A multi-dimensional data analysis method for breeding screening

Abstract

The present invention discloses a multi-dimensional data analysis method for breeding screening, which relates to the technical field of breeding screening and includes the following steps: spreading the chili seeds of the batch to be detected on the detection area, and dividing the detection area into multiple sub-areas. By introducing the texture complexity index and the porosity index, and combining with a pre-trained machine learning model, the present invention intelligently analyzes the surface microstructure characteristics of chili seeds, achieving an improvement in the screening accuracy. The system flexibly adjusts the number and position of detection windows, accurately captures complex microstructure characteristics, avoids misjudging high-quality seeds as low-quality seeds, and improves the screening accuracy. At the same time, it intelligently divides simple and complex surface structures, optimizes resource allocation, saves detection costs and time, and provides an efficient and economical screening scheme for the breeding process.

Description

Technical Field

[0001] The present invention relates to the technical field of breeding and screening, and in particular to a multidimensional data analysis method for breeding and screening. Background Art

[0002] Multidimensional data analysis for breeding screening refers to the collection and processing of data from multiple dimensions to evaluate and select seeds with superior genetic characteristics. In the screening process of pepper seeds, this analysis can help breeders comprehensively consider the physical characteristics of seeds, such as seed size, shape, color, density, and moisture content, and then judge their germination potential and growth performance. With high-precision pepper seed screening equipment, breeders can quickly measure and analyze these physical characteristics, thereby improving screening efficiency and accuracy.

[0003] In pepper seed quality screening, multidimensional data analysis can reveal the relationship between different seed characteristics and help breeders identify the best seed sources. For example, using optical sensors and imaging technology, pepper seed screening equipment can capture the surface characteristics and internal structure of seeds to assess their quality. Combined with machine learning algorithms, these data can not only be used to monitor the physical properties of seeds in real time, but also predict the performance of different seeds in the future growth process through the analysis of historical data, and further optimize breeding decisions.

[0004] The prior art has the following deficiencies:

[0005] The surface microstructure of pepper seeds refers to the subtle morphology and texture characteristics of the outer surface of the seeds, including surface bumps, pores and lines. These microstructures not only affect the appearance and touch of seeds, but also play an important role in the germination and growth of seeds. For example, certain microstructures help to enhance the adhesion between seeds and soil, improve the permeability of water and air, and thus promote rapid germination of seeds under suitable conditions. In addition, microstructures may also affect the resistance of seeds to diseases and environmental stress, thereby improving the survival rate and healthy growth of seeds.

[0006] In existing image recognition technology, a fixed number of observation areas are usually randomly selected for a batch of pepper seeds, and high-resolution image acquisition is performed to identify the surface microstructure features of the seeds. However, when the surface microstructure of some pepper seeds is too complex, the fixed number of observation areas may not be able to accurately capture these features. This deficiency may cause high-quality seeds to be misjudged as low-quality seeds, thus affecting the screening results, and then using unsuitable seeds during the planting process, ultimately affecting the growth and yield of the crop.

[0007] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the Invention

[0008] The object of the present invention is to provide a multi-dimensional data analysis method for breeding screening. By introducing a texture complexity index and a porosity index, and combining a pre-trained machine learning model to intelligently analyze the surface microstructure characteristics of pepper seeds, the screening accuracy is improved. The system flexibly adjusts the number and position of detection windows, accurately captures complex microstructure characteristics, avoids misjudging high-quality seeds as low-quality seeds, and improves the screening accuracy. At the same time, it intelligently divides simple and complex surface structures, optimizes resource allocation, saves detection costs and time, and provides an efficient and economical screening solution for the breeding process to solve the problems in the above-mentioned background technology.

[0009] To achieve the above object, the present invention provides the following technical solution: A multi-dimensional data analysis method for breeding screening, comprising the following steps:

[0010] Lay the pepper seeds of the batch to be detected flat in the detection area, and divide the detection area into multiple sub-areas;

[0011] In the divided sub-areas, randomly extract sub-areas based on the preset number of detection windows for data collection to obtain the seed surface structure information;

[0012] Extract features from the seed surface structure information in the sub-areas. Based on the extracted seed surface microstructure characteristics, under the detection window, use the pre-learned machine learning model to intelligently evaluate the complexity of the seed surface structure in this sub-area and identify different surface structure types of the seeds;

[0013] Based on the evaluation results of the machine learning model, divide the surface structure of pepper seeds into complex surface structures and simple surface structures;

[0014] For the identified simple surface structures, continue to randomly extract sub-areas for data collection based on the preset number of detection windows;

[0015] For complex surface structures, based on the preset number of detection windows and according to the evaluation results, dynamically adjust the number and position of detection windows to comprehensively capture complex microstructure characteristics.

[0016] Preferably, feature extraction is performed on the seed surface structure information in the sub-region. The extracted features include the degree of diversity change of the surface texture and the distribution characteristics of the number of pores on the surface. Under the monitoring window, after analyzing the degree of diversity change of the extracted surface texture and the distribution characteristics of the number of pores on the surface, a texture complexity index and a porosity index are generated respectively. The texture complexity index is used to measure the diversity and irregularity of the seed surface texture, and the porosity index is used to measure the number, size and distribution characteristics of the pores on the seed surface. The texture complexity index and the porosity index are input into a pre-trained machine learning model, and a microstructure coefficient is generated through the machine learning model. The complexity of the seed surface structure is intelligently evaluated through the microstructure coefficient, and different surface structure types of the seeds are identified.

[0017] Preferably, under the detection window, the specific steps for analyzing the degree of diversity change of the surface texture and generating the texture complexity index are as follows:

[0018] Under the detection window, the seed surface data collected is preprocessed, and the texture features are extracted using the gray-level co-occurrence matrix technique for the preprocessed data. The calculation expression is as follows:

[0019]

[0020] , where G(i,j) is an element of the gray-level co-occurrence matrix, representing the frequency of adjacent occurrences of pixels with gray value i and gray value j in the image, and Q(i,j|x,y) represents the conditional probability of the occurrence of pixel pairs with gray value i and gray value j at position (x,y);

[0021] The texture energy and texture contrast are calculated using the extracted gray-level co-occurrence matrix. The calculation expression for the texture energy is as follows:

[0022]

[0023] where E is the texture energy, measuring the uniformity of the texture;

[0024] The calculation expression for the texture contrast is as follows:

[0025]

[0026] , where C is the texture contrast, describing the degree of change of the texture;

[0027] The entropy value of the texture is calculated to further evaluate the complexity of the texture. The calculation expression is as follows:

[0028]

[0029] Wherein, H is the texture entropy value, and ∈ is a small constant used to avoid the zero value problem in logarithmic calculations;

[0030] Combining the texture energy E, texture contrast C, and texture entropy value H, a texture complexity index is generated, and the calculation expression is as follows:

[0031]

[0032] Wherein, TCI is the texture complexity index, max(E) is the maximum value of texture energy, max(C) is the maximum value of texture contrast, max(H) is the maximum value of texture entropy, A is the adjustment parameter of texture energy E, B is the adjustment parameter of texture contrast C, F is the power adjustment parameter of texture contrast, δ is the offset adjustment parameter of texture entropy H, λ is the power adjustment parameter of texture entropy H, and η is a non-zero positive number used to control the smoothness of the final result.

[0033] Preferably, under the detection window, the specific steps for analyzing the distribution characteristics of the number of pores on the surface and generating the porosity index are as follows:

[0034] Under the detection window, the acquired image information is preprocessed, and the preprocessed image information is used with an image processing algorithm to identify pores. The expression is as follows:

[0035]

[0036] Wherein, A p is the total pore area, N p is the number of pores, A y is the area of the y-th pore;

[0037] The calculation formula for porosity is:

[0038]

[0039] Wherein, P is the porosity, and A t is the total area of the observation region;

[0040] Based on the porosity, the shape factor of the pores is introduced, and the porosity is corrected by its influence. The calculation expression is as follows:

[0041]

[0042] Wherein, S is the shape factor, N p is the number of pores, L y is the perimeter of the y-th pore, and A y is the area of the y-th pore;

[0043] Based on the influence of the shape factor S, the porosity is corrected. The calculation expression is as follows:

[0044]

[0045] Wherein, P adj is the corrected porosity;

[0046] Finally, a porosity index is generated, and the calculation expression is as follows:

[0047]

[0048] Wherein, PI is the porosity index, k is the normalization constant, 1 - P adj represents the proportion of the non-porous area in the detection area, and ω is the adjustment parameter, used to adjust the influence of the corrected porosity P adj of, ω is used to adjust the influence of the non-porous area in the detection area, e -ρS is the exponential weight of the shape factor, ρ is the weight of the shape factor, e is the natural base, and δ is the uniformity adjustment parameter.

[0049] Preferably, after analyzing the seed surface structure information in the sub-region under the detection window, the generated microstructure coefficient is compared with a preset microstructure coefficient reference threshold to classify the seed surface structure state in the sub-region;

[0050] If the microstructure coefficient is greater than or equal to the preset microstructure coefficient reference threshold, the seed surface structure state of this sub-region is classified as a complex surface structure;

[0051] If the microstructure coefficient is less than the preset microstructure coefficient reference threshold, the seed surface structure state of this sub-region is classified as a simple surface structure.

[0052] Preferably, for the complex surface structure, based on the preset number of detection windows, according to the evaluation result, the number and position of the detection windows are dynamically adjusted to comprehensively capture the complex microstructure features. The specific steps are as follows:

[0053] According to the evaluation result, that is, the complex surface structure, the number of detection windows is dynamically adjusted, considering the influence of the changes in the texture complexity index TCI and the porosity index PI on the number of detection windows, and a weighted adjustment method is adopted. The calculation expression is as follows:

[0054]

[0055] Wherein, N adjusted is the adjusted number of detection windows, N initial is the preset number of detection windows, TCI rt is the preset texture complexity index reference threshold, PI rtis a preset reference threshold for porosity index, and α and β are adjustment factors, respectively reflecting the influence of texture complexity and porosity change on the number of detection windows;

[0056] After adjusting the number of windows, determine the positions of the new windows. Based on the distribution of complex surface features, identify the regions with significant changes in surface features through the clustering analysis method. The expression is as follows:

[0057] P selected =Top(N adjusted ,{P r |r∈rank(C r )})

[0058] In the formula, P selected is the set of selected detection window positions, P r is the candidate detection window position, the r-th position in the set of all candidate detection window positions, P r ={P1, P2, P3, ……, P u}, u is the total number of candidate detection windows, C r is the complexity evaluation of the candidate detection window P r , is the complexity index of the r-th candidate detection window position P r , rank(C r ) is the sorting of the complexity evaluation;

[0059] Based on the feedback of complex surface features, re-evaluate the positions of the windows. If some positions do not fully cover the features, position adjustment can be performed. The calculation expression is as follows:

[0060] P revised =P selected +ΔP

[0061] In the formula, P revised is the adjusted detection window position, and ΔP is the position offset used to adjust the window position after feedback;

[0062] Finally, according to the new number and positions of the windows, perform loop optimization. The calculation expression is as follows:

[0063]

[0064] , where N final is the finally adjusted number of detection windows, γ is the feedback adjustment factor, C new is the current complexity index, and C old is the complexity index of the previous evaluation.

[0065] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0066] By introducing the texture complexity index and porosity index and inputting them into a pre-trained machine learning model, the present invention can intelligently analyze the surface microstructure characteristics of pepper seeds. This method enables the screening process to not only focus on the basic morphology of the seed surface but also accurately identify different microstructure types and their complexity. This automated evaluation process avoids the limitations of traditional random sampling methods and reduces the risk of high-quality seeds being misjudged as inferior seeds due to their complex microstructures not being fully captured. At the same time, by dynamically adjusting the number and position of detection windows, the system can be flexibly adjusted according to the actual complexity of the microstructure to ensure that complex microstructure characteristics are fully captured, thereby improving the accuracy and reliability of the screening process.

[0067] By intelligently classifying the complexity of the seed surface structure and dividing the surface microstructure into "complex surface structure" and "simple surface structure", the present invention can allocate suitable acquisition and detection resources for different seed characteristics. For simple surface structures, only random acquisition based on a fixed number of detection windows is required, saving unnecessary resource investment; while for complex surface structures, by dynamically adjusting the number and position of the detection windows for acquisition, the seed characteristics can be analyzed more comprehensively to achieve efficient utilization of resources. This method significantly improves the overall detection efficiency, not only reducing the time and cost consumption during the screening process but also ensuring that resources can be concentrated on the identification of complex surface structures, ultimately providing a high-quality and cost-effective screening solution for the breeding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0069] Figure 1 It is a method flow chart of a multi-dimensional data analysis method for breeding screening according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0070] Now, the exemplary embodiments will be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more complete and comprehensive, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0071] The present invention provides a multi-dimensional data analysis method for breeding screening as shown in Figure 1 and includes the following steps:

[0072] Lay the chili seeds of the batch to be detected flat in the detection area, and divide the detection area into multiple sub-areas;

[0073] Divide the detection area into multiple sub-areas. The division is based on the quantity and size of the seeds. Usually, a grid method is adopted to make the size of each sub-area appropriate for subsequent random sampling. The divided sub-areas not only contribute to the systematic management of samples, but also improve the efficiency and accuracy of image acquisition, ensuring that the collected data can comprehensively reflect the characteristics of the entire batch of seeds.

[0074] In the divided sub-areas, randomly select sub-areas for data collection based on the preset number of detection windows to obtain the surface structure information of the seeds;

[0075] In the divided sub-areas, conduct random sampling based on the preset number of detection windows. This process aims to ensure that each sub-area can represent the characteristics of the entire batch of seeds. In specific operations, a random number generation algorithm can be used to select specific sub-areas so that the probability of each sub-area being selected is equal. Such random sampling not only avoids sample selection bias, but also effectively covers the surface characteristics of various types of seeds, thereby enhancing the representativeness and reliability of the data.

[0076] After random sampling, enter the data collection stage. Usually, a high-resolution image acquisition device, such as a microscope or a high-resolution camera, is used to take images of the selected sub-areas. At this time, it is necessary to ensure that the light source is uniform and the focal length is appropriately adjusted to clearly capture the microstructural characteristics on the seed surface. Through image processing software, the collected images can be further preprocessed, such as denoising and enhancing contrast, to improve the accuracy of subsequent feature extraction. This method combining random sampling and efficient data collection helps to obtain high-quality sample data and lay a solid foundation for subsequent analysis and evaluation.

[0077] Extract the features of the surface structure information of the seeds in the sub-areas. Based on the extracted microstructural characteristics of the seed surface, under the detection window, use the pre-trained machine learning model to intelligently evaluate the complexity of the surface structure of the seeds in this sub-area and identify different surface structure types of the seeds;

[0078] Extract the feature of the seed surface structure information under the sub-region. The extracted features include the degree of diversity change of the surface texture and the distribution characteristics of the number of pores existing on the surface. Under the monitoring window, after analyzing the degree of diversity change of the extracted surface texture and the distribution characteristics of the number of pores existing on the surface, a texture complexity index and a porosity index are generated respectively. The texture complexity index is used to measure the diversity and irregularity of the seed surface texture, and the porosity index is used to measure the number, size and distribution characteristics of the pores existing on the seed surface. Input the texture complexity index and the porosity index into the pre-trained machine learning model, and generate a microstructure coefficient through the machine learning model. Evaluate the complexity of the seed surface structure intelligently through the microstructure coefficient, and identify different surface structure types of the seeds;

[0079] When the degree of diversity change of the surface texture obtained under the sub-region is abnormal, it can indeed indicate that the microstructure of the seed surface in this sub-region is in a complex state. This is because the surface microstructure of the seed consists of various morphological features, including an uneven surface, pores of different sizes and shapes, and diverse texture patterns. Under normal circumstances, the diversity of the surface texture will be maintained within a certain range, reflecting the health status and consistency of the seeds. However, when this diversity increases significantly or shows abnormalities, it means that the morphological features on the seed surface may become more complex, which may be caused by genetic mutations, environmental factors, or abnormal changes during the seed development process. The complex microstructure may include more delicate patterns, abnormal protrusions, and higher roughness, etc., making the identification and classification of surface features more difficult. Therefore, the abnormal surface texture diversity not only reflects the complexity of the microstructure, but may also indicate significant differences in the morphology of the seeds, which complicates the judgment of seed quality, thus affecting the effective screening and evaluation of the seeds in this sub-region. By monitoring this change, seeds with complex surface microstructures can be better identified, providing an important basis for subsequent research and applications.

[0080] Under the detection window, the specific steps for analyzing the degree of diversity change of the surface texture and generating the texture complexity index are as follows:

[0081] Under the detection window, preprocess the collected seed surface data, and use the gray-level co-occurrence matrix (GLCM) technology to extract texture features for the preprocessed data. The calculation expression is as follows:

[0082]

[0083] , where G(i, j) is an element of the gray-level co-occurrence matrix, representing the frequency of occurrence of pixels with gray values i and j adjacent to each other in the image, and Q(i, j|x, y) represents the conditional probability of the occurrence of pixel pairs with gray values i and j at position (x, y);

[0084] The gray-level co-occurrence matrix (GLCM) is a statistical method used to describe the gray-level spatial relationship of an image. It reflects the texture features of the image by calculating the co-occurrence frequency of pixel pairs with different gray levels in the image. Specifically, each element G(i, j) of the gray-level co-occurrence matrix represents the number of times or the probability of the occurrence of pixel pairs with gray values i and j adjacent to each other in the image. By defining the relative positions between pixels (such as horizontal, vertical, or diagonal directions), the gray-level co-occurrence matrix can reveal the spatial relationship between pixels in the image and capture the texture information of the image.

[0085] Here, the role of the gray-level co-occurrence matrix is to extract the texture features of the surface of pepper seeds for generating the texture complexity index. By calculating indicators such as the energy, contrast, and entropy of the gray-level co-occurrence matrix, the texture complexity of the seed surface can be quantified, and simple and complex surface structures can be distinguished. These texture features can help evaluate the quality of the seeds because the texture characteristics of the seed surface are closely related to its germination and growth potential. As a basic tool for texture analysis, the gray-level co-occurrence matrix can provide key data support for seed screening.

[0086] Calculate the texture energy and texture contrast using the extracted gray-level co-occurrence matrix. The expression for calculating the texture energy is as follows:

[0087]

[0088] , where E is the texture energy, measuring the uniformity of the texture;

[0089] The expression for calculating the texture contrast is as follows:

[0090]

[0091] , where C is the texture contrast, describing the degree of change of the texture;

[0092] The texture energy is used to measure the uniformity and repeatability of the texture in the image. The higher the texture energy value, the higher the frequency of occurrence of pixel pairs with the same gray value in the image, that is, the more uniform and repetitive the texture of the image. For the surface texture of pepper seeds, a high texture energy usually means that the surface structure is relatively flat or regular, which may indicate that the seeds have better quality because surface uniformity may contribute to the uniform absorption of water and nutrients.

[0093] Texture contrast is used to describe the degree of texture change in an image, reflecting the size of the grayscale value difference in the image. The higher the contrast value, the greater the grayscale difference in the image, and the more complex or irregular the texture. For the seed surface, a higher texture contrast may mean that there are more subtle bumps or texture features on the surface, which may affect the contact between the seed and the soil and its ability to absorb water. Combining texture energy and contrast can more comprehensively evaluate the surface quality of seeds, thereby helping to screen out seeds with excellent germination potential.

[0094] Calculate the entropy value of the texture and further evaluate the complexity of the texture. The calculation expression is as follows:

[0095]

[0096] , where H is the texture entropy value, ∈ is a small constant used to avoid the zero value problem in logarithmic calculation;

[0097] Combining texture energy E, texture contrast C and texture entropy value H, the texture complexity index is generated. The calculation expression is as follows:

[0098]

[0099] , where TCI is the texture complexity index, max(E) is the maximum texture energy, max(C) is the maximum texture contrast, max(H) is the maximum texture entropy, A is the adjustment parameter of texture energy E, B is the adjustment parameter of texture contrast C, F is the power adjustment parameter of texture contrast, δ is the offset adjustment parameter of texture entropy value H, λ is the power adjustment parameter of texture entropy value H, and η is a non-zero positive number used to control the smoothness of the final result;

[0100] The expression for calculating the texture complexity index shows that, under the detection window, the larger the value of the texture complexity index generated after analyzing the degree of diversity change of the surface texture, the more complex the seed surface microstructure is. This is because a higher complexity index usually reflects the presence of diverse morphological features on the seed surface, such as bumps and lines of different shapes, sizes and arrangements, indicating that the complexity of the surface structure has increased. Relatively speaking, a lower complexity index indicates that the surface texture is relatively simple, lacking obvious diversity and change, indicating that the seed surface microstructure is in a simple structure. Therefore, the texture complexity index can effectively quantify and reflect the complexity of the seed surface microstructure, thereby providing an important basis for the analysis of seed characteristics.

[0101] When the distribution characteristics of the number of pores on the seed surface obtained in a sub-region show anomalies, this usually indicates that the microstructure of the seed surface in this sub-region is in a complex state. The number and distribution characteristics of pores are one of the important indicators for evaluating the surface microstructure. Under normal circumstances, the pores on the seed surface should exhibit a certain regularity and uniformity. However, when the number of pores significantly increases or decreases, or their distribution shows uneven and randomized characteristics, it indicates that the formation of the surface microstructure may be affected by various factors, such as genetic variation, environmental conditions, or different growth stages. Such abnormal pore distributions often reflect complex changes in the surface in terms of morphology, texture, or physical properties, which may lead to an increase in the diversity of surface features and the formation of complex microstructures. Complex microstructures may include pores of multiple sizes and shapes, and this diversity makes the interaction between the surface and the surrounding environment more complex. Therefore, the abnormal distribution characteristics of the number of pores are not only a direct manifestation of the complexity of the microstructure but also a response of the seed to environmental stress or other biological factors, further emphasizing the complex state of the seed surface microstructure.

[0102] Under the detection window, the specific steps for analyzing the distribution characteristics of the number of pores existing on the surface and generating the porosity index are as follows:

[0103] Under the detection window, preprocess the acquired image information, and use an image processing algorithm to identify pores for the preprocessed image information. The expression is as follows:

[0104]

[0105] In the formula, A p is the total pore area, N p is the number of pores, and A y is the area of the y-th pore;

[0106] The calculation formula for porosity is:

[0107]

[0108] In the formula, P is the porosity, and A t is the total area of the observation region;

[0109] Based on the porosity, introduce the shape factor of the pores and correct it through its influence on the porosity. The calculation expression is as follows:

[0110]

[0111] In the formula, S is the shape factor, N p is the number of pores, L y is the perimeter of the y-th pore, and A y is the area of the y-th pore;

[0112] The shape factor of pores plays a crucial role in the calculation of porosity index, as it not only reflects the geometric characteristics of pores but also affects the permeability of pores to water, air, and nutrients. The shape factor is usually represented by the ratio of the perimeter to the area of pores. Pores with more complex shapes generally have a larger surface area-to-volume ratio, which helps to increase the contact area between seeds and soil, enhancing their adhesion and stability. Additionally, pores with complex shapes can often better regulate the exchange of water and gases, thus enhancing the adaptability and resistance of seeds during the growth process. Therefore, the introduction of the shape factor enables the porosity index to more comprehensively reflect the actual performance of seeds and provide a more accurate quality assessment.

[0113] Based on the influence of the shape factor S, the porosity is corrected, and the calculation expression is as follows:

[0114]

[0115] In the formula, P adj is the corrected porosity;

[0116] Finally, the porosity index is generated, and the calculation expression is as follows:

[0117]

[0118] In the formula, PI is the porosity index, k is the normalization constant used to adjust the final index to a suitable range, 1 - P adj represents the proportion of the non-pore area in the detection area, and ω are adjustment parameters, used to adjust the influence of the corrected porosity P adj and ω is used to adjust the influence of the non-pore area in the detection area. e -ρS is the exponential weight of the shape factor, ρ is the weight of the shape factor, e is the natural base, and δ is the uniformity adjustment parameter used to control the influence of the pore area distribution uniformity on the index;

[0119] It can be seen from the calculation expression of the porosity index that under the detection window, the larger the value of the porosity index generated after analyzing the distribution characteristics of the number of pores on the surface, the more complex the microstructure of the seed surface is usually. This is because a higher porosity means there are more pores and irregularities on the surface, and this diversity reflects the complexity and variability of the surface structure. In contrast, if the porosity index is low, it indicates that there are fewer pores and a more regular morphology on the seed surface, showing a relatively simple microstructure.

[0120] The machine learning model is not limited herein, and any machine learning model that can comprehensively analyze the texture complexity index TCI and the porosity index PI to generate the microstructure coefficient MC is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation manner:

[0121] The calculation formula for generating the microstructure coefficient MC is as follows:

[0122]

[0123] In the formula, k1 and k2 are respectively the preset proportionality coefficients of the texture complexity index TCI and the porosity index PI, and both k1 and k2 are greater than 0.

[0124] It can be seen from the calculation expression of the microstructure coefficient that under the detection window, the larger the value of the texture complexity index generated after analyzing the degree of diversity change of the surface texture, and the larger the value of the porosity index generated after analyzing the distribution characteristics of the pore quantity existing on the surface, that is, the larger the value of the microstructure coefficient generated under the detection window, it indicates that the microstructure of the seed surface in this sub-region is in a complex state, and vice versa, it indicates that the microstructure of the seed surface in this sub-region is relatively simple.

[0125] Based on the evaluation results of the machine learning model, the surface structure of pepper seeds is divided into a complex surface structure and a simple surface structure;

[0126] The microstructure coefficient generated after analyzing the seed surface structure information in the sub-region under the detection window is compared and analyzed with the preset microstructure coefficient reference threshold to divide the state of the seed surface structure in the sub-region;

[0127] If the microstructure coefficient is greater than or equal to the preset microstructure coefficient reference threshold, the state of the seed surface structure in this sub-region is divided into a complex surface structure;

[0128] If the microstructure coefficient is less than the preset microstructure coefficient reference threshold, the state of the seed surface structure in this sub-region is divided into a simple surface structure;

[0129] The complex surface structure refers to the presence of diverse and irregular morphological features on the surface of pepper seeds, including pores of various sizes and shapes, uneven textures, and complex microstructures. The simple surface structure, on the other hand, shows relatively uniform and regular surface features, usually including fewer pores and relatively smooth textures;

[0130] For the identified simple surface structure, continue to randomly extract sub-regions for data collection based on the preset number of detection windows;

[0131] This process can effectively reduce the interference to complex structures while ensuring the continuity and stability of the screening process. Through continuous acquisition, the screening strategy of seeds can be further verified and optimized. This process can effectively reduce the interference to complex structures while ensuring the continuity and stability of the screening process. Through continuous acquisition, the screening strategy of seeds can be further verified and optimized.

[0132] For complex surface structures, based on the preset number of detection windows, multi-scale image analysis is used. According to the evaluation results, the number and position of the detection windows are dynamically adjusted to comprehensively capture complex micro-structure features;

[0133] For complex surface structures, based on the preset number of detection windows, according to the evaluation results, the number and position of the detection windows are dynamically adjusted to comprehensively capture complex micro-structure features. The specific steps are as follows:

[0134] According to the evaluation results, that is, the complex surface structure, the number of detection windows is dynamically adjusted. Considering the influence of the changes in the texture complexity index TCI and the porosity index PI on the number of detection windows, a weighted adjustment method is adopted, and the calculation formula is as follows:

[0135]

[0136] In the formula, N adjusted is the adjusted number of detection windows, N initial is the preset number of detection windows, TCI rt is the preset reference threshold of the texture complexity index, PI rt is the preset reference threshold of the porosity index, and α and β are adjustment factors, respectively reflecting the influence of texture complexity and porosity changes on the number of detection windows;

[0137] After adjusting the number of windows, determine the position of the new window. Based on the distribution of complex surface features, identify the regions with large changes in surface features through the clustering analysis method. The formula is as follows:

[0138] P selected = Top(N adjusted ,{P r |r∈rank(C r )})

[0139] In the formula, P selected is the set of selected detection window positions, P r is the candidate detection window position, the r-th position in the set of all candidate detection window positions, P r ={P1, P2, P3, ……, P u}, u is the total number of candidate detection windows, C r is the candidate detection window P rThe complexity evaluation is the position P of the r-th candidate detection window r The complexity index, rank(C r ) is the ranking of the complexity evaluation;

[0140] Select the N adjusted (i.e., the adjusted number of detection windows) positions with the highest complexity from the candidate detection window positions to form P selected (the selected detection window) set. This method ensures that under limited resources, the most complex and representative surface areas can be preferentially detected, so as to comprehensively capture the complex microstructural features of the seed surface and improve the accuracy of analysis.

[0141] Based on the feedback of the complex surface features, re-evaluate the positions of the windows. If some positions do not fully cover the features, the positions can be adjusted. The calculation formula is as follows:

[0142] P revised = P selected + ΔP

[0143] In the formula, P revised is the adjusted detection window position, and ΔP is the position offset used to adjust the window position after feedback;

[0144] Finally, according to the new number and position of the windows, perform cyclic optimization. The calculation formula is as follows:

[0145]

[0146] In the formula, N final is the finally adjusted number of detection windows, γ is the feedback adjustment factor, and this parameter controls the sensitivity of the adjustment amplitude, that is, when the surface complexity changes greatly, the degree of adjusting the number of windows, C new is the current complexity index, C old is the complexity index of the previous evaluation, which is the result of the previous complexity analysis and serves as a benchmark for complexity changes.

[0147] By introducing the texture complexity index and the porosity index and inputting them into a pre-trained machine learning model, the present invention can perform intelligent analysis on the surface microstructural features of pepper seeds. This method enables the screening process to not only focus on the basic morphology of the seed surface but also accurately identify different microstructural types and their complexities. This automated evaluation process avoids the limitations of traditional random sampling methods and reduces the risk of high-quality seeds being misjudged as inferior seeds because their complex microstructures are not fully captured. At the same time, by dynamically adjusting the number and position of the detection windows, the system can be flexibly adjusted according to the actual complexity of the microstructures to ensure that the complex microstructural features are fully captured, thereby improving the accuracy and reliability of the screening process.

[0148] The present invention intelligently classifies the complexity of the seed surface structure, dividing the surface microstructure into "complex surface structure" and "simple surface structure", which can allocate suitable acquisition and detection resources for different seed characteristics. For the simple surface structure, only random acquisition based on a fixed number of detection windows is required, saving unnecessary resource investment; while for the complex surface structure, by dynamically adjusting the number and position of the detection windows for acquisition, the seed characteristics can be analyzed more comprehensively, realizing the efficient utilization of resources. This method significantly improves the overall detection efficiency, not only reducing the time and cost consumption during the screening process, but also ensuring that resources can be concentrated on the identification of complex surface structures, ultimately providing a high-quality and cost-effective screening solution for the breeding process.

[0149] Only some exemplary embodiments of the present invention have been described in the above general manner. Undoubtedly, for those of ordinary skill in the art, various different ways can be used to modify the described embodiments without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A multi-dimensional data analysis method for breeding screening, characterized in that, Including the following steps: Lay the chili seeds of the batch to be detected flat in the detection area, and divide the detection area into multiple sub-areas; In the divided sub-areas, randomly select sub-areas based on the preset number of detection windows for data collection to obtain the seed surface structure information; Extract features from the seed surface structure information in the sub-areas. Based on the extracted seed surface microstructure features, under the detection window, use the pre-trained machine learning model to intelligently evaluate the complexity of the seed surface structure in this sub-area and identify different surface structure types of the seeds; Based on the evaluation results of the machine learning model, divide the surface structure of the chili seeds into complex surface structures and simple surface structures; For the identified simple surface structures, continue to randomly select sub-areas for data collection based on the preset number of detection windows; For the complex surface structures, based on the preset number of detection windows and according to the evaluation results, dynamically adjust the number and position of the detection windows to comprehensively capture the complex microstructure features; Extract features from the seed surface structure information in the sub-areas. The extracted features include the degree of diversity change of the surface texture and the distribution characteristics of the number of pores on the surface. Under the monitoring window, after analyzing the degree of diversity change of the extracted surface texture and the distribution characteristics of the number of pores on the surface, generate a texture complexity index and a porosity index respectively. The texture complexity index is used to measure the diversity and irregularity of the seed surface texture, and the porosity index is used to measure the number, size and distribution characteristics of the pores on the seed surface. Input the texture complexity index and the porosity index into the pre-trained machine learning model, generate a microstructure coefficient through the machine learning model, and use the microstructure coefficient to intelligently evaluate the complexity of the seed surface structure and identify different surface structure types of the seeds.

2. The multi-dimensional data analysis method for breeding screening according to claim 1, characterized in that Under the detection window, the specific steps for analyzing the degree of diversity change of the surface texture to generate the texture complexity index are as follows: Under the detection window, preprocess the collected seed surface data, and use the gray-level co-occurrence matrix technology to extract texture features for the preprocessed data. The calculation expression is as follows: ， In the formula, is an element of the gray-level co-occurrence matrix, representing the frequency of adjacent occurrences of pixels with gray-level value and gray-level value in the image. represents the conditional probability of the occurrence of a pixel pair with gray-level value at position and ; Use the extracted gray-level co-occurrence matrix to calculate the texture energy and texture contrast. The texture energy calculation expression is as follows: ， In the formula, is the texture energy, which measures the uniformity of the texture; The texture contrast calculation expression is as follows: , where is the texture contrast, which describes the degree of texture change; Calculate the entropy value of the texture to further evaluate the complexity of the texture. The calculation expression is as follows: ， In the formula, is the texture entropy value, is a small constant used to avoid the zero-value problem in logarithmic calculations; Combined texture energy , texture contrast and texture entropy value , generate a texture complexity index, and the calculation expression is as follows: ， Wherein, is the texture complexity index, is the maximum texture energy, is the maximum texture contrast, is the maximum texture entropy value, is the texture energy of the adjustment parameter, is the texture contrast of the adjustment parameter, is the texture contrast power adjustment parameter, is the texture entropy value of the offset adjustment parameter, is the texture entropy value of the power adjustment parameter, is a non-zero positive number used to control the smoothness of the final result.

3. A multi-dimensional data analysis method for breeding screening according to claim 1, characterized in that Under the detection window, the specific steps for analyzing the distribution characteristics of the number of pores on the surface to generate the porosity index are as follows: Under the detection window, preprocess the obtained image information, and use an image processing algorithm to identify pores for the preprocessed image information. The expression is as follows: ， In the formula, is the total pore area, is the number of pores, is the -th pore area; The calculation formula for porosity is: ， In the formula, is the porosity, is the total area of the observation region; On the basis of porosity, introduce the shape factor of the pores and correct it through its influence on porosity. The calculation expression is as follows: ， Wherein, is the shape factor, is the number of pores, is the perimeter of the th pore, and is the area of the th pore; Porosity correction is performed based on the influence of the shape factor , and the calculation expression is as follows: ， In the formula, is the corrected porosity; Finally, generate the porosity index. The calculation expression is as follows: , where is the porosity index, is the normalization constant, represents the proportion of non-porous regions in the detection area, and are adjustment parameters, used to adjust the influence of the corrected porosity , used to adjust the influence of non-porous regions in the detection area, is the exponential weight of the shape factor, is the weight of the shape factor, is the natural base, is the uniformity adjustment parameter.

4. A multi-dimensional data analysis method for breeding screening according to claim 1, characterized in that: Compare and analyze the microstructure coefficient generated after analyzing the seed surface structure information in the sub-areas under the detection window with the pre-set microstructure coefficient reference threshold to divide the seed surface structure state in the sub-areas; If the microstructure coefficient is greater than or equal to a preset reference threshold of the microstructure coefficient, the surface structure state of the sub-region seed is classified as a complex surface structure; If the microstructure coefficient is less than the preset reference threshold of the microstructure coefficient, the surface structure state of the sub-region seed is classified as a simple surface structure.

5. A multi-dimensional data analysis method for breeding screening according to claim 4, characterized in that: For a complex surface structure, based on the preset number of detection windows, the number and position of the detection windows are dynamically adjusted according to the evaluation results to comprehensively capture complex microstructure features. The specific steps are as follows: Dynamically adjust the number of detection windows according to the evaluation results, i.e., complex surface structures, considering the changes in the texture complexity index and the porosity index on the number of detection windows, and adopt a weighted adjustment method. The calculation expression is as follows: ， In the formula, is the adjusted number of detection windows, is the preset number of detection windows, is the preset reference threshold of the texture complexity index, is the preset reference threshold of the porosity index, and are adjustment factors, respectively reflecting the influence of the changes in texture complexity and porosity on the number of detection windows; After adjusting the number of windows, determine the position of the new window. Based on the distribution of complex surface features, identify the regions with large changes in surface features through a clustering analysis method. The expression is as follows: ， In the formula, is the set of selected detection window positions, is the candidate detection window position, the th position in the set of positions of all candidate detection windows, , is the total number of candidate detection windows, is the complexity evaluation of the candidate detection window , which is the complexity index of the th candidate detection window position ; is the sorting of the complexity evaluation. According to the feedback of complex surface features, re-evaluate the position of the window. If some positions do not fully cover the features, position adjustment can be carried out. The calculation expression is as follows: ， In the formula, is the adjusted detection window position, is the position offset used to adjust the window position after feedback; Finally, according to the new number and position of the windows, perform cyclic optimization. The calculation expression is as follows: ， In the formula, is the number of detection windows after the final adjustment, is the feedback adjustment factor, is the current complexity index, is the complexity index of the previous evaluation.

Images