, an extraction module, a construction module and an analysis module;

[0007] The acquisition module is used to acquire image data of muskmelon;

[0008] The extraction module is used to extract image features from the image data;

[0009] The construction module is used to construct an analysis model based on the image features;

[0010] The analysis module is used to complete the analysis of muskmelon breeding using the analysis model.

[0011] Preferably, the workflow of the acquisition module includes: using a 3D fixed-focus lens to shoot plants of different varieties and types one by one in the acquisition area to obtain clear and bright image information for displaying the shape phenotypic characteristics of the plants; the shape phenotypic characteristics include: the results presented by the development of organs such as roots, stems, leaves, flowers, fruits and seeds of the plant body.

[0012] Preferably, the workflow of the extraction module includes: first, preprocessing the image data to obtain a processed image; then, using a threshold processing-based segmentation method to divide the processed image into regions or fragments, each region or fragment corresponds to a specific object or feature in the image; after the image segmentation is completed, the extraction module identifies and extracts relevant features by setting boundary conditions from the segmented image.

[0013] Preferably, the workflow of the construction module includes: constructing an analysis model using a grey analysis decision model and constructing an evaluation index system:

[0014]

[0015] In the formula, x i (k) represents the kth data of the i-th indicator, max{x i (k)} and min{x i (k)} represent x i (k) maximum and minimum values;

[0016] Create a reference sequence X 0 and compare the series X i , where X 0 represents the ideal state of the decision-making goal; X i Represents the performance of each scheme on each evaluation index, and calculates the grey correlation formula:

[0017] First, calculate the absolute value of the difference between the reference sequence and the comparison sequence:

[0018] ΔX 0 (k) = |X 0 (k)-X i (k)|

[0019] Then calculate the correlation coefficient:

[0020]

[0021] Where ρ represents the resolution coefficient; its value is between (0, 1); min i ΔX i(k)max i ΔX i (k) represents the minimum and maximum difference between all comparison series and reference series at the kth data respectively;

[0022] Finally, calculate the relevance:

[0023]

[0024] Among them, r i It represents the correlation between the ith solution and the reference sequence, and n represents the number of data.

[0025] Preferably, the analysis module uses the analysis model to analyze the collected data. Specifically, it extracts the detailed characteristic element indicators of leaves, internodes and fruits of different muskmelon resources, performs calculation analysis through the analysis model and symmetric simulation, and pushes the decision result.

[0026] The present invention also provides a grey decision-making and symmetry simulation analysis method for melon breeding intelligence analysis, which is applied to the above system and comprises the following steps:

[0027] Collect image data of cantaloupe;

[0028] extracting image features from the image data;

[0029] Based on the image features, construct an analysis model;

[0030] The analysis model is used to complete the analysis of muskmelon breeding.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention collects crop information that simulates the morphological characteristics and growth patterns of real plants, saves labor, collects data accurately, and makes decision-making analysis based on the collected information to give the optimal solution, preserve excellent variety resources, and accelerate the breeding process. It is easy to use and effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0034] Figure 1 A schematic diagram of the system structure of an embodiment of the present invention;

[0035] Figure 2 The figure is a schematic diagram of a method flow of an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] Embodiment 1

[0039] like Figure 1 As shown, it is a schematic diagram of the system structure of this embodiment, including: an acquisition module, an extraction module, a construction module and an analysis module; the acquisition module is used to collect image data of muskmelon; the extraction module is used to extract image features in the image data; the construction module is used to build an analysis model based on image features; the analysis module is used to use the analysis model to complete the analysis of muskmelon breeding. The workflow of the analysis system of this embodiment includes: collecting image data of muskmelon, determining the type and quantity index of the collected crop, selecting the collection area (leaves, internodes, fruits, etc.), collecting real morphological characteristics and growth law information, and using gray analysis models and symmetrical simulation of crop element indicators. According to the breeding goals, give the preferred solution, retain excellent variety resources, and ensure that the crop variety data information is scientific and reliable. Accelerate the breeding process.

[0040] The following will describe in detail how the present invention solves the technical problems in practical work in conjunction with this embodiment.

[0041] In this embodiment, the acquisition module uses a 3D fixed-focus lens, selects a suitable focal length, and uses the wide-angle feature to highlight the real morphological characteristics and growth law information of the plant at close range. These image data are clear and bright photos of the detailed features of the plant's leaves, internodes, and fruits.

[0042] In the collection area, plants of different varieties and types are photographed one by one to obtain clear and bright image information in order to show the main shape phenotypic characteristics of the plants. The plant body is generally composed of organs such as roots, stems, leaves, flowers, fruits and seeds. The formation of plant phenotypic characteristics is the result of the common presentation of these organs after development. The characteristics of morphological structure are often closely related to morphological traits, reproductive traits and quality traits. For example: the main shape phenotypic characteristics collected by the cantaloupe breeding institute include stems, leaves, flowers, fruits and seeds. The collected cantaloupe images are sorted and classified to ensure that the phenotypic characteristics of each plant have clear data records.

[0043] The image features in the image data are extracted using the extraction module. The image data is first preprocessed, including image enhancement and image restoration, to improve the visual quality of the image. By increasing contrast, reducing noise and removing artifacts, as well as removing degradations such as blur, noise and distortion from the image.

[0044] The processed image is then divided into regions or segments, each of which corresponds to a specific object or feature in the image. This embodiment adopts a segmentation method based on threshold processing. First, a grayscale histogram is calculated for the image to analyze the distribution of different grayscale values ​​in the image, and a threshold is determined based on the distribution. The image is binarized using the determined threshold and the image is segmented into two parts, the target and the background, based on the result of the threshold processing. Finally, the segmented image is post-processed, such as removing noise, filling holes, etc., to improve the segmentation quality.

[0045] After the image segmentation is completed, the extraction module identifies and extracts relevant features by setting boundary conditions from the segmented image, such as the aspect ratio of the leaves, the length of the internodes, the stem thickness, the aspect ratio of the fruit, and the aspect ratio of the inner cavity, etc. This embodiment calculates by setting symmetric boundary conditions of the shapes of the plant, leaves, stems, and fruits of the muskmelon to capture all features of the muskmelon, such as the plant, leaves, stems, the distance between each leaf on the plant stem, the fruit size, the appearance, and the shape of the inner cavity.

[0046] Specifically, based on the collected data, CAD software was used to construct a geometric model of the cantaloupe plant, leaves, stems, and fruits. During the modeling process, symmetry needs to be taken into account to ensure that the model can accurately reflect the physical properties of the cantaloupe. The symmetry axis of the model is determined by observing the growth characteristics of the cantaloupe, and boundary conditions are set on the symmetry axis of the model. These conditions will limit the deformation of the model and keep it symmetrical.

[0047] Finite element analysis (FEA) software was used to numerically analyze the geometric model. During the analysis, boundary conditions were applied to simulate the morphological changes of the melon at different growth stages. The entire growth process of the melon from seed to mature plant was simulated by the simulation software; during the simulation process, boundary conditions were continuously adjusted to reflect the actual changes in the growth of the melon. During the simulation process, data on the morphological characteristics of the plant, leaves, stems, fruits, etc. were collected; these data were analyzed to determine the growth patterns and characteristics of the melon.

[0048] The lowest buckling mode of symmetrical structures such as leaves, stems, and fruits of different varieties of watermelon plants can be symmetrical or asymmetrical. Symmetry refers to the phenomenon that the same parts of an object or figure are regularly repeated under certain transformation conditions, that is, the invariant phenomenon under certain transformation conditions. For example, the leaves growing on the main stems of some varieties of watermelon or watermelon grow symmetrically, while the leaves growing on the main stems of some varieties of watermelon or watermelon grow asymmetrically. For example, in the process of watermelon variety selection, a large number of variety resources use specific commands to conduct quantitative and systematic analysis of all characteristics of various indicators such as plants, leaves, stems, distance between each leaf on the plant stem, fruit size, appearance and shape of the inner cavity of different variety resources, providing breeders with a basis for measuring the advantages and disadvantages of different germplasm resources.

[0049] Using the construction module, the analysis model is constructed using the grey analysis decision model. The grey analysis decision model is a decision analysis method based on grey system theory. The grey analysis decision model has the advantages of small amount of data required, simple calculation, and no strict requirements on data distribution. It can process and analyze limited and incomplete data, calculate the correlation between each factor and the decision target, predict the impact of the result on the decision target, and then comprehensively evaluate and compare multiple decision plans to select the best plan. The process of building the model includes:

[0050] First, determine the decision-making goal based on the output results of the above modules, such as selecting the best watermelon variety. Construct an evaluation index system, including all relevant evaluation indicators, such as yield, quality, disease resistance, etc. Dimensionless data is processed and published as follows:

[0051]

[0052] In the formula, x i (k) represents the kth data of the i-th indicator, max{x i (k)} and min{x i (k)} represent x i The maximum and minimum values ​​of (k).

[0053] Create a reference sequence X 0 and compare the series X i , where X 0 represents the ideal state of the decision-making goal; X i Represents the performance of each solution on each evaluation index. The formula for calculating grey correlation degree is:

[0054] First, calculate the absolute value of the difference between the reference sequence and the comparison sequence:

[0055] ΔX 0 (k) = |X 0(k)-X i (k)|

[0056] Then calculate the correlation coefficient:

[0057]

[0058] Where ρ represents the resolution coefficient; its value is between (0, 1); min i ΔX i (k)max i ΔX i (k) represents the minimum and maximum difference between all comparison series and reference series at the kth data.

[0059] Finally, calculate the relevance:

[0060]

[0061] Among them, r i It represents the correlation between the ith solution and the reference sequence, and n represents the number of data.

[0062] The model divides each plan into different categories according to the degree of correlation, usually divided into excellent, good, medium, and poor.

[0063] Finally, the analysis module uses the analysis model to analyze the collected data, extracts the detailed characteristic element indicators of leaves, internodes, and fruits of different cantaloupe resources, performs calculations and analysis through gray analysis models and symmetric simulation, and pushes the decision results.

[0064] Embodiment 2

[0065] like Figure 2 As shown, the grey decision-making and symmetry simulation analysis melon breeding intelligent analysis method provided in this embodiment comprises the following steps:

[0066] S1. Collect image data of cantaloupe.

[0067] Use a 3D fixed-focus lens, choose the appropriate focal length, and take close-up shots using the wide-angle feature to highlight the capture of real morphological features and growth pattern information. These image data are clear and bright photos of the detailed features of plant leaves, internodes, and fruits.

[0068] In the collection area, plants of different varieties and types are photographed one by one to obtain clear and bright image information in order to show the main shape phenotypic characteristics of the plants. The plant body is generally composed of organs such as roots, stems, leaves, flowers, fruits and seeds. The formation of plant phenotypic characteristics is the result of the common presentation of these organs after development. The characteristics of morphological structure are often closely related to morphological traits, reproductive traits and quality traits. For example: the main shape phenotypic characteristics collected by the cantaloupe breeding institute include stems, leaves, flowers, fruits and seeds. The collected cantaloupe images are sorted and classified to ensure that the phenotypic characteristics of each plant have clear data records.

[0069] S2. Extract image features from image data.

[0070] Extract image features from image data. First, preprocess the image data, including image enhancement and image restoration, to improve the visual quality of the image. By increasing contrast, reducing noise and removing artifacts, as well as removing degradation such as blur, noise and distortion from the image.

[0071] The processed image is then divided into regions or segments, each of which corresponds to a specific object or feature in the image. This embodiment adopts a segmentation method based on threshold processing. First, a grayscale histogram is calculated for the image to analyze the distribution of different grayscale values ​​in the image, and a threshold is determined based on the distribution. The image is binarized using the determined threshold and the image is segmented into two parts, the target and the background, based on the result of the threshold processing. Finally, the segmented image is post-processed, such as removing noise, filling holes, etc., to improve the segmentation quality.

[0072] After the image segmentation is completed, boundary conditions are set from the segmented image to identify and extract relevant features, such as the aspect ratio of the leaves, the length of the internodes, the stem thickness, the aspect ratio of the fruit, and the aspect ratio of the inner cavity. In this embodiment, symmetric boundary conditions of the shapes of the plant, leaves, stems, and fruits of the muskmelon are set for calculation to capture all features of the muskmelon plant, leaves, stems, the distance between each leaf on the plant stem, the fruit size, the appearance, and the shape of the inner cavity.

[0073] Specifically, based on the collected data, CAD software was used to construct a geometric model of the cantaloupe plant, leaves, stems, and fruits. During the modeling process, symmetry needs to be taken into account to ensure that the model can accurately reflect the physical properties of the cantaloupe. The symmetry axis of the model is determined by observing the growth characteristics of the cantaloupe, and boundary conditions are set on the symmetry axis of the model. These conditions will limit the deformation of the model and keep it symmetrical.

[0074] Finite element analysis (FEA) software was used to numerically analyze the geometric model. During the analysis, boundary conditions were applied to simulate the morphological changes of the melon at different growth stages. The entire growth process of the melon from seed to mature plant was simulated by the simulation software; during the simulation process, boundary conditions were continuously adjusted to reflect the actual changes in the growth of the melon. During the simulation process, data on the morphological characteristics of the plant, leaves, stems, fruits, etc. were collected; these data were analyzed to determine the growth patterns and characteristics of the melon.

[0075] The lowest buckling mode of symmetrical structures such as leaves, stems, and fruits of different varieties of watermelon plants can be symmetrical or asymmetrical. Symmetry refers to the phenomenon that the same parts of an object or figure are regularly repeated under certain transformation conditions, that is, the invariant phenomenon under certain transformation conditions. For example, the leaves growing on the main stems of some varieties of watermelon or watermelon grow symmetrically, while the leaves growing on the main stems of some varieties of watermelon or watermelon grow asymmetrically. For example, in the process of watermelon variety selection, a large number of variety resources use specific commands to conduct quantitative and systematic analysis of all characteristics of various indicators such as plants, leaves, stems, distance between each leaf on the plant stem, fruit size, appearance and shape of the inner cavity of different variety resources, providing breeders with a basis for measuring the advantages and disadvantages of different germplasm resources.

[0076] S3. Build an analysis model based on image features.

[0077] The grey analysis decision model is used to construct the analysis model. The grey analysis decision model is a decision analysis method based on grey system theory. The grey analysis decision model has the advantages of small amount of data required, simple calculation, and no strict requirements on data distribution. It can process and analyze limited and incomplete data, calculate the correlation between each factor and the decision target, predict the impact of the result on the decision target, and then comprehensively evaluate and compare multiple decision plans to select the best plan. The process of building the model includes:

[0078] First, determine the decision-making goal based on the output results of the above modules, such as selecting the best watermelon variety. Construct an evaluation index system, including all relevant evaluation indicators, such as yield, quality, disease resistance, etc. Dimensionless data is processed and published as follows:

[0079]

[0080] In the formula, x i (k) represents the kth data of the i-th indicator, max{x i (k)} and min{x i (k)} represent x i The maximum and minimum values ​​of (k).

[0081] Create a reference sequence X 0 and compare the series X i , where X 0 represents the ideal state of the decision-making goal; X i Represents the performance of each solution on each evaluation index. The formula for calculating grey correlation degree is:

[0082] First, calculate the absolute value of the difference between the reference sequence and the comparison sequence:

[0083] ΔX 0 (k) = |X 0 (k)-X i (k)|

[0084] Then calculate the correlation coefficient:

[0085]

[0086] Where ρ represents the resolution coefficient; its value is between (0, 1); min i ΔX i (k)max i ΔX i (k) represents the minimum and maximum difference between all comparison series and reference series at the kth data.

[0087] Finally, calculate the relevance:

[0088]

[0089] Among them, r i It represents the correlation between the ith solution and the reference sequence, and n represents the number of data.

[0090] The model divides each plan into different categories according to the degree of correlation, usually divided into excellent, good, medium, and poor.

[0091] S4. Use the analytical model to complete the analysis of muskmelon breeding.

[0092] Finally, the collected data were analyzed using an analytical model, and detailed characteristic element indicators of leaves, internodes, and fruits of different cantaloupe resources were extracted. The grey analysis model and symmetric simulation were used to perform computational analysis and push the decision results.

[0093] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A grey decision-making and symmetry simulation analysis system for melon breeding, characterized in that: include: Acquisition module, extraction module, construction module and analysis module; The acquisition module is used to acquire image data of muskmelon; The extraction module is used to extract image features from the image data; The construction module is used to construct an analysis model based on the image features; The analysis module is used to complete the analysis of muskmelon breeding using the analysis model.

2. The grey decision-making and symmetry simulation analysis melon breeding intelligent analysis system according to claim 1, characterized in that: The workflow of the acquisition module includes: using a 3D fixed-focus lens to shoot plants of different varieties and types one by one in the acquisition area to obtain clear and bright image information for displaying the shape phenotypic characteristics of the plants; the shape phenotypic characteristics include: the results presented by the development of the plant's roots, stems, leaves, flowers, fruits and seeds.

3. The grey decision-making and symmetry simulation analysis melon breeding intelligent analysis system according to claim 1, characterized in that: The workflow of the extraction module includes: first, preprocessing the image data to obtain a processed image; then, using a segmentation method based on threshold processing to divide the processed image into regions or segments, each region or segment corresponds to a specific object or feature in the image; after the image segmentation is completed, the extraction module identifies and extracts relevant features by setting boundary conditions from the segmented image.

4. The grey decision-making and symmetry simulation analysis melon breeding intelligent analysis system according to claim 1, characterized in that: The workflow of the construction module includes: constructing an analysis model using a grey analysis decision model and constructing an evaluation index system: In the formula, x i (k) represents the kth data of the i-th indicator, max{x i (k)} and min{x i (k)} represent x i (k) maximum and minimum values; Create a reference sequence X0 and a comparison sequence X i , where X0 represents the ideal state of the decision-making target; X i Represents the performance of each scheme on each evaluation index, and calculates the grey correlation formula: First, calculate the absolute value of the difference between the reference sequence and the comparison sequence: ΔX0(k)=|X0(k)-X i (k)| Then calculate the correlation coefficient: Where, ρ represents the resolution coefficient; its value is between (0,1); min i ΔX i (k)max i ΔX i (k) represents the minimum and maximum difference between all comparison series and reference series at the kth data respectively; Finally, calculate the relevance: Among them, r i It represents the correlation between the ith solution and the reference sequence, and n represents the number of data.

5. The grey decision-making and symmetry simulation analysis melon breeding intelligent analysis system according to claim 1, characterized in that: The analysis module uses the analysis model to analyze the collected data. Specifically, it extracts the detailed characteristic element indicators of leaves, internodes and fruits of different muskmelon resources, performs calculation analysis through the analysis model and symmetric simulation, and pushes the decision results.

6. A grey decision-making and symmetry simulation analysis method for melon breeding intelligence analysis, the method is applied to the system according to any one of claims 1 to 5, characterized in that the steps include: Collect image data of cantaloupe; extracting image features from the image data; Based on the image features, construct an analysis model; The analysis model is used to complete the analysis of muskmelon breeding.

Images