Spinach sowing positioning method and system based on image data analysis

Abstract

The invention discloses a spinach sowing positioning method and system based on image data analysis, and relates to the technical field of spinach sowing positioning, and the method comprises the steps: obtaining a soil image, determining a screenshot size according to an actual area, and intercepting a plurality of image blocks for analysis; analyzing the soil color and texture features of each block, and calculating an average value to construct a soil comprehensive feature representation model; adjusting model parameters to obtain a soil comprehensive characteristic expression model for comparison and a corresponding sowing interval; classifying the models, establishing a reference expression model set, analyzing and determining mark factor parameters of characteristic factors, and constructing a reference expression model library; analyzing a currently shot soil image, constructing a current soil comprehensive feature representation model, and determining a comparison model with the highest coincidence degree by comparing models in a reference representation model library, so as to accurately determine the position of spinach sowing; according to the technical scheme, the sowing accuracy and efficiency are improved, and powerful support is provided for intelligent agriculture.

Description

Technical Field

[0001] The invention relates to the technical field of spinach sowing positioning, and in particular to a spinach sowing positioning method and system based on image data analysis. Background Art

[0002] In the modern agricultural production system, accurate and efficient seeding is one of the key steps to ensure the healthy growth of crops and increase yields. Especially in the planting process of leafy crops such as spinach, the uniform distribution of seeds has a vital impact on subsequent growth management, resource utilization and final yield. However, traditional seeding methods, whether manual seeding or simple mechanized operations, face challenges such as insufficient positioning accuracy, low operating efficiency and difficulty in adapting to complex and changing field environments.

[0003] Traditional sowing methods mainly rely on farmers' planting experience and simple mechanical devices. This method is incapable of coping with different soil conditions, weather changes and crop growth needs. Due to the lack of effective positioning methods, the distribution of seeds is often not uniform, resulting in too dense seeds in some areas and sparse seeds in other areas. This uneven sowing density not only increases the difficulty of field management, such as the difficulty in ensuring the accuracy of weeding, irrigation, fertilization and other operations, but may also lead to waste of resources and reduced crop yields. Summary of the invention

[0004] The purpose of the present invention is to provide a positioning method and system capable of accurately positioning the planting position of spinach.

[0005] The present invention discloses a spinach sowing positioning method based on image data analysis, comprising:

[0006] Step S100, taking a picture of the sowing soil area as a whole to obtain a sowing soil image, determining a screenshot size when randomly screenshotting the sowing soil image based on the actual area of ​​the sowing soil area, and capturing a number of soil image blocks for analysis;

[0007] Step S200, analyzing each soil image block for analysis, determining the soil color feature and soil texture feature corresponding to each soil image block for analysis, and calculating the average soil color feature, average soil texture feature and average soil structure feature of a plurality of soil image blocks for analysis;

[0008] Step S300, constructing a soil comprehensive characteristic expression model, adjusting the parameters of the soil comprehensive characteristic expression model according to the average soil color characteristics and the average soil texture characteristics, obtaining a soil comprehensive characteristic expression model for comparison, and configuring the corresponding spinach sowing interval for the soil comprehensive characteristic expression model for comparison based on the spinach sowing experiment record;

[0009] Step S400, classifying the soil comprehensive characteristic expression models for comparison with the same spinach sowing intervals, and classifying the soil comprehensive characteristic expression models for comparison based on the similarity between the soil comprehensive characteristic expression models for comparison, to obtain a reference expression model set;

[0010] Step S500, analyzing the soil comprehensive characteristic performance model for comparison in the reference performance model set, determining the marker factor parameter of each characteristic factor, marking the reference performance model set with the marker factor parameter, and constructing a plurality of reference performance model sets into a reference performance model library;

[0011] Step S600, analyze the currently taken sowing soil image, and construct the current soil comprehensive characteristic expression model based on the analysis result, and determine the characteristic factor parameters in the current soil comprehensive characteristic expression model, and find the corresponding reference expression model set in the reference expression model library based on the parameter difference value of the characteristic factor parameters and the corresponding marker factor parameters, and compare each comparison soil comprehensive characteristic expression model in the reference expression model set with the current soil comprehensive characteristic expression model, and determine the degree of consistency between the two based on the comparison result, and determine the suitable comparison soil comprehensive characteristic expression model based on the degree of consistency, and determine the position of spinach sowing based on the spinach sowing interval corresponding to the comparison soil comprehensive characteristic expression model.

[0012] In some embodiments disclosed in the present invention, the method for determining the soil color feature and the soil texture feature of each soil image block for analysis includes:

[0013] Methods for determining soil color characteristics include:

[0014] Step S201, performing color analysis on the soil image block for analysis, determining a color histogram of the soil image block for analysis, and associating the color histogram with the soil image block for analysis;

[0015] Methods for determining soil texture characteristics include:

[0016] Step S202, based on the data record of soil particle size and soil texture, set corresponding several soil particle size parameters and several soil texture parameters for each historical grayscale change vector comparison template, record the historical grayscale change vector comparison template with the set soil particle size parameters and soil texture parameters as grayscale change vector comparison data, and construct the several grayscale change vector comparison data into a grayscale change vector comparison database;

[0017] Step S203, matching the real-time grayscale change vector comparison template using the grayscale change vector comparison database, and determining the soil particle size parameters and soil texture parameters corresponding to the real-time grayscale change vector comparison template based on the matching result;

[0018] The method of constructing a grayscale change vector comparison template includes:

[0019] Grayscale processing is performed on the soil block for analysis, and a number of marker pixels are randomly selected, the marker grayscale values ​​of the marker pixels are analyzed, and the side grayscale values ​​of the side pixels next to the marker pixels are analyzed, and based on the positions of the marker pixels and the side pixels, the marker grayscale values ​​and the side grayscale values ​​are constructed into a grayscale value matrix;

[0020] A grayscale change vector judgment template is set for the grayscale value matrix, the grayscale change vector judgment template includes a number of template units corresponding to the positions of matrix factors in the grayscale value matrix, and vector pointing lines are set between the template units;

[0021] Based on the grayscale change vector, the vector pointing line in the judgment template is calculated, the vector change value between the corresponding matrix factors in the grayscale value matrix is ​​calculated, and the vector change value is associated with the corresponding vector pointing line to obtain the grayscale change vector comparison template.

[0022] In some embodiments disclosed in the present invention, the method for constructing a soil comprehensive characteristic expression model includes:

[0023] Step S301, constructing a soil particle simulation sphere for soil particles, and adjusting the diameter of the soil particle simulation sphere based on a number of soil particle size parameters;

[0024] Step S302, a number of soil particle simulation spheres are associated and combined to form a soil particle simulation sphere array, and based on a number of soil texture parameters, the distance between the soil particle simulation spheres in the soil particle simulation sphere array is adjusted to form an adjusted soil particle simulation sphere array, which is recorded as a soil comprehensive characteristic expression model.

[0025] In some embodiments disclosed in the present invention, the method for adjusting the diameter of the soil particle simulation sphere based on several soil particle size parameters includes:

[0026] Step S3011, constructing a spinach sowing interval variation sequence based on the spinach sowing experiment record, wherein the spinach sowing interval variation sequence includes a plurality of spinach sowing intervals, and the spinach sowing intervals are arranged according to a monotonic variation trend;

[0027] Step S3012, constructing a soil particle size change sequence for each soil particle size parameter, wherein the soil particle size change sequence includes a plurality of soil particle size parameters, and the soil particle size parameters are arranged according to a monotonic change trend;

[0028] Step S3013, based on the spinach sowing experiment record, establishing a correspondence between the spinach sowing interval in the spinach sowing interval variation sequence and the soil particle size parameter in the soil particle size variation sequence, and performing segment association on the correspondence between the spinach sowing interval and the soil particle size parameter to form a correspondence between the spinach sowing interval segments and the soil particle size parameter segments;

[0029] The method of segmentally associating the corresponding relationship between the spinach sowing interval and the soil particle size parameter includes:

[0030] The spinach sowing interval change sequence is analyzed. If there are adjacent spinach sowing intervals whose change amounts are greater than or equal to a preset value, the intervals between the spinach sowing intervals are identified as interval boundary points. Based on the interval boundary points, several spinach sowing interval segments in the spinach sowing interval change sequence are determined. Based on the corresponding relationship between the spinach sowing interval and the soil particle size parameter in the spinach sowing interval segment, the corresponding relationship between the spinach sowing interval segment and the soil particle size parameter segment is determined. The spinach sowing interval segments mapped by several soil particle size parameter segments under different performance states are analyzed to establish a soil particle size parameter segment group---spinach sowing interval segment corresponding table.

[0031] Step S3014, setting a standard diameter for the soil particle simulation sphere, dividing the standard diameter evenly to obtain a plurality of diameter segments, and associating each spinach sowing interval segment with a corresponding diameter segment according to the order of the spinach sowing interval segments;

[0032] Step S3015, obtaining several current soil particle size parameters, and determining the soil particle size parameter section to which each soil particle size parameter belongs, and substituting the determined soil particle size parameter section into the soil particle size parameter section group---spinach sowing interval section corresponding table, determining the corresponding spinach sowing interval section, and determining the first adjustment reference diameter of the soil simulation sphere based on the correspondence between the spinach sowing interval section and the diameter section, the first adjustment reference diameter is the overall length of the diameter section before the determined diameter section;

[0033] Step S3016, analyze the influencing characteristics of each soil particle size parameter, and based on the analysis results, determine the correction method for the first adjustment reference diameter to obtain the second adjustment reference diameter, and use the second adjustment reference diameter to adjust the diameter of the soil particle simulation sphere.

[0034] In some embodiments disclosed in the present invention, the correction method for the first adjustment reference diameter includes:

[0035] Step S30161, determining a mapping sub-segment of each soil particle size parameter segment in the soil particle size parameter segment group that is relatively equivalent to the spinach sowing interval segment, and determining the sub-segment position of the obtained soil particle size parameter in the mapping sub-segment, and recording the front segment of the sub-segment position as the front segment of the mapping sub-segment;

[0036] Step S30162, calculating the first segment occupancy ratio of the mapping subsegment relative to the soil particle size parameter segment, and calculating the second segment occupancy ratio of the front segment of the mapping subsegment relative to the mapping subsegment, determining the diameter segment interception segment of the diameter segment based on the first segment occupancy ratio and the second segment occupancy ratio, and adding the diameter segment interception segment and the first adjustment reference diameter to obtain the second adjustment reference diameter;

[0037] Among them, the expression for calculating the second adjusted reference diameter is:

[0038]

[0039] Where L2 is the second adjustment reference diameter, L1 is the first adjustment reference diameter, J is the diameter segment intercept length conversion adjustment coefficient, d is the length of the diameter segment, m 1-i is the first segment occupancy ratio corresponding to the i-th type of soil particle size parameter, m 2-i is the second section occupancy ratio corresponding to the i-th type of soil particle size parameter, c is the adjustment constant affecting the section occupancy ratio, and n is the total number of soil particle size parameter types.

[0040] In some embodiments disclosed in the present invention, based on a number of soil texture parameters, a method for adjusting the distance between soil particle simulation spheres in a soil particle simulation sphere array includes:

[0041] Step S3021, analyzing the spinach sowing experiment record, and constructing a spinach sowing interval change sequence based on the analysis result, wherein the spinach sowing interval change sequence includes a plurality of spinach sowing intervals, and the spinach sowing intervals are arranged according to a monotonic change trend;

[0042] Step S3022, analyzing the spinach sowing interval variation sequence, if there are adjacent spinach sowing intervals whose variation is greater than or equal to a preset value, the spinach sowing intervals are identified as interval boundary points, and based on the interval boundary points, a number of spinach sowing interval sections in the spinach sowing interval variation sequence are determined;

[0043] Step S3023, based on the analysis of the spinach sowing experiment record, determining a plurality of soil texture parameter groups corresponding to each spinach sowing interval section, the soil texture parameter group including a plurality of soil texture parameters;

[0044] Step S3024, evenly dividing the distance between the soil particle simulation spheres to obtain a plurality of inter-sphere distance segments, and establishing a corresponding relationship between the inter-sphere distance segments and the spinach sowing intervals based on the arrangement order of the spinach sowing intervals in the spinach sowing interval variation sequence;

[0045] Step S3025, obtaining several current soil texture parameters and substituting them into different soil texture parameter groups. If the texture parameter difference between each soil texture parameter of the same type is less than or equal to a preset value, the soil texture parameter group is identified as a reference soil texture parameter group.

[0046] Step S3026, determine the spinach sowing interval segment corresponding to the reference soil texture parameter group, and based on the correspondence between the inter-ball distance segment and the spinach sowing interval, determine the inter-ball distance segment corresponding to the reference soil texture parameter group, and the total length of several inter-ball distance segments before the inter-ball distance segment is recognized as the distance between the soil particle simulation spheres.

[0047] In some embodiments disclosed in the present invention, the method for determining the similarity between the soil comprehensive performance models for comparison includes:

[0048] Step S401, determining characteristic factors of a soil comprehensive characteristic representation model for comparison, the characteristic factors including the spherical size of a soil particle simulation sphere and the spatial size of a model space jointly constructed between the soil particle simulation spheres;

[0049] Step S402, setting a number of sphere size comparison intervals for sphere size, and setting a number of space size comparison intervals for space size, and randomly combining the sphere size comparison intervals and the space size comparison intervals to obtain a number of size comparison interval groups;

[0050] Step S403, classifying the soil comprehensive performance models for comparison based on the size comparison interval group to which the sphere size and space size of the soil comprehensive characteristic performance model for comparison belong, to obtain a reference performance model set.

[0051] In some embodiments disclosed in the present invention, a method for analyzing a soil comprehensive characteristic performance model for comparison in a reference performance model set to determine a marker factor parameter of each characteristic factor includes:

[0052] Step S501, determining characteristic factors to be analyzed, including the sphere size of the soil particle simulation sphere and the space size of the model space jointly constructed between the soil particle simulation spheres;

[0053] Step S502, analyzing each reference expression model in the reference expression model set, defining a number of marker sphere sizes, including establishing a sphere size reference line, and mapping each sphere size on the sphere size reference line in the form of a sphere size mapping point, analyzing the aggregation characteristics of the sphere size mapping points, and if the number of sphere size mapping points is greater than or equal to a preset value within a preset size interval, then identifying the midpoint of the size interval as the marker sphere size mapping point, and identifying the sphere size corresponding to the marker sphere size mapping point as the marker sphere size;

[0054] Step S503, determining the space size corresponding to each sphere mapping point in the preset size interval, and calculating the average value of these space sizes, which is recorded as the marked space size.

[0055] In some embodiments disclosed in the present invention, the method for determining the degree of consistency between the comparison soil comprehensive characteristic expression model and the current soil comprehensive characteristic expression model includes:

[0056] Step S601, determining the comparison sphere size and the comparison space size of the soil comprehensive characteristic expression model, and determining the current sphere size and the current space size of the current soil comprehensive characteristic expression model;

[0057] Step S602, compare the current sphere size with the comparison sphere size, and construct a first comparison operator based on the comparison result, compare the current space size with the comparison space size, and construct a second comparison operator based on the comparison result, randomly select several other comparison comprehensive feature expression models, and analyze the difference scale characteristics between them and the current comprehensive feature expression model respectively;

[0058] Step S603, based on the difference scale feature, the first comparison operator and the second comparison operator are modified to obtain the degree of agreement between the comparison soil comprehensive characteristic expression model and the current soil comprehensive characteristic expression model;

[0059] The expression for calculating the degree of fit is:

[0060]

[0061] Among them, W is the degree of fit, W MAXis the preset maximum degree of fit, K1 is the weight coefficient of the sphere size parameter difference, Δh is the sphere size parameter difference, K2 is the weight coefficient of the space size parameter difference, ΔV is the space size parameter difference, γ(x) is the difference scale judgment function between the xth randomly selected soil comprehensive characteristic expression model for comparison and the current soil comprehensive characteristic expression model, if the sphere size parameter difference and space size parameter difference of the two are less than or equal to the preset value, then γ(x) outputs 1, otherwise it outputs 0, P is the difference characteristic impact adjustment coefficient, and b is the difference characteristic impact adjustment constant.

[0062] In some embodiments disclosed in the present invention, a spinach sowing positioning system based on image data analysis includes:

[0063] The first module is used to take an overall photo of the sowing soil area to obtain a sowing soil image, determine the size of a random screenshot of the sowing soil image based on the actual area of ​​the sowing soil area, and capture a number of soil image blocks for analysis;

[0064] The second module is used to analyze each soil image block for analysis, determine the soil color characteristics and soil texture characteristics corresponding to each soil image block for analysis, and calculate the average soil color characteristics, average soil texture characteristics and average soil structure characteristics of several soil image blocks for analysis;

[0065] The third module is used to construct a soil comprehensive characteristic performance model. The soil comprehensive characteristic performance model is adjusted according to the average soil color characteristics and the average soil texture characteristics to obtain a soil comprehensive characteristic performance model for comparison. Based on the spinach sowing experiment record, the corresponding spinach sowing interval is configured for the soil comprehensive characteristic performance model for comparison;

[0066] The fourth module is used to classify the soil comprehensive characteristic performance models for comparison with the same spinach sowing interval, and to classify the soil comprehensive characteristic performance models for comparison based on the similarity between the soil comprehensive characteristic performance models for comparison, so as to obtain a reference performance model set;

[0067] The fifth module is used to analyze the soil comprehensive characteristic performance model for comparison in the reference performance model set, determine the marker factor parameter of each characteristic factor, mark the reference performance model set with the marker factor parameter, and construct several reference performance model sets into a reference performance model library;

[0068] The sixth module is used to analyze the currently taken sowing soil image, and to construct the current soil comprehensive characteristic expression model based on the analysis results, and to determine the characteristic factor parameters in the current soil comprehensive characteristic expression model, and to find the corresponding reference expression model set in the reference expression model library based on the parameter difference value of the characteristic factor parameters and the corresponding marker factor parameters, and to compare each comparison soil comprehensive characteristic expression model in the reference expression model set with the current soil comprehensive characteristic expression model, and to determine the degree of consistency between the two based on the comparison results, and to determine the suitable comparison soil comprehensive characteristic expression model based on the degree of consistency, and to determine the position of spinach sowing based on the spinach sowing interval corresponding to the comparison soil comprehensive characteristic expression model.

[0069] The invention discloses a spinach sowing positioning method and system based on image data analysis, and relates to the technical field of spinach sowing positioning. The method comprises the following steps: obtaining a soil image, determining a screenshot size according to an actual area, and intercepting a plurality of image blocks for analysis; analyzing the soil color and texture characteristics of each block, and calculating an average value to construct a soil comprehensive characteristic expression model; adjusting model parameters to obtain a soil comprehensive characteristic expression model for comparison and a corresponding sowing interval thereof; classifying the models, establishing a reference expression model set, analyzing and determining the marker factor parameters of characteristic factors, and constructing a reference expression model library; analyzing a soil image currently taken, constructing a current soil comprehensive characteristic expression model, and determining a comparison model with the highest degree of coincidence by comparing the models in the reference expression model library, thereby accurately determining the position of spinach sowing; the technical scheme of the invention improves the accuracy and efficiency of sowing, and provides strong support for intelligent agriculture.

[0070] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a method step diagram of a spinach sowing positioning method based on image data analysis disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0073] The following will be combined with the accompanying drawings and specific embodiments to clearly and completely describe the technical solution of the present invention. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and cannot be understood as limiting the scope of protection of the present invention. Those skilled in the art in this field can make some non-essential improvements and adjustments based on the content of the present invention described below. In the present invention, unless otherwise clearly specified and limited, the technical terms used in the present invention should be the common meanings understood by the technical personnel described in the present invention.

[0074] The present invention discloses a spinach sowing positioning method based on image data analysis, see Figure 1 ,include:

[0075] Step S100, taking a picture of the sowing soil area as a whole to obtain a sowing soil image, determining a screenshot size when taking a random screenshot of the sowing soil image based on the actual area of ​​the sowing soil area, and capturing a number of soil image blocks for analysis.

[0076] This step takes a picture of the sowing soil area as a whole to obtain an image containing information such as soil color and texture. Subsequently, the screenshot size is determined based on the actual area of ​​the soil area to ensure that the image block captured for analysis is representative but not too large to cause inefficiency in processing. By taking random screenshots, different parts of the soil area can be captured, providing diverse data for subsequent analysis, which helps to more accurately reflect the overall characteristics of the soil.

[0077] Step S200 , analyzing each soil image block for analysis, determining the soil color feature and soil texture feature corresponding to each soil image block for analysis, and calculating the average soil color feature and average soil texture feature of a plurality of soil image blocks for analysis.

[0078] In this step, image processing technology is used to extract the color and texture features of each soil image block for analysis. Color features reflect the color information of the soil, such as depth, lightness, etc., while texture features describe the physical properties of the soil surface, such as roughness and particle distribution. Calculating the average value of these features helps to eliminate local differences and obtain more stable and reliable comprehensive soil features. These features will provide key data support for the subsequent construction of a comprehensive soil feature representation model.

[0079] In some embodiments disclosed in the present invention, the method for determining the soil color feature and the soil texture feature of each soil image block for analysis includes:

[0080] Methods for determining soil color characteristics include:

[0081] Step S201 , color analysis is performed on the soil image block for analysis, a color histogram of the soil image block for analysis is determined, and the color histogram is associated with the soil image block for analysis and recorded.

[0082] Color feature is one of the important attributes to describe the appearance of soil. It reflects the composition, moisture, oxidation degree and other information of soil. In this step, the color histogram can be obtained by color analysis of the analytical soil image block. The color histogram is a method to count the frequency of occurrence of various colors in an image, which can intuitively show the distribution of soil color. The color histogram is associated with the corresponding analytical soil image block to provide basic data for subsequent feature extraction and model construction.

[0083] Methods for determining soil texture characteristics include:

[0084] Step S202, based on the data records of soil particle size and soil texture, set corresponding several soil particle size parameters and several soil texture parameters for each historical grayscale change vector comparison template, record the historical grayscale change vector comparison template with soil particle size parameters and soil texture parameters as grayscale change vector comparison data, and construct several grayscale change vector comparison data into a grayscale change vector comparison database.

[0085] Soil texture features describe the physical properties of the soil surface, such as roughness, particle distribution, and pore structure, and are of great significance for understanding the physical properties of soil and predicting soil behavior. In this step, by constructing a grayscale change vector comparison database and using the database to match the real-time grayscale change vector comparison template, the soil particle size parameters and soil texture parameters can be determined, thereby extracting the soil texture features.

[0086] Step S203, matching the real-time grayscale change vector comparison template with the grayscale change vector comparison database, and determining the soil particle size parameters and soil texture parameters corresponding to the real-time grayscale change vector comparison template based on the matching result.

[0087] The method of constructing a grayscale change vector comparison template includes:

[0088] The soil block for analysis is grayed, and several marker pixels are randomly selected. The marker gray values ​​of the marker pixels are analyzed, and the side gray values ​​of the side pixels next to the marker pixels are analyzed. Based on the positions of the marker pixels and the side pixels, the marker gray values ​​and the side gray values ​​are constructed into a gray value matrix.

[0089] A grayscale change vector judgment template is set for the grayscale value matrix. The grayscale change vector judgment template includes a number of template units corresponding to the positions of matrix factors in the grayscale value matrix, and vector pointing lines are set between the template units.

[0090] Based on the grayscale change vector, the vector pointing line in the judgment template is calculated, the vector change value between the corresponding matrix factors in the grayscale value matrix is ​​calculated, and the vector change value is associated with the corresponding vector pointing line to obtain the grayscale change vector comparison template.

[0091] Step S300, constructing a soil comprehensive characteristic expression model, adjusting the parameters of the soil comprehensive characteristic expression model according to the average soil color characteristics and the average soil texture characteristics, obtaining a soil comprehensive characteristic expression model for comparison, and configuring the corresponding spinach sowing interval for the soil comprehensive characteristic expression model for comparison based on the spinach sowing experiment record.

[0092] Based on the characteristic data extracted in step S200, a soil comprehensive characteristic expression model is constructed. The model aims to describe the comprehensive characteristics of the soil by mathematical methods. By adjusting the parameters of the model, such as color feature weights, texture feature weights, etc., the model can more accurately reflect the actual state of the soil. At the same time, combined with the spinach sowing experiment record, the corresponding spinach sowing interval is configured for the model, so that the model can guide the selection of spinach sowing positions in practical applications.

[0093] In some embodiments disclosed in the present invention, the method for constructing a soil comprehensive characteristic expression model includes:

[0094] Step S301: construct a soil particle simulation sphere for soil particles, and adjust the diameter of the soil particle simulation sphere based on a number of soil particle size parameters.

[0095] Step S301 is the first step in building a soil comprehensive characteristic representation model, the core of which is to simulate the morphology of soil particles. In real soil, particles have different morphologies, but in order to simplify the model and capture the main features, soil particles are usually abstracted into simple geometric shapes, such as spheres.

[0096] Constructing a soil particle simulation sphere: First, a basic sphere model is selected as the simulation object of soil particles. The sphere model has the advantages of being easy to handle and able to reflect the basic shape of particles.

[0097] Adjust the diameter of the simulated sphere: Next, the diameter of the simulated sphere is finely adjusted based on the soil particle size parameters (such as average diameter, standard deviation, etc.) previously extracted from the actual soil image. The purpose of this step is to ensure that the simulated sphere is consistent in size with the real soil particles, thereby more accurately reflecting the physical properties of the soil.

[0098] In some embodiments disclosed in the present invention, the method for adjusting the diameter of the soil particle simulation sphere based on several soil particle size parameters includes:

[0099] Step S3011, constructing a spinach sowing interval variation sequence based on the spinach sowing experiment record, wherein the spinach sowing interval variation sequence includes a plurality of spinach sowing intervals, and the spinach sowing intervals are arranged according to a monotonic variation trend.

[0100] Step S3012: construct a soil particle size variation sequence for each soil particle size parameter. The soil particle size variation sequence includes a plurality of soil particle size parameters, and the soil particle size parameters are arranged according to a monotonic variation trend.

[0101] Step S3013, based on the spinach sowing experiment record, establish the correspondence between the spinach sowing interval in the spinach sowing interval change sequence and the soil particle size parameters in the soil particle size change sequence, and perform segment association on the correspondence between the spinach sowing interval and the soil particle size parameters to form a correspondence between the spinach sowing interval segments and the soil particle size parameter segments.

[0102] The method of segmentally associating the corresponding relationship between the spinach sowing interval and the soil particle size parameter includes:

[0103] The spinach sowing interval change sequence is analyzed. If there are adjacent spinach sowing intervals whose change amount is greater than or equal to a preset value, the spinach sowing intervals are identified as interval boundary points. Based on the interval boundary points, several spinach sowing interval segments in the spinach sowing interval change sequence are determined. Based on the correspondence between the spinach sowing intervals and the soil particle size parameters in the spinach sowing interval segments, the correspondence between the spinach sowing interval segments and the soil particle size parameter segments is determined. The spinach sowing interval segments mapped by several soil particle size parameter segments under different performance states are analyzed, and a soil particle size parameter segment group-spinach sowing interval segment correspondence table is established.

[0104] Step S3014, setting a standard diameter for the soil particle simulation sphere, dividing the standard diameter evenly to obtain a number of diameter segments, and associating each spinach sowing interval segment with a corresponding diameter segment according to the order of the spinach sowing interval segments.

[0105] Step S3015, obtain several current soil particle size parameters, and determine the soil particle size parameter segment to which each soil particle size parameter belongs, and substitute the determined soil particle size parameter segment into the soil particle size parameter segment group---spinach sowing interval segment correspondence table, determine the corresponding spinach sowing interval segment, and based on the correspondence between the spinach sowing interval segment and the diameter segment, determine the first adjustment reference diameter of the soil simulation sphere, the first adjustment reference diameter is the overall length of the diameter segment before the determined diameter segment.

[0106] Step S3016, analyze the influencing characteristics of each soil particle size parameter, and based on the analysis results, determine the correction method for the first adjustment reference diameter to obtain the second adjustment reference diameter, and use the second adjustment reference diameter to adjust the diameter of the soil particle simulation sphere.

[0107] In some embodiments disclosed in the present invention, the correction method for the first adjustment reference diameter includes:

[0108] Step S30161, determine the mapping sub-segment of each soil particle size parameter segment in the soil particle size parameter segment group that is relatively equivalent to the spinach sowing interval segment, and judge the sub-segment position of the acquired soil particle size parameter in the mapping sub-segment, and record the front segment of the sub-segment position as the front segment of the mapping sub-segment.

[0109] Step S30162, calculate the first segment occupancy ratio of the mapping sub-segment relative to the soil particle size parameter segment, and calculate the second segment occupancy ratio of the front segment of the mapping sub-segment relative to the mapping sub-segment, based on the first segment occupancy ratio and the second segment occupancy ratio, determine the diameter segment cut-off segment of the diameter segment, and add the diameter segment cut-off segment and the first adjusted reference diameter to obtain the second adjusted reference diameter.

[0110] Among them, the expression for calculating the second adjusted reference diameter is:

[0111]

[0112] Where L2 is the second adjustment reference diameter, L1 is the first adjustment reference diameter, J is the diameter segment intercept length conversion adjustment coefficient, d is the length of the diameter segment, m 1-i is the first segment occupancy ratio corresponding to the i-th type of soil particle size parameter, m 2-i is the second section occupancy ratio corresponding to the i-th type of soil particle size parameter, c is the adjustment constant affecting the section occupancy ratio, and n is the total number of soil particle size parameter types.

[0113] Step S302, a number of soil particle simulation spheres are associated and combined to form a soil particle simulation sphere array, and based on a number of soil texture parameters, the distance between the soil particle simulation spheres in the soil particle simulation sphere array is adjusted to form an adjusted soil particle simulation sphere array, which is recorded as a soil comprehensive characteristic expression model.

[0114] Step S302 is to further simulate the arrangement and interaction between soil particles based on step S301 to form a soil model with a specific texture.

[0115] Combining simulated spheres to form an array: combining the plurality of simulated spheres of soil particles obtained in step S301 according to a certain rule or in a random distribution manner to form a simulated sphere array. This array can be regarded as a simplified and representative soil sample.

[0116] Adjust the distance between simulated spheres: Next, the distance between soil particles in the simulated sphere array is finely adjusted based on the soil texture parameters extracted previously (such as porosity, contact between particles, etc.). These adjustments are designed to simulate the spatial relationship and interaction between particles in real soil, thereby reflecting the texture and structural characteristics of the soil.

[0117] Forming a soil comprehensive characteristic performance model: After the above adjustments, the obtained soil particle simulation sphere array is the soil comprehensive characteristic performance model. This model not only takes into account the size of soil particles, but also the arrangement and interaction between particles, so it can more comprehensively reflect the comprehensive characteristics of the soil.

[0118] In some embodiments disclosed in the present invention, based on a number of soil texture parameters, a method for adjusting the distance between soil particle simulation spheres in a soil particle simulation sphere array includes:

[0119] Step S3021, analyzing the spinach sowing experiment record, and constructing a spinach sowing interval variation sequence based on the analysis result, wherein the spinach sowing interval variation sequence includes a plurality of spinach sowing intervals, and the spinach sowing intervals are arranged according to a monotonic variation trend.

[0120] Step S3022, analyzing the spinach sowing interval variation sequence, if there are adjacent spinach sowing intervals whose variation is greater than or equal to a preset value, the spinach sowing intervals are identified as interval boundary points, and based on the interval boundary points, several spinach sowing interval segments in the spinach sowing interval variation sequence are determined.

[0121] Step S3023, based on the analysis of the spinach sowing experiment records, several soil texture parameter groups corresponding to each spinach sowing interval section are determined, and the soil texture parameter groups include several soil texture parameters.

[0122] Step S3024, evenly divide the distance between the soil particle simulation spheres to obtain a number of inter-sphere distance segments, and establish a corresponding relationship between the inter-sphere distance segments and the spinach sowing interval based on the arrangement order of the spinach sowing interval in the spinach sowing interval change sequence.

[0123] Step S3025, obtaining several current soil texture parameters and substituting them into different soil texture parameter groups. If the texture parameter difference between each soil texture parameter of the same type is less than or equal to a preset value, the soil texture parameter group is identified as a reference soil texture parameter group.

[0124] Step S3026, determine the spinach sowing interval segment corresponding to the reference soil texture parameter group, and based on the correspondence between the inter-ball distance segment and the spinach sowing interval, determine the inter-ball distance segment corresponding to the reference soil texture parameter group, and the total length of several inter-ball distance segments before the inter-ball distance segment is recognized as the distance between the soil particle simulation spheres.

[0125] Step S400, classifying the soil comprehensive characteristic expression models for comparison with the same spinach sowing intervals once, and classifying the soil comprehensive characteristic expression models for comparison twice based on the similarity between the soil comprehensive characteristic expression models for comparison, to obtain a reference expression model set.

[0126] This step simplifies the management and retrieval process of the model by comparing and classifying the soil comprehensive characteristic performance models. The initial classification is based on the spinach sowing interval, and the models with similar sowing intervals are classified into one category. The secondary classification is based on the similarity between the models, such as feature similarity, sowing interval proximity, etc., to further refine the classification. Through classification, a reference performance model set can be formed. The models in each set have similar characteristic performances, which facilitates the rapid matching of the most suitable model in subsequent practical applications.

[0127] In some embodiments disclosed in the present invention, the method for determining the similarity between the soil comprehensive performance models for comparison includes:

[0128] Step S401, determining characteristic factors of the soil comprehensive characteristic representation model for comparison, the characteristic factors including the spherical size of the soil particle simulation sphere and the spatial size of the model space jointly constructed between the soil particle simulation spheres.

[0129] Step S402, setting a number of sphere size comparison intervals for sphere size, setting a number of space size comparison intervals for space size, and randomly combining the sphere size comparison intervals and the space size comparison intervals to obtain a number of size comparison interval groups.

[0130] Step S403, classifying the soil comprehensive performance models for comparison based on the size comparison interval group to which the sphere size and space size of the soil comprehensive characteristic performance model for comparison belong, to obtain a reference performance model set.

[0131] Step S500, analyze the soil comprehensive characteristic performance model for comparison in the reference performance model set, determine the marker factor parameters of each characteristic factor, mark the reference performance model set with the marker factor parameters, and construct several reference performance model sets into a reference performance model library.

[0132] This step deeply analyzes each model in the reference performance model set to determine the marker factor parameters in its characteristic factors. These marker factor parameters are highly representative and discriminative, and can accurately reflect the characteristic performance of the model. Subsequently, these marker factor parameters are used to mark the model to form a unique identity. Integrating the marked models into the reference performance model library can greatly improve the retrieval speed and matching accuracy of the model.

[0133] In some embodiments disclosed in the present invention, a method for analyzing a soil comprehensive characteristic performance model for comparison in a reference performance model set to determine a marker factor parameter of each characteristic factor includes:

[0134] Step S501, determining characteristic factors to be analyzed, including the sphere size of the soil particle simulation sphere and the space size of the model space jointly constructed between the soil particle simulation spheres.

[0135] Step S502, analyze each reference expression model in the reference expression model set, define a number of marker sphere sizes, including establishing a sphere size reference line, and mapping each sphere size on the sphere size reference line in the form of a sphere size mapping point, analyzing the aggregation characteristics of the sphere size mapping points, if within a preset size range, the number of sphere size mapping points is greater than or equal to a preset value, then the midpoint of the size range is identified as the marker sphere size mapping point, and the sphere size corresponding to the marker sphere size mapping point is identified as the marker sphere size.

[0136] Step S503, determining the space size corresponding to each sphere mapping point in the preset size interval, and calculating the average value of these space sizes, which is recorded as the marked space size.

[0137] Step S600, analyze the currently taken sowing soil image, and construct the current soil comprehensive characteristic expression model based on the analysis result, and determine the characteristic factor parameters in the current soil comprehensive characteristic expression model, and find the corresponding reference expression model set in the reference expression model library based on the parameter difference value of the characteristic factor parameters and the corresponding marker factor parameters, and compare each comparison soil comprehensive characteristic expression model in the reference expression model set with the current soil comprehensive characteristic expression model, and determine the degree of consistency between the two based on the comparison result, and determine the suitable comparison soil comprehensive characteristic expression model based on the degree of consistency, and determine the position of spinach sowing based on the spinach sowing interval corresponding to the comparison soil comprehensive characteristic expression model.

[0138] In practical applications, an image of the current sowing soil area is taken and a current soil comprehensive characteristic performance model is constructed. The model is compared with the model in the reference performance model library, the difference between the characteristic factor parameters is calculated, and the most matching reference performance model set is found. Subsequently, the models are compared one by one in the set to determine the comparison soil comprehensive characteristic performance model with the highest degree of consistency with the current soil comprehensive characteristic performance model. Finally, according to the spinach sowing interval corresponding to the model, the sowing position of spinach is accurately guided. This process realizes intelligent and precise decision-making of sowing positions, which helps to improve sowing efficiency and crop growth quality.

[0139] In some embodiments disclosed in the present invention, the method for determining the degree of consistency between the comparison soil comprehensive characteristic expression model and the current soil comprehensive characteristic expression model includes:

[0140] Step S601, determining the comparison sphere size and the comparison space size of the soil comprehensive characteristic expression model, and determining the current sphere size and the current space size of the current soil comprehensive characteristic expression model.

[0141] Step S602, compare the current sphere size with the comparison sphere size, and construct a first comparison operator based on the comparison result, compare the current space size with the comparison space size, and construct a second comparison operator based on the comparison result, randomly select several other comprehensive feature expression models for comparison, and analyze the difference scale characteristics between them and the current comprehensive feature expression model.

[0142] Step S603: based on the difference scale feature, the first comparison operator and the second comparison operator are modified to obtain the degree of agreement between the comparison soil comprehensive characteristic expression model and the current soil comprehensive characteristic expression model.

[0143] The expression for calculating the degree of fit is:

[0144]

[0146] Among them, W is the degree of fit, W MAX is the preset maximum degree of fit, K1 is the weight coefficient of the sphere size parameter difference, Δh is the sphere size parameter difference, K2 is the weight coefficient of the space size parameter difference, ΔV is the space size parameter difference, γ(x) is the difference scale judgment function between the xth randomly selected soil comprehensive characteristic expression model for comparison and the current soil comprehensive characteristic expression model, if the sphere size parameter difference and space size parameter difference of the two are less than or equal to the preset value, then γ(x) outputs 1, otherwise it outputs 0, P is the difference characteristic impact adjustment coefficient, and b is the difference characteristic impact adjustment constant.

[0147] In some embodiments disclosed in the present invention, a spinach sowing positioning system based on image data analysis includes:

[0148] The first module is used to take an overall photo of the sowing soil area to obtain a sowing soil image, determine the size of a random screenshot of the sowing soil image based on the actual area of ​​the sowing soil area, and capture a number of soil image blocks for analysis;

[0149] The second module is used to analyze each soil image block for analysis, determine the soil color characteristics and soil texture characteristics corresponding to each soil image block for analysis, and calculate the average soil color characteristics, average soil texture characteristics and average soil structure characteristics of several soil image blocks for analysis;

[0150] The third module is used to construct a soil comprehensive characteristic performance model. The soil comprehensive characteristic performance model is adjusted according to the average soil color characteristics and the average soil texture characteristics to obtain a soil comprehensive characteristic performance model for comparison. Based on the spinach sowing experiment record, the corresponding spinach sowing interval is configured for the soil comprehensive characteristic performance model for comparison;

[0151] The fourth module is used to classify the soil comprehensive characteristic performance models for comparison with the same spinach sowing interval, and to classify the soil comprehensive characteristic performance models for comparison based on the similarity between the soil comprehensive characteristic performance models for comparison, so as to obtain a reference performance model set;

[0152] The fifth module is used to analyze the soil comprehensive characteristic performance model for comparison in the reference performance model set, determine the marker factor parameter of each characteristic factor, mark the reference performance model set with the marker factor parameter, and construct several reference performance model sets into a reference performance model library;

[0153] The sixth module is used to analyze the currently taken sowing soil image, and to construct the current soil comprehensive characteristic expression model based on the analysis results, and to determine the characteristic factor parameters in the current soil comprehensive characteristic expression model, and to find the corresponding reference expression model set in the reference expression model library based on the parameter difference value of the characteristic factor parameters and the corresponding marker factor parameters, and to compare each comparison soil comprehensive characteristic expression model in the reference expression model set with the current soil comprehensive characteristic expression model, and to determine the degree of consistency between the two based on the comparison results, and to determine the suitable comparison soil comprehensive characteristic expression model based on the degree of consistency, and to determine the position of spinach sowing based on the spinach sowing interval corresponding to the comparison soil comprehensive characteristic expression model.

[0154] The invention discloses a spinach sowing positioning method and system based on image data analysis, and relates to the technical field of spinach sowing positioning. The method comprises the following steps: obtaining a soil image, determining a screenshot size according to an actual area, and intercepting a plurality of image blocks for analysis; analyzing the soil color and texture characteristics of each block, and calculating an average value to construct a soil comprehensive characteristic expression model; adjusting model parameters to obtain a soil comprehensive characteristic expression model for comparison and a corresponding sowing interval thereof; classifying the models, establishing a reference expression model set, analyzing and determining the marker factor parameters of characteristic factors, and constructing a reference expression model library; analyzing a soil image currently taken, constructing a current soil comprehensive characteristic expression model, and determining a comparison model with the highest degree of coincidence by comparing the models in the reference expression model library, thereby accurately determining the position of spinach sowing; the technical scheme of the invention improves the accuracy and efficiency of sowing, and provides strong support for intelligent agriculture.

[0155] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A spinach sowing positioning method based on image data analysis, characterized in that: include: Step S100, taking a picture of the sowing soil area as a whole to obtain a sowing soil image, determining a screenshot size when randomly screenshotting the sowing soil image based on the actual area of ​​the sowing soil area, and capturing a number of soil image blocks for analysis; Step S200, analyzing each soil image block for analysis, determining the soil color feature and soil texture feature corresponding to each soil image block for analysis, and calculating the average soil color feature and average soil texture feature of a plurality of soil image blocks for analysis; Step S300, constructing a soil comprehensive characteristic expression model, adjusting the parameters of the soil comprehensive characteristic expression model according to the average soil color characteristics and the average soil texture characteristics, obtaining a soil comprehensive characteristic expression model for comparison, and configuring the corresponding spinach sowing interval for the soil comprehensive characteristic expression model for comparison based on the spinach sowing experiment record; Step S400, classifying the soil comprehensive characteristic expression models for comparison with the same spinach sowing intervals, and classifying the soil comprehensive characteristic expression models for comparison based on the similarity between the soil comprehensive characteristic expression models for comparison, to obtain a reference expression model set; Step S500, analyzing the soil comprehensive characteristic performance model for comparison in the reference performance model set, determining the marker factor parameter of each characteristic factor, marking the reference performance model set with the marker factor parameter, and constructing a plurality of reference performance model sets into a reference performance model library; Step S600, analyze the currently taken sowing soil image, and construct the current soil comprehensive characteristic expression model based on the analysis result, and determine the characteristic factor parameters in the current soil comprehensive characteristic expression model, and find the corresponding reference expression model set in the reference expression model library based on the parameter difference value of the characteristic factor parameters and the corresponding marker factor parameters, and compare each comparison soil comprehensive characteristic expression model in the reference expression model set with the current soil comprehensive characteristic expression model, and determine the degree of consistency between the two based on the comparison result, and determine the suitable comparison soil comprehensive characteristic expression model based on the degree of consistency, and determine the position of spinach sowing based on the spinach sowing interval corresponding to the comparison soil comprehensive characteristic expression model.

2. The spinach sowing positioning method based on image data analysis according to claim 1, characterized in that: The method for determining the soil color characteristics and soil texture characteristics of each soil image block for analysis includes: Methods for determining soil color characteristics include: Step S201, performing color analysis on the soil image block for analysis, determining a color histogram of the soil image block for analysis, and associating the color histogram with the soil image block for analysis; Methods for determining soil texture characteristics include: Step S202, based on the data record of soil particle size and soil texture, set corresponding several soil particle size parameters and several soil texture parameters for each historical grayscale change vector comparison template, record the historical grayscale change vector comparison template with the set soil particle size parameters and soil texture parameters as grayscale change vector comparison data, and construct the several grayscale change vector comparison data into a grayscale change vector comparison database; Step S203, matching the real-time grayscale change vector comparison template using the grayscale change vector comparison database, and determining the soil particle size parameters and soil texture parameters corresponding to the real-time grayscale change vector comparison template based on the matching result; The method of constructing a grayscale change vector comparison template includes: Grayscale processing is performed on the soil block for analysis, and a number of marker pixels are randomly selected, the marker grayscale values ​​of the marker pixels are analyzed, and the side grayscale values ​​of the side pixels next to the marker pixels are analyzed, and based on the positions of the marker pixels and the side pixels, the marker grayscale values ​​and the side grayscale values ​​are constructed into a grayscale value matrix; A grayscale change vector judgment template is set for the grayscale value matrix, the grayscale change vector judgment template includes a number of template units corresponding to the positions of matrix factors in the grayscale value matrix, and vector pointing lines are set between the template units; Based on the grayscale change vector, the vector pointing line in the judgment template is calculated, the vector change value between the corresponding matrix factors in the grayscale value matrix is ​​calculated, and the vector change value is associated with the corresponding vector pointing line to obtain the grayscale change vector comparison template.

3. The spinach sowing positioning method based on image data analysis according to claim 1, characterized in that: Methods for constructing a comprehensive soil characteristic performance model include: Step S301, constructing a soil particle simulation sphere for soil particles, and adjusting the diameter of the soil particle simulation sphere based on a number of soil particle size parameters; Step S302, a number of soil particle simulation spheres are associated and combined to form a soil particle simulation sphere array, and based on a number of soil texture parameters, the distance between the soil particle simulation spheres in the soil particle simulation sphere array is adjusted to form an adjusted soil particle simulation sphere array, which is recorded as a soil comprehensive characteristic expression model.

4. The spinach sowing positioning method based on image data analysis according to claim 3 is characterized in that: The method for adjusting the diameter of the soil particle simulation sphere based on several soil particle size parameters includes: Step S3011, constructing a spinach sowing interval variation sequence based on the spinach sowing experiment record, wherein the spinach sowing interval variation sequence includes a plurality of spinach sowing intervals, and the spinach sowing intervals are arranged according to a monotonic variation trend; Step S3012, constructing a soil particle size change sequence for each soil particle size parameter, wherein the soil particle size change sequence includes a plurality of soil particle size parameters, and the soil particle size parameters are arranged according to a monotonic change trend; Step S3013, based on the spinach sowing experiment record, establishing a correspondence between the spinach sowing interval in the spinach sowing interval variation sequence and the soil particle size parameter in the soil particle size variation sequence, and performing segment association on the correspondence between the spinach sowing interval and the soil particle size parameter to form a correspondence between the spinach sowing interval segments and the soil particle size parameter segments; The method of segmentally associating the corresponding relationship between the spinach sowing interval and the soil particle size parameter includes: The spinach sowing interval change sequence is analyzed. If there are adjacent spinach sowing intervals whose change amounts are greater than or equal to a preset value, the intervals between the spinach sowing intervals are identified as interval boundary points. Based on the interval boundary points, several spinach sowing interval segments in the spinach sowing interval change sequence are determined. Based on the corresponding relationship between the spinach sowing interval and the soil particle size parameter in the spinach sowing interval segment, the corresponding relationship between the spinach sowing interval segment and the soil particle size parameter segment is determined. The spinach sowing interval segments mapped by several soil particle size parameter segments under different performance states are analyzed to establish a soil particle size parameter segment group---spinach sowing interval segment corresponding table. Step S3014, setting a standard diameter for the soil particle simulation sphere, dividing the standard diameter evenly to obtain a plurality of diameter segments, and associating each spinach sowing interval segment with a corresponding diameter segment according to the order of the spinach sowing interval segments; Step S3015, obtaining several current soil particle size parameters, and determining the soil particle size parameter section to which each soil particle size parameter belongs, and substituting the determined soil particle size parameter section into the soil particle size parameter section group---spinach sowing interval section corresponding table, determining the corresponding spinach sowing interval section, and determining the first adjustment reference diameter of the soil simulation sphere based on the correspondence between the spinach sowing interval section and the diameter section, the first adjustment reference diameter is the overall length of the diameter section before the determined diameter section; Step S3016, analyze the influencing characteristics of each soil particle size parameter, and based on the analysis results, determine the correction method for the first adjustment reference diameter to obtain the second adjustment reference diameter, and use the second adjustment reference diameter to adjust the diameter of the soil particle simulation sphere.

5. The spinach sowing positioning method based on image data analysis according to claim 4, characterized in that: The correction methods for the first adjustment reference diameter include: Step S30161, determining a mapping sub-segment of each soil particle size parameter segment in the soil particle size parameter segment group that is relatively equivalent to the spinach sowing interval segment, and determining the sub-segment position of the obtained soil particle size parameter in the mapping sub-segment, and recording the front segment of the sub-segment position as the front segment of the mapping sub-segment; Step S30162, calculating the first segment occupancy ratio of the mapping subsegment relative to the soil particle size parameter segment, and calculating the second segment occupancy ratio of the front segment of the mapping subsegment relative to the mapping subsegment, determining the diameter segment interception segment of the diameter segment based on the first segment occupancy ratio and the second segment occupancy ratio, and adding the diameter segment interception segment and the first adjustment reference diameter to obtain the second adjustment reference diameter; Among them, the expression for calculating the second adjusted reference diameter is: Where L2 is the second adjustment reference diameter, L1 is the first adjustment reference diameter, J is the diameter segment intercept length conversion adjustment coefficient, d is the length of the diameter segment, m 1-i is the first segment occupancy ratio corresponding to the i-th type of soil particle size parameter, m 2-i is the second section occupancy ratio corresponding to the i-th type of soil particle size parameter, c is the adjustment constant affecting the section occupancy ratio, and n is the total number of soil particle size parameter types.

6. The spinach sowing positioning method based on image data analysis according to claim 3, characterized in that: The method for adjusting the distance between soil particle simulation spheres in the soil particle simulation sphere array based on a plurality of soil texture parameters comprises: Step S3021, analyzing the spinach sowing experiment record, and constructing a spinach sowing interval change sequence based on the analysis result, wherein the spinach sowing interval change sequence includes a plurality of spinach sowing intervals, and the spinach sowing intervals are arranged according to a monotonic change trend; Step S3022, analyzing the spinach sowing interval variation sequence, if there are adjacent spinach sowing intervals whose variation is greater than or equal to a preset value, the spinach sowing intervals are identified as interval boundary points, and based on the interval boundary points, a number of spinach sowing interval sections in the spinach sowing interval variation sequence are determined; Step S3023, based on the analysis of the spinach sowing experiment record, determining a plurality of soil texture parameter groups corresponding to each spinach sowing interval section, the soil texture parameter group including a plurality of soil texture parameters; Step S3024, evenly dividing the distance between the soil particle simulation spheres to obtain a plurality of inter-sphere distance segments, and establishing a corresponding relationship between the inter-sphere distance segments and the spinach sowing intervals based on the arrangement order of the spinach sowing intervals in the spinach sowing interval variation sequence; Step S3025, obtaining several current soil texture parameters and substituting them into different soil texture parameter groups. If the texture parameter difference between each soil texture parameter of the same type is less than or equal to a preset value, the soil texture parameter group is identified as a reference soil texture parameter group. Step S3026, determine the spinach sowing interval segment corresponding to the reference soil texture parameter group, and based on the correspondence between the inter-ball distance segment and the spinach sowing interval, determine the inter-ball distance segment corresponding to the reference soil texture parameter group, and the total length of several inter-ball distance segments before the inter-ball distance segment is recognized as the distance between the soil particle simulation spheres.

7. The spinach sowing positioning method based on image data analysis according to claim 3 is characterized in that: Methods for determining the similarity between the soil comprehensive performance models used for comparison include: Step S401, determining characteristic factors of a soil comprehensive characteristic representation model for comparison, the characteristic factors including the spherical size of a soil particle simulation sphere and the spatial size of a model space jointly constructed between the soil particle simulation spheres; Step S402, setting a plurality of sphere size comparison intervals for sphere size, and setting a plurality of space size comparison intervals for space size, and randomly combining the sphere size comparison intervals and the space size comparison intervals to obtain a plurality of size comparison interval groups; Step S403, classifying the soil comprehensive performance models for comparison based on the size comparison interval group to which the sphere size and space size of the soil comprehensive characteristic performance model for comparison belong, to obtain a reference performance model set.

8. The spinach sowing positioning method based on image data analysis according to claim 3 is characterized in that: The method of analyzing the soil comprehensive characteristic performance model for comparison in the reference performance model set and determining the marker factor parameters of each characteristic factor includes: Step S501, determining characteristic factors to be analyzed, including the sphere size of the soil particle simulation sphere and the space size of the model space jointly constructed between the soil particle simulation spheres; Step S502, analyzing each reference expression model in the reference expression model set, defining a number of marker sphere sizes, including establishing a sphere size reference line, and mapping each sphere size on the sphere size reference line in the form of a sphere size mapping point, analyzing the aggregation characteristics of the sphere size mapping points, and if the number of sphere size mapping points is greater than or equal to a preset value within a preset size interval, then identifying the midpoint of the size interval as a marker sphere size mapping point, and identifying the sphere size corresponding to the marker sphere size mapping point as the marker sphere size; Step S503, determining the space size corresponding to each sphere mapping point in the preset size interval, and calculating the average value of these space sizes, which is recorded as the marked space size.

9. The spinach sowing positioning method based on image data analysis according to claim 3, characterized in that: Methods for determining the degree of agreement between the comparative soil comprehensive characteristic expression model and the current soil comprehensive characteristic expression model include: Step S601, determining the comparison sphere size and the comparison space size of the soil comprehensive characteristic expression model, and determining the current sphere size and the current space size of the current soil comprehensive characteristic expression model; Step S602, compare the current sphere size with the comparison sphere size, and construct a first comparison operator based on the comparison result, compare the current space size with the comparison space size, and construct a second comparison operator based on the comparison result, randomly select several other comparison comprehensive feature expression models, and analyze the difference scale characteristics between them and the current comprehensive feature expression model respectively; Step S603, based on the difference scale feature, the first comparison operator and the second comparison operator are modified to obtain the degree of agreement between the comparison soil comprehensive characteristic expression model and the current soil comprehensive characteristic expression model; The expression for calculating the degree of fit is: Among them, W is the degree of fit, W MAX is the preset maximum degree of fit, K1 is the weight coefficient of the sphere size parameter difference, Δh is the sphere size parameter difference, K2 is the weight coefficient of the space size parameter difference, ΔV is the space size parameter difference, γ(x) is the difference scale judgment function between the x-th randomly selected soil comprehensive characteristic expression model for comparison and the current soil comprehensive characteristic expression model, if the sphere size parameter difference and space size parameter difference of the two are less than or equal to the preset value, then γ(x) outputs 1, otherwise it outputs 0, P is the difference characteristic impact adjustment coefficient, and b is the difference characteristic impact adjustment constant.

10. A spinach sowing positioning system based on image data analysis, characterized in that: The spinach sowing positioning method for executing any one of claims 1 to 9 comprises: The first module is used to take an overall photo of the sowing soil area to obtain a sowing soil image, determine the size of a random screenshot of the sowing soil image based on the actual area of ​​the sowing soil area, and capture a number of soil image blocks for analysis; The second module is used to analyze each soil image block for analysis, determine the soil color characteristics and soil texture characteristics corresponding to each soil image block for analysis, and calculate the average soil color characteristics and average soil texture characteristics of several soil image blocks for analysis; The third module is used to construct a soil comprehensive characteristic performance model. The soil comprehensive characteristic performance model is adjusted according to the average soil color characteristics and the average soil texture characteristics to obtain a soil comprehensive characteristic performance model for comparison. Based on the spinach sowing experiment record, the corresponding spinach sowing interval is configured for the soil comprehensive characteristic performance model for comparison; The fourth module is used to classify the soil comprehensive characteristic performance models for comparison with the same spinach sowing interval, and to classify the soil comprehensive characteristic performance models for comparison based on the similarity between the soil comprehensive characteristic performance models for comparison, so as to obtain a reference performance model set; The fifth module is used to analyze the soil comprehensive characteristic performance model for comparison in the reference performance model set, determine the marker factor parameter of each characteristic factor, mark the reference performance model set with the marker factor parameter, and construct several reference performance model sets into a reference performance model library; The sixth module is used to analyze the currently taken sowing soil image, and to construct the current soil comprehensive characteristic expression model based on the analysis results, and to determine the characteristic factor parameters in the current soil comprehensive characteristic expression model, and to find the corresponding reference expression model set in the reference expression model library based on the parameter difference value of the characteristic factor parameters and the corresponding marker factor parameters, and to compare each comparison soil comprehensive characteristic expression model in the reference expression model set with the current soil comprehensive characteristic expression model, and to determine the degree of consistency between the two based on the comparison results, and to determine the suitable comparison soil comprehensive characteristic expression model based on the degree of consistency, and to determine the position of spinach sowing based on the spinach sowing interval corresponding to the comparison soil comprehensive characteristic expression model.

Images