Method and device for measuring physiological parameters of peanut seeds based on CT images

Abstract

The embodiment of the present application provides a method and device for measuring the physiological parameters of peanut seeds based on CT images. By collecting a CT tomography sequence of peanut seeds and combining an image segmentation algorithm, a binary image of the fruit and pod area is obtained. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and a triangular mesh partitioning algorithm is used to achieve accurate three-dimensional reconstruction. The length, width and height parameters are obtained through spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated in combination with a surface mesh model, and the volume and void ratio are obtained through depth dimension integration. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and device for measuring physiological parameters of peanut kernels based on CT images. Background Art

[0002] The measurement of physiological parameters of peanut kernels is of great significance for agricultural breeding and quality evaluation. Traditional measurement methods mainly rely on manual measurement or two-dimensional image analysis, which have obvious limitations in measurement accuracy and efficiency.

[0003] From the perspective of measurement methods, existing solutions mainly use tools such as vernier calipers and two-dimensional images for measurement, which makes it difficult to accurately obtain the internal structural characteristics of peanut seeds. In particular, when measuring parameters such as porosity and internal morphology, traditional methods cannot achieve non-destructive testing, affecting the accuracy of measurement and the integrity of the sample.

[0004] From the perspective of data processing, traditional measurement methods lack standardized 3D reconstruction and parameter extraction processes. Manual measurement has subjective errors, and simple image analysis is difficult to accurately reflect the spatial characteristics of seeds. At the same time, existing methods are difficult to obtain multiple physiological parameters at the same time, which increases the measurement workload.

[0005] From the application effect point of view, the existing technology is difficult to meet the requirements of seed breeding and quality assessment for measurement accuracy. Especially in batch measurement scenarios, traditional methods are inefficient and it is difficult to ensure the consistency and repeatability of measurement results.

[0006] Therefore, how to achieve high-precision automated measurement of peanut seed physiological parameters is an urgent problem to be solved in current agricultural measurement technology. This is not only related to the scientific nature of seed industry research, but also an important foundation for improving the level of agricultural modernization. Summary of the invention

[0007] In response to the problems in the prior art, the present application provides a method and device for measuring physiological parameters of peanut kernels based on CT images, which can achieve high-precision automated measurement of physiological parameters of peanut kernels.

[0008] In order to solve at least one of the above problems, the present application provides the following technical solutions:

[0009] In a first aspect, the present application provides a method for measuring physiological parameters of peanut seeds based on CT images, comprising:

[0010] Collecting a sequence of CT tomographic scan images of peanut kernels, obtaining binary images of the fruit region and pod region of the peanut kernels through an image segmentation algorithm, converting the pixel coordinates of the binary images into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, calculating the depth coordinate value in the three-dimensional coordinate system according to CT scan layer number information, and generating spatial point cloud data including the fruit region and the pod region;

[0011] Performing three-dimensional reconstruction processing on the spatial point cloud data, using a triangular meshing algorithm to construct a surface mesh model of the fruit region and the pod region, extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection point of the major axis vector with the three-dimensional model as a length parameter, determining the intersection point of the minor axis vector with the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection point of the normal vector with the three-dimensional model as a height parameter;

[0012] The surface area parameters of the fruit region and the pod region are calculated based on the surface mesh model, and the volume parameters of the fruit region and the pod region are obtained by performing depth dimension integration operation on the fruit region and the pod region areas of each layer of the CT tomography image sequence. The void ratio parameter is calculated according to the difference between the fruit region volume parameter and the pod region volume parameter, and the length parameter, width parameter, height parameter, surface area parameter, volume parameter and void ratio parameter are written into a peanut seed physiological characteristic database.

[0013] Furthermore, the CT tomography image sequence of peanut kernels is collected, binary images of the fruit region and pod region of the peanut kernels are obtained by an image segmentation algorithm, and pixel coordinates of the binary images are converted into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, including:

[0014] The CT scanning device performs multi-angle tomographic scanning on peanut kernels through a rotating X-ray tube, converts the scanning signal into a digital image sequence, extracts the edge contours of the fruit region and the pod region from the digital image sequence using a threshold segmentation and region growing algorithm, and generates a binary image set containing labels of the fruit region and the pod region;

[0015] The geometric calibration parameters and imaging calibration parameters of the CT device are read, the pixel coordinates in the binary image set are transformed into a spatial coordinate system according to the calibration parameter matrix, the fruit region coordinate data and the pod region coordinate data in the three-dimensional coordinate system of the physical world are generated, and the coordinate data are written into the spatial data buffer area.

[0016] Furthermore, the depth coordinate value in the three-dimensional coordinate system is calculated according to the CT scanning layer number information to generate spatial point cloud data including the fruit area and the pod area, including:

[0017] Reading the inter-slice spacing parameter and the number of scanning layers of the CT scanning sequence, performing depth dimension mapping on the coordinate data in the spatial data buffer area, calculating the depth values ​​of the coordinate points of the fruit area and the pod area in each layer of the image, and combining the depth coordinate values ​​with the original two-dimensional coordinate data to generate three-dimensional coordinate data;

[0018] The three-dimensional coordinate data are classified and sorted according to the label information of the fruit area and the pod area, a point cloud data structure including the spatial position and the regional attributes is constructed, the point cloud data is written into the three-dimensional data storage area, and a point cloud data index table is established.

[0019] Further, performing three-dimensional reconstruction processing on the spatial point cloud data, using a triangular meshing algorithm to construct a surface mesh model of the fruit region and the pod region, and extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, includes:

[0020] Reading the point cloud data in the three-dimensional data storage area, performing noise filtering and spatial smoothing on the point cloud data, constructing an initial mesh model based on a Delaunay triangulation algorithm, and generating a fine surface mesh model of a fruit region and a pod region through mesh boundary optimization and a surface fitting algorithm;

[0021] A boundary extraction algorithm is executed on the surface mesh model to calculate the curvature and normal vector distribution of the mesh vertices, identify the characteristic edges and contour lines of the mesh surface, and store the extracted spatial contour data in a geometric feature data table according to the region type.

[0022] Further, the calculating of the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection of the major axis vector and the three-dimensional model as a length parameter, determining the intersection of the minor axis vector and the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection of the normal vector and the three-dimensional model as a height parameter, comprises:

[0023] Performing principal direction analysis on the contour data in the geometric feature data table, calculating the covariance matrix of the contour point set, obtaining the eigenvector corresponding to the maximum eigenvalue as the major axis vector and the eigenvector corresponding to the minimum eigenvalue as the minor axis vector through eigenvalue decomposition, and calculating the coordinates of the intersection of the major axis vector and the minor axis vector with the surface mesh model;

[0024] A cross multiplication operation is performed based on the major axis vector and the minor axis vector to obtain a normal vector, the normal vector is normalized and then intersected with the surface mesh model, the Euclidean distance between the intersection points is calculated to obtain a height parameter, the major axis vector intersection distance is used as a length parameter, the minor axis vector intersection distance is used as a width parameter, and the parameter data is written into the size feature table.

[0025] Furthermore, the surface area parameters of the fruit area and the pod area are calculated based on the curved mesh model, and the volume parameters of the fruit area and the pod area are obtained by performing a depth dimension integration operation on the fruit area and the pod area of ​​each layer of the CT tomography image sequence, including:

[0026] Reading the triangular face data of the surface mesh model, calculating the area of ​​each triangular face and performing cumulative summation to obtain surface area parameters of the fruit region and the pod region respectively, performing connected domain labeling on the binary regions in the CT tomography image sequence, and calculating the pixel area of ​​the fruit region and the pod region in each layer of the image;

[0027] The pixel area is multiplied by the pixel resolution of the CT device to obtain the actual area value, and a depth dimension integral function is constructed according to the CT scanning layer spacing parameters. An integral operation is performed on the fruit area and pod area of ​​each layer of the image to obtain the volume parameters of the fruit area and pod area, and the surface area parameters and volume parameters are written into the morphological feature table.

[0028] Furthermore, the gap ratio parameter is calculated according to the difference between the fruit area volume parameter and the pod area volume parameter, and the length parameter, width parameter, height parameter, surface area parameter, volume parameter and gap ratio parameter are written into the peanut seed physiological characteristic database, including:

[0029] Reading the fruit region volume parameter and the pod region volume parameter in the morphological feature table, calculating the difference between the two volume parameters to obtain the cavity volume, dividing the cavity volume by the pod region volume parameter to calculate the void ratio parameter, and generating a void feature record according to the calculation result;

[0030] Length parameters, width parameters, height parameters, surface area parameters, and volume parameters are extracted from the size feature table and the morphological feature table, combined with the gap ratio parameters in the gap feature record to construct a physiological feature data structure, and the physiological feature data is written into the parameter table of the peanut seed feature database.

[0031] In a second aspect, the present application provides a device for measuring physiological parameters of peanut seeds based on CT images, comprising:

[0032] A point cloud data construction module is used to collect a sequence of CT tomographic scan images of peanut kernels, obtain binary images of the fruit region and pod region of the peanut kernels through an image segmentation algorithm, convert the pixel coordinates of the binary images into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, calculate the depth coordinate value in the three-dimensional coordinate system according to the CT scan layer number information, and generate spatial point cloud data including the fruit region and the pod region;

[0033] a three-dimensional data analysis module, for performing three-dimensional reconstruction processing on the spatial point cloud data, constructing a surface mesh model of the fruit region and the pod region using a triangular mesh subdivision algorithm, extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection point of the major axis vector with the three-dimensional model as a length parameter, determining the intersection point of the minor axis vector with the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection point of the normal vector with the three-dimensional model as a height parameter;

[0034] The physiological parameter measurement module is used to calculate the surface area parameters of the fruit area and the pod area based on the surface mesh model, obtain the volume parameters of the fruit area and the pod area by performing depth dimension integration operation on the fruit area and the pod area area of ​​each layer of the CT tomography image sequence, calculate the gap ratio parameter according to the difference between the fruit area volume parameter and the pod area volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and gap ratio parameter into the peanut seed physiological characteristic database.

[0035] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for measuring physiological parameters of peanut seeds based on CT images are implemented.

[0036] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for measuring physiological parameters of peanut kernels based on CT images.

[0037] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for measuring physiological parameters of peanut kernels based on CT images.

[0038] It can be seen from the above technical solution that the present application provides a method and device for measuring the physiological parameters of peanut seeds based on CT images. By collecting the CT tomography sequence of peanut seeds and combining the image segmentation algorithm, the binary images of the fruit and pod areas are obtained. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and the triangular mesh subdivision algorithm is used to achieve accurate three-dimensional reconstruction. The length, width and height parameters are obtained through spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated by combining the surface mesh model, and the volume and void ratio are obtained by depth dimension integration. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is one of the flow charts of the method for measuring physiological parameters of peanut seeds based on CT images in the embodiment of the present application;

[0041] Figure 2 This is a second flow chart of the method for measuring physiological parameters of peanut seeds based on CT images in an embodiment of the present application;

[0042] Figure 3 This is a third flow chart of the method for measuring physiological parameters of peanut seeds based on CT images in the embodiment of the present application;

[0043] Figure 4 This is a fourth flow chart of the method for measuring physiological parameters of peanut seeds based on CT images in an embodiment of the present application;

[0044] Figure 5 This is a fifth flow chart of the method for measuring physiological parameters of peanut seeds based on CT images in the embodiment of the present application;

[0045] Figure 6 This is a sixth flow chart of the method for measuring physiological parameters of peanut seeds based on CT images in the embodiment of the present application;

[0046] Figure 7 FIG7 is a flow chart of a method for measuring physiological parameters of peanut seeds based on CT images in an embodiment of the present application;

[0047] Figure 8 It is a structural diagram of a peanut seed physiological parameter measurement device based on CT images in an embodiment of the present application;

[0048] Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0049] Reference numerals:

[0050] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0052] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0053] Taking into account the problems existing in the prior art, the present application provides a method and device for measuring the physiological parameters of peanut seeds based on CT images. By collecting a CT tomography sequence of peanut seeds and combining an image segmentation algorithm, a binary image of the fruit and pod area is obtained. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and a triangular mesh partitioning algorithm is used to achieve accurate three-dimensional reconstruction. The length, width and height parameters are obtained through spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated by combining a surface mesh model, and the volume and void ratio are obtained by integrating the depth dimension. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

[0054] In order to achieve high-precision automated measurement of physiological parameters of peanut seeds, the present application provides an embodiment of a method for measuring physiological parameters of peanut seeds based on CT images, see Figure 1 The method for measuring physiological parameters of peanut kernels based on CT images specifically includes the following contents:

[0055] Step S101: collecting a sequence of CT tomographic scan images of peanut kernels, obtaining binary images of the fruit region and pod region of the peanut kernels through an image segmentation algorithm, converting the pixel coordinates of the binary images into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, calculating the depth coordinate value in the three-dimensional coordinate system according to CT scan layer number information, and generating spatial point cloud data including the fruit region and pod region;

[0056] Optionally, this embodiment first performs tomographic imaging of the peanut kernel sample by industrial CT equipment. During the scanning process, the X-ray tube rotates around the sample to collect multi-angle projection data, and the projection data at each angle is digitized to form a high-resolution grayscale image sequence. By adjusting the tube voltage and tube current parameters of the CT device, it is ensured that a clear image of the internal structure of the peanut kernel is obtained, especially the boundary features between the fruit and the pod.

[0057] In order to accurately extract the fruit and pod regions of peanut seeds, this embodiment adopts a multi-stage image segmentation strategy. First, the CT image sequence is pre-processed for noise reduction, and Gaussian filtering is used to eliminate image noise while maintaining edge features. Then, based on the image grayscale histogram features, an adaptive threshold segmentation method is used to preliminarily separate the target area and background. Considering the complexity of the internal structure of peanut seeds, this embodiment further introduces a region growing algorithm, using grayscale similarity and spatial adjacency as growth criteria to accurately extract the boundary contours of fruits and pods.

[0058] For the segmented regions, this embodiment applies morphological processing technology to optimize the boundary contour. Small noise points are removed by opening operations, and boundary gaps are filled by closing operations, and finally a complete set of binary images is generated. In each binary image, the fruit region and the pod region are marked with different label values, respectively, for convenience of subsequent processing.

[0059] To achieve accurate conversion from pixel coordinates to physical world coordinates, this embodiment reads the geometric calibration data of the CT device, including parameters such as focal length, detector pixel size, and rotation center offset. A coordinate transformation matrix is ​​constructed based on these parameters to map the two-dimensional pixel coordinates to the three-dimensional space of the real physical scale. Considering the possible distortion of the CT device, this embodiment also introduces a nonlinear correction term to ensure the accuracy of the coordinate transformation.

[0060] After obtaining the plane coordinates, this embodiment calculates the depth coordinate value of each data point according to the inter-slice spacing parameter and the number of layers of the CT scan. Considering the possible position deviation between different scanning layers, an interpolation algorithm is used to optimize the depth coordinates to ensure the continuity and smoothness of the spatial points.

[0061] By integrating plane coordinates and depth information, this embodiment generates a three-dimensional point cloud data structure containing position coordinates and regional attributes. Each point in the point cloud data carries the label information of the fruit area or pod area, and records its precise position in the physical world. In order to improve the efficiency of subsequent processing, this embodiment establishes a spatial index structure of the point cloud data to support fast regional query and feature extraction.

[0062] The technical solution of this embodiment solves the problem of difficulty in obtaining accurate internal structure in traditional peanut seed detection, and realizes accurate segmentation and spatial reconstruction of fruit and pod regions. By introducing CT imaging and precise coordinate transformation, a three-dimensional point cloud expression with real physical scale is obtained, laying the foundation for subsequent morphological feature measurement. This solution is particularly suitable for the evaluation of germplasm resources in the process of peanut breeding, and can obtain the internal structural characteristics of seeds without loss, providing more comprehensive and accurate phenotypic data support.

[0063] Step S102: performing three-dimensional reconstruction processing on the spatial point cloud data, constructing a surface mesh model of the fruit region and the pod region using a triangular meshing algorithm, extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection point of the major axis vector with the three-dimensional model as a length parameter, determining the intersection point of the minor axis vector with the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection point of the normal vector with the three-dimensional model as a height parameter;

[0064] Optionally, after obtaining the spatial point cloud data of peanut kernels, this embodiment first pre-processes the point cloud data to improve the quality of three-dimensional reconstruction. The statistical outlier filtering method is used to remove noise points, and the abnormal points that do not conform to the normal distribution assumption are eliminated by analyzing the distance distribution characteristics of the points in the neighborhood of each point. At the same time, the moving least squares method is applied to smooth the point cloud to reduce data noise while maintaining edge features.

[0065] In order to construct an accurate surface mesh model, this embodiment adopts an improved Delaunay triangulation algorithm. First, an initial triangular mesh is constructed on the point cloud projection plane, and then the mesh topology is adjusted through iterative optimization. During the optimization process, the curvature variation characteristics of the peanut seed surface are focused on, and the triangulation density is increased in areas with larger curvature to ensure accurate description of the surface details of the fruit and pod.

[0066] For the boundary area of ​​the mesh model, this embodiment introduces a boundary preservation strategy. By analyzing the local connection relationship of the mesh vertices, the characteristic points at the model boundary are identified, and the smoothness of the boundary contour is optimized by using a curve fitting method. For characteristic areas such as the connection between fruits and pods, the sampling point density is increased to improve the reconstruction accuracy.

[0067] After obtaining the surface mesh model, this embodiment determines the main directional characteristics of the peanut seed through the principal component analysis method. The mesh vertex coordinates are organized into a covariance matrix, and the three main directions are obtained through eigenvalue decomposition. The eigenvector corresponding to the largest eigenvalue is defined as the major axis vector, indicating the main growth direction of the seed; the eigenvector corresponding to the second largest eigenvalue is defined as the minor axis vector, indicating the lateral distribution direction of the seed.

[0068] In order to accurately measure the length and width parameters of the seed, this embodiment calculates the intersection of the major axis vector and the minor axis vector with the surface mesh model. Using the ray-triangle patch intersection test algorithm, rays are emitted from the center of the seed in the positive and negative directions to obtain the coordinates of the intersection with the mesh model. The Euclidean distance between the intersection points is defined as the length and width parameters of the seed, respectively.

[0069] When determining the height parameter, this embodiment obtains the normal vector by cross-product operation of the major axis vector and the minor axis vector, and the normal vector is perpendicular to the main plane and points to the thickness direction of the seed. After the normal vector is normalized, the ray intersection test is also used to obtain the intersection with the grid model, and the distance between the intersections is defined as the height parameter of the seed.

[0070] The technical solution of this embodiment solves the problem that it is difficult to accurately obtain three-dimensional morphological characteristics in traditional peanut seed measurement methods. Through accurate three-dimensional reconstruction and feature direction extraction, automatic measurement of the main size parameters of the seeds is achieved. This solution is particularly suitable for rapid detection and analysis of large quantities of peanut seeds, overcoming the subjectivity and inefficiency of manual measurement, while providing more comprehensive morphological characteristic data, which can be used for phenotypic analysis in the process of germplasm resource evaluation and variety breeding.

[0071] In the process of obtaining the morphological characteristics of peanut seeds, this embodiment obtains complete surface geometric information through three-dimensional reconstruction technology, and realizes accurate measurement of size parameters in combination with principal direction analysis, providing reliable data support for subsequent germplasm resource evaluation. Compared with traditional measurement methods, this solution has higher accuracy and repeatability, and can support breeding experts to make more accurate phenotypic analysis and breeding decisions.

[0072] Step S103: Calculate the surface area parameters of the fruit region and the pod region based on the surface mesh model, obtain the volume parameters of the fruit region and the pod region by performing depth dimension integration operation on the fruit region and the pod region areas of each layer of the CT tomography image sequence, calculate the void ratio parameter according to the difference between the fruit region volume parameter and the pod region volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and void ratio parameter into the peanut seed physiological characteristic database.

[0073] Optionally, after obtaining the curved mesh model of the peanut seed, the present embodiment first calculates the surface area parameters of the fruit region and the pod region. For each triangular mesh patch, the vector cross product method is used to calculate its area, and the total surface area is obtained by accumulating the areas of all patches. Considering the possible irregularities of the mesh model, the present embodiment introduces a mesh optimization strategy to merge triangular patches with too small an area and split patches with too large an area to ensure the accuracy of the area calculation.

[0074] In the volume parameter calculation, this embodiment adopts the layered integration method. First, the CT tomography sequence is registered according to the scanning layer spacing to ensure the correspondence between adjacent layers. For the fruit area and pod area obtained by segmentation in each layer of CT image, the pixel counting method is used to obtain its two-dimensional area. Combined with the voxel resolution parameter of the CT device, the number of pixels is converted into the actual physical area.

[0075] In order to improve the accuracy of volume calculation, this embodiment adopts an interpolation strategy between adjacent layers. By analyzing the shape changes of the target area in the continuous layers, the cubic spline interpolation method is used to estimate the area changes of the transition area between layers. This method is particularly suitable for processing the irregular shapes formed by peanut kernels during the growth process, and can more accurately reflect the actual volume of the kernels.

[0076] After obtaining the area of ​​each layer, this embodiment calculates the total volume by numerical integration of the depth dimension. Considering the gradual change characteristics of the cross-sectional shape of the peanut seed, the trapezoidal integration method is used to accumulate the area sequence. By introducing an adaptive step size strategy, the sampling density is increased in areas with drastic shape changes, thereby improving the integration accuracy.

[0077] The calculation of the gap ratio parameter is an important innovation of this embodiment. By comparing the volume difference between the pod area and the fruit area, the fullness of the peanut grain during development can be evaluated. This embodiment first calculates the volume difference, and then divides it by the total volume of the pod to obtain a standardized gap ratio index. This parameter is of great significance for evaluating the development quality and physiological state of peanut grains.

[0078] In order to facilitate subsequent analysis and application, this embodiment designs a special data storage structure. In the peanut seed physiological characteristics database, an independent data record is established for each sample, including metadata fields such as unique identifier, collection time, variety information, etc. The measured parameters such as length, width, height, surface area, volume and void ratio are stored in a unified format, and indexes are established to support fast query and statistical analysis.

[0079] The technical solution of this embodiment solves the problem that it is difficult to simultaneously obtain multi-dimensional morphological parameters in traditional peanut seed detection. By combining CT imaging and three-dimensional reconstruction technology, a complete characterization of the internal and external structure of the seed is achieved. This solution can not only obtain conventional external dimension parameters, but also accurately measure the internal void characteristics, providing a new quantitative indicator for breeding experts to evaluate the quality of seed development.

[0080] The characteristic parameter system established in this embodiment comprehensively reflects the morphological and structural characteristics of peanut seeds, especially the degree of seed filling through the gap ratio parameter, which has important guiding significance for germplasm resource evaluation and variety breeding. The establishment of the database facilitates the management and analysis of large-scale germplasm resources and supports breeding experts to conduct more systematic and in-depth research.

[0081] From the above description, it can be seen that the peanut seed physiological parameter measurement method based on CT images provided in the embodiment of the present application can obtain binary images of the fruit and pod area by collecting the CT tomography sequence of peanut seeds and combining the image segmentation algorithm. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and the triangular mesh subdivision algorithm is used to achieve accurate three-dimensional reconstruction. The length, width and height parameters are obtained by spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated by combining the surface mesh model, and the volume and void ratio are obtained by depth dimension integration. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

[0082] In one embodiment of the peanut seed physiological parameter measurement method based on CT images of the present application, see Figure 2 , and can also include the following:

[0083] Step S201: the CT scanning device performs multi-angle tomographic scanning on peanut kernels through a rotating X-ray tube, converts the scanning signals into a digital image sequence, extracts edge contours of the fruit region and the pod region from the digital image sequence using a threshold segmentation and region growing algorithm, and generates a binary image set containing labels of the fruit region and the pod region;

[0084] Step S202: Read the geometric calibration parameters and imaging calibration parameters of the CT device, transform the pixel coordinates in the binary image set into a spatial coordinate system according to the calibration parameter matrix, generate fruit region coordinate data and pod region coordinate data in the three-dimensional coordinate system of the physical world, and write the coordinate data into a spatial data cache.

[0085] Optionally, this embodiment first performs tomographic scanning on the peanut kernel sample by a rotating X-ray tube. During the scanning process, the X-ray tube rotates around the sample at a constant angular velocity, emitting an X-ray beam at each angular position and receiving a penetration signal by the detector. By adjusting the tube voltage and tube current parameters, the ray energy is optimized to obtain the best image contrast, especially for displaying the density difference between the fruit and pod tissues in the peanut kernel.

[0086] In the signal acquisition process, this embodiment uses a high-precision analog-to-digital converter to convert the analog signal received by the detector into a digital signal. In order to improve the signal quality, multi-stage signal amplification and filtering processing are introduced to effectively suppress the interference of electronic noise. The digital signal is processed by a reconstruction algorithm to generate a grayscale image sequence reflecting the internal structure of the peanut kernel.

[0087] Image segmentation is a key link in this embodiment. First, the image sequence is preprocessed, and the median filter is used to remove discrete noise while maintaining edge features. By analyzing the grayscale histogram features of the image, the maximum inter-class variance method is used to determine the initial segmentation threshold, and the image is divided into the target area and the background area.

[0088] In order to improve the accuracy of segmentation, this embodiment designs a two-stage segmentation strategy. Based on the threshold segmentation, an improved region growing algorithm is introduced. The algorithm starts with the seed point obtained by the initial segmentation, and gradually expands the region boundary according to the similarity of the pixel gray value and the spatial continuity criterion. During the growing process, the similarity threshold is dynamically adjusted to adapt to the grayscale change characteristics of different regions.

[0089] For the boundary area between the fruit and the pod, this embodiment particularly considers the processing of the tissue transition zone. By calculating the grayscale gradient of the local area, the boundary position is accurately located at the position with the larger gradient. At the same time, the morphological processing technology is introduced to optimize the smoothness of the boundary contour using the opening and closing operations.

[0090] After completing the region segmentation, this embodiment assigns different label values ​​to the fruit region and the pod region to generate a binary image set. Through connected region analysis, the noise region with too small area is eliminated to ensure the integrity and accuracy of the segmentation result.

[0091] Spatial coordinate conversion is a key step in achieving 3D reconstruction from 2D images. This embodiment first reads the geometric calibration data of the CT device, including parameters such as source-detector distance, detector pixel size, and rotation center coordinates. Based on these parameters, a projection matrix is ​​constructed to establish a mapping relationship between the pixel coordinate system and the physical world coordinate system.

[0092] In the coordinate conversion process, this embodiment takes into account the systematic error of the CT device. The imaging calibration parameters obtained by the calibration plate compensate for the geometric distortion and projection distortion of the device. For each marker point, its position in the physical world coordinate system is calculated through the inverse transformation of the projection matrix.

[0093] In order to improve data processing efficiency, this embodiment designs a special spatial data cache structure. In the cache area, the coordinate data of the fruit area and the pod area are organized according to the spatial index to support fast spatial query and feature extraction operations.

[0094] The technical solution of this embodiment solves the problems of inaccurate image segmentation and inaccurate spatial positioning in traditional peanut seed detection. Through precise CT imaging and efficient image processing algorithms, accurate segmentation of the fruit and pod regions is achieved. Combined with strict spatial coordinate transformation, it lays the foundation for subsequent three-dimensional reconstruction and feature extraction. This solution is particularly suitable for research scenarios that require accurate acquisition of the internal structure of peanut seeds, and can provide high-quality spatial morphological data support.

[0095] In one embodiment of the peanut seed physiological parameter measurement method based on CT images of the present application, see Figure 3 , and can also include the following:

[0096] Step S301: reading the inter-slice spacing parameter and the number of scanning layers of the CT scanning sequence, performing depth dimension mapping on the coordinate data in the spatial data buffer area, calculating the depth values ​​of the coordinate points of the fruit area and the pod area in each layer of the image, and combining the depth coordinate values ​​with the original two-dimensional coordinate data to generate three-dimensional coordinate data;

[0097] Step S302: Classify and sort the three-dimensional coordinate data according to the label information of the fruit area and the pod area, construct a point cloud data structure including spatial position and regional attributes, write the point cloud data into the three-dimensional data storage area, and establish a point cloud data index table.

[0098] Optionally, this embodiment first extracts the inter-slice spacing and scanning layer number information from the scanning parameters of the CT device, which determine the accuracy and resolution of the three-dimensional reconstruction of the peanut kernel. The inter-slice spacing parameter reflects the physical distance between adjacent scanning layers, and the scanning layer number determines the sampling density in the depth direction. In order to ensure the reconstruction accuracy, this embodiment selects appropriate scanning parameters according to the actual size of the peanut kernel to ensure that the sampling points can fully express the morphological characteristics of the kernel.

[0099] In the depth dimension mapping process, this embodiment uses a linear mapping method to convert the layer number into a physical depth value. Considering the possible inter-layer non-uniformity of CT scanning, a correction factor is introduced to compensate for the depth value. For each coordinate point on the scanning layer, the corresponding depth coordinate is calculated according to the number of the layer to which it belongs, realizing the mapping from the two-dimensional plane to the three-dimensional space.

[0100] In order to deal with the data interpolation problem in the depth direction, this embodiment designs an adaptive interpolation strategy. In the transition area between the fruit and the pod, by analyzing the contour changes of the adjacent layers, the cubic spline interpolation method is used to generate the coordinate data of the transition layer to improve the continuity of the reconstructed model.

[0101] In the process of coordinate data integration, this embodiment uses a unified data structure to store three-dimensional coordinate information. Each coordinate point contains the position information of the three spatial dimensions of X, Y, and Z, while retaining the label attribute of the area to which it belongs. By establishing the topological relationship between the coordinate points, a complete three-dimensional space description is formed.

[0102] The organization of point cloud data is an important innovation of this embodiment. According to the label information of the fruit area and the pod area, the coordinate points are classified and stored. In addition to the spatial position information, each point also records attribute data such as density and grayscale. This multi-attribute point cloud structure provides rich information support for subsequent feature extraction and analysis.

[0103] In order to improve data access efficiency, this embodiment designs a multi-level point cloud index structure. In the spatial dimension, an octree structure is used to spatially divide the point cloud data to support fast regional query and neighborhood search. At the same time, an attribute index table is established to facilitate data screening according to regional labels or other attributes.

[0104] This embodiment adopts a compression strategy when storing point cloud data, and reduces storage space occupancy by encoding and compressing coordinate values ​​and attribute data. At the same time, an efficient data decompression algorithm is designed to ensure that the original data can be quickly restored when needed.

[0105] In the process of index table construction, this embodiment takes into account the needs of subsequent analysis. In addition to the basic spatial index, special index structures such as feature point index and boundary point index are also established to support rapid positioning and extraction of key features. Through reasonable index design, the efficiency of subsequent processing steps is significantly improved.

[0106] The technical solution of this embodiment solves the problems of complex data organization and low retrieval efficiency in traditional three-dimensional reconstruction. Through precise depth mapping and efficient data organization, the complete expression of the three-dimensional morphology of peanut kernels is achieved. This solution is particularly suitable for breeding research that requires a large number of sample analyses and can provide fast and accurate data support.

[0107] This embodiment provides a reliable data basis for the analysis of peanut kernel morphological characteristics by establishing a complete three-dimensional point cloud data structure. Reasonable data organization and indexing mechanism not only improves data processing efficiency, but also facilitates subsequent feature extraction and analysis, supporting breeding experts to conduct more in-depth research.

[0108] In one embodiment of the peanut seed physiological parameter measurement method based on CT images of the present application, see Figure 4 , and can also include the following:

[0109] Step S401: reading the point cloud data in the three-dimensional data storage area, performing noise filtering and spatial smoothing on the point cloud data, constructing an initial mesh model based on the Delaunay triangulation algorithm, and generating fine surface mesh models of the fruit area and the pod area through mesh boundary optimization and surface fitting algorithms;

[0110] Step S402: Execute a boundary extraction algorithm on the surface mesh model, calculate the curvature and normal vector distribution of the mesh vertices, identify the characteristic edges and contour lines of the mesh surface, and store the extracted spatial contour data in a geometric feature data table according to the region type.

[0111] Optionally, this embodiment first pre-processes the point cloud data, and uses a statistical outlier analysis method to identify and remove noise points. By calculating the local density and neighborhood distribution characteristics of each point, an adaptive threshold is set to effectively remove abnormal points generated during the sampling process. For the fine structure of the surface of the peanut seed, a local weighted average method is used for smoothing, which improves the surface smoothness while maintaining the morphological characteristics.

[0112] When constructing the initial mesh model, this embodiment uses an improved Delaunay triangulation algorithm. First, feature points are selected from the point cloud data as initial seed points, and a triangular mesh is constructed by gradually expanding. In order to adapt to the complex curvature changes on the surface of the peanut seed, curvature constraints are introduced in the triangulation process to ensure that the generated triangular facets can better fit the actual shape.

[0113] Mesh optimization is a key step to improve model accuracy. This embodiment designs a multi-level optimization strategy. First, the topological structure of the mesh is optimized, too small triangular facets are merged, and severely distorted facets are split. Then, the sampling density is increased in the transition area between the fruit and the pod through local re-meshing technology to improve the accuracy of the boundary description.

[0114] In the surface fitting process, this embodiment adopts a surface reconstruction method based on NURBS. By analyzing the distribution characteristics of mesh vertices, a control point mesh is constructed, and a smooth parametric surface is obtained by least square fitting. In order to process the local features of the peanut seed surface, an adaptive weight strategy is introduced to increase the fitting weight in the area with obvious features.

[0115] Boundary extraction is an important innovation of this embodiment. By calculating the Gaussian curvature and mean curvature of the mesh vertices, the characteristic points and characteristic lines on the surface are identified. In areas where the curvature changes dramatically, a dynamic programming method is used to track the characteristic edges to form continuous contour lines. For the boundary between fruits and pods, the change of depth information is specially considered to improve the accuracy of boundary positioning.

[0116] In the normal vector distribution analysis, this embodiment adopts the local coordinate system transformation method. By calculating the normal vector of each vertex and the angle between adjacent facets, the sudden change position and gradual change area of ​​the surface are identified. This method is particularly suitable for describing the wrinkles and concave-convex features on the surface of peanut seeds.

[0117] In order to accurately extract the spatial contour, this embodiment designs a contour tracking algorithm based on curvature flow. Starting from the curvature extreme point, continuous contour lines are searched along the curvature gradient direction. By setting a suitable stop criterion, over-segmentation or missed extraction of contour lines is avoided.

[0118] In the process of organizing geometric feature data, this embodiment establishes a multi-level feature description structure. In addition to basic spatial coordinates, differential geometric features such as curvature and normal vector are also recorded to provide rich feature descriptions for subsequent morphological analysis. Through reasonable data organization, fast feature matching and retrieval are supported.

[0119] The technical solution of this embodiment solves the problems of low model accuracy and inaccurate feature extraction in traditional 3D reconstruction. Through fine mesh modeling and feature extraction algorithms, high-fidelity reconstruction of peanut seed morphology is achieved. This solution is particularly suitable for research scenarios that require accurate analysis of seed morphological characteristics and can provide reliable morphological data support.

[0120] This embodiment provides a new technical means for the digital protection of peanut germplasm resources by establishing a complete geometric feature description system. Accurate morphological feature data not only helps to quantitatively describe variety characteristics, but also provides a scientific basis for the evaluation and utilization of germplasm resources.

[0121] In one embodiment of the peanut seed physiological parameter measurement method based on CT images of the present application, see Figure 5 , and can also include the following:

[0122] Step S501: performing principal direction analysis on the contour data in the geometric feature data table, calculating the covariance matrix of the contour point set, obtaining the eigenvector corresponding to the maximum eigenvalue as the major axis vector and the eigenvector corresponding to the minimum eigenvalue as the minor axis vector through eigenvalue decomposition, and calculating the intersection coordinates of the major axis vector and the minor axis vector with the surface mesh model;

[0123] Step S502: Perform a cross multiplication operation based on the major axis vector and the minor axis vector to obtain a normal vector, normalize the normal vector and intersect it with the surface mesh model, calculate the Euclidean distance between the intersection points to obtain a height parameter, use the major axis vector intersection distance as a length parameter, and the minor axis vector intersection distance as a width parameter, and write the parameter data into the size feature table.

[0124] Optionally, this embodiment first reads the contour data of peanut seeds from the geometric feature data table, which reflects the morphological distribution of the seeds in three-dimensional space. In order to accurately describe the main growth direction and morphological characteristics of the seeds, the main direction analysis method is used to process the contour point set. By calculating the centroid of the point set, all contour points are centered relative to the centroid to eliminate the influence of position offset.

[0125] In the process of calculating the covariance matrix, this embodiment takes into account the spatial distribution characteristics of the point set. Each contour point is given a weight based on the local curvature, so that it has a greater contribution in areas with more significant features. By constructing a three-dimensional covariance matrix, the spatial distribution characteristics of the seeds are encoded into the matrix.

[0126] Eigenvalue decomposition is a key step in determining the main direction. This embodiment uses an improved power iteration method to calculate the eigenvalues, and ensures the calculation accuracy by setting a convergence threshold. The eigenvector corresponding to the maximum eigenvalue represents the main extension direction of the seed, that is, the major axis direction; the eigenvector corresponding to the minimum eigenvalue represents the lateral growth direction, that is, the minor axis direction.

[0127] In order to accurately locate the intersection of the major axis and the minor axis with the seed surface, this embodiment designs a ray tracing algorithm with an adaptive step size. Starting from the centroid, bidirectional tracing is performed along the eigenvector direction, and the intersection position is accurately located by bisection. During the tracing process, the local curvature change of the surface is taken into account, and a smaller tracing step size is used in the high curvature area.

[0128] The calculation of the normal vector uses a vector cross multiplication operation to ensure that the obtained vector is perpendicular to both the major axis and the minor axis. Through normalization processing, a unit normal vector is obtained for subsequent height measurement. This embodiment particularly considers the direction consistency of the normal vector to ensure that it points to the outer surface of the seed.

[0129] When calculating geometric parameters, this embodiment adopts an improved Euclidean distance calculation method. For the length parameter, the spatial distance between the two intersection points of the long axis vector is calculated; the width parameter is determined by the distance between the intersection points of the short axis vector. In order to improve the measurement accuracy, the influence of the surface curvature is considered in the distance calculation.

[0130] The measurement of the height parameter is the innovation of this embodiment. By calculating the intersection of the normal vector and the curved surface, the maximum distance of the seed in the direction perpendicular to the major and minor axis plane is obtained. This measurement method can accurately reflect the thickness characteristics of the seed and avoid the error of the traditional orthogonal projection method.

[0131] In the process of organizing parameter data, this embodiment establishes a complete size feature table. In addition to the basic length, width and height parameters, it also records the curvature distribution in various directions, volume and other derived features, providing comprehensive data support for the morphological feature analysis of the seed grains.

[0132] The technical solution of this embodiment solves the problems of inaccurate determination of the main direction and large measurement errors of size parameters in traditional measurement methods. Through accurate main direction analysis and geometric parameter calculation, accurate quantification of peanut grain morphological characteristics is achieved. This solution is particularly suitable for digital characterization of germplasm resources and can provide objective and reliable morphological data for breeding experts.

[0133] The size characteristic system established in this embodiment provides a standardized measurement method for the characteristic description of peanut varieties. Through accurate morphological parameter measurement, it is not only convenient for morphological comparison between varieties, but also provides a scientific basis for the classification and screening of germplasm resources.

[0134] In one embodiment of the peanut seed physiological parameter measurement method based on CT images of the present application, see Figure 6 , and can also include the following:

[0135] Step S601: reading the triangular face data of the surface mesh model, calculating the area of ​​each triangular face and performing cumulative summation to obtain surface area parameters of the fruit region and the pod region respectively, performing connected domain labeling on the binary regions in the CT tomography image sequence, and calculating the pixel area of ​​the fruit region and the pod region in each layer of the image;

[0136] Step S602: Multiply the pixel area by the pixel resolution of the CT device to obtain the actual area value, construct a depth dimension integral function according to the CT scanning layer spacing parameters, perform integral operation on the fruit area and pod area of ​​each layer of the image, obtain the volume parameters of the fruit area and pod area, and write the surface area parameters and volume parameters into the morphological feature table.

[0137] Optionally, the present embodiment first reads triangular facet data from the surface mesh model, and these facets constitute a sliced ​​representation of the surface of the peanut seed. For each triangular facet, its area is calculated by a vector cross multiplication method. During the calculation process, the spatial direction of the facet is particularly considered to ensure the accuracy of the area calculation. By accumulating the areas of all triangular facets, the surface areas of the fruit region and the pod region are obtained respectively.

[0138] In the process of surface area calculation, this embodiment adopts an adaptive grid subdivision strategy. For areas with large curvature, the number of facets is increased by local subdivision to improve the accuracy of area calculation. This method is particularly suitable for wrinkles and concave-convex areas on the surface of peanut seeds, and can more accurately reflect the surface morphological characteristics.

[0139] The processing of CT tomographic images is the basis of volume calculation. This embodiment adopts an improved connected domain labeling algorithm to perform regional segmentation on the binarized image. Through eight-neighborhood connectivity analysis, the fruit and pod regions in each tomographic image are accurately identified, avoiding missed detection and false detection of regional labels.

[0140] In the pixel area calculation, this embodiment takes into account the imaging characteristics of the CT device. The pixel unit is converted into the actual physical size through the device calibration parameters. In order to improve the calculation accuracy, the sub-pixel level area estimation method is used in the processing of boundary pixels to reduce the discretization error.

[0141] The integration of the depth dimension is the key innovation of volume calculation. This embodiment designs a depth integration function based on spline interpolation, taking into account the morphological changes between adjacent layers, and constructs a smooth transition function to achieve more accurate volume estimation. In areas with large inter-layer spacing, virtual layers are generated by interpolation to improve integration accuracy.

[0142] In the process of calculating volume parameters, this embodiment adopts a composite integration method. First, the area of ​​the region is calculated on each fault plane, and then the total volume is calculated by a numerical integration method in combination with the layer spacing parameter. In order to deal with the inter-layer inhomogeneity that may exist in tomography, an adaptive weight coefficient is introduced for correction.

[0143] The organization and management of morphological characteristics is an important part of this embodiment. A unified morphological characteristic table is established, which includes basic parameters such as surface area and volume, and morphological indexes derived from these parameters. These indexes reflect the morphological characteristics of the seeds, such as compactness and roundness, and provide a basis for the quantitative description of variety characteristics.

[0144] This embodiment also considers the evaluation and control of measurement errors. By analyzing the noise level and resolution limit of the CT image, an error estimation model is established. In the parameter calculation process, multiple sampling and statistical analysis are performed to provide a reliable error range estimation.

[0145] The technical solution of this embodiment solves the problems of low accuracy and low efficiency in traditional morphological measurement. Through accurate surface area and volume calculation methods, accurate quantification of peanut seed morphological characteristics is achieved. This solution is particularly suitable for batch evaluation of germplasm resources and can provide fast and accurate morphological data.

[0146] The morphological characteristic measurement system established in this embodiment provides a reliable tool for phenotypic analysis of peanut varieties. Accurate morphological parameters not only contribute to the objective description of variety characteristics, but also provide quantitative indicators for the evaluation and screening of germplasm resources, supporting scientific decision-making in breeding work.

[0147] In one embodiment of the peanut seed physiological parameter measurement method based on CT images of the present application, see Figure 7 , and can also include the following:

[0148] Step S701: reading the fruit region volume parameter and the pod region volume parameter in the morphological feature table, calculating the difference between the two volume parameters to obtain the cavity volume, dividing the cavity volume by the pod region volume parameter to calculate the void ratio parameter, and generating a void feature record according to the calculation result;

[0149] Step S702: extracting length parameters, width parameters, height parameters, surface area parameters, and volume parameters from the size feature table and the morphological feature table, combining them with the gap ratio parameters in the gap feature record to construct a physiological feature data structure, and writing the physiological feature data into the parameter table of the peanut seed feature database.

[0150] Optionally, this embodiment first reads the volume parameter data in the morphological characteristic table to obtain the volume information of the fruit area and the pod area respectively. These two volume parameters reflect the spatial distribution characteristics of the internal structure of the peanut seed. By calculating the difference between the volumes of the two areas, the cavity volume inside the seed is obtained, which is of great significance for evaluating the fullness of the seed.

[0151] In the process of calculating the void ratio, this embodiment adopts a normalization processing method. By dividing the cavity volume by the total volume of the pod area, a standardized void ratio parameter is obtained. This calculation method eliminates the influence of seed size differences, making the void characteristics between seeds of different sizes comparable.

[0152] The generation of void feature records is the innovation of this embodiment. In addition to the basic void proportion, it also includes a spatial feature description of the cavity distribution. By analyzing the position distribution of the cavity inside the seed, the uniformity and maturity of the seed development can be judged. For example, when the seed is well developed, the void distribution should be relatively uniform and the proportion should be moderate.

[0153] In the process of constructing the physiological characteristic data structure, this embodiment adopts a multi-dimensional feature fusion strategy. The external morphological parameters (length, width, height) of the seed grains are correlated and analyzed with the internal structural parameters (surface area, volume, void ratio) to form a complete feature description. This multi-dimensional feature expression method can fully reflect the physiological state of the seed grains.

[0154] Parameter correlation analysis is an important part of this embodiment. By studying the relationship between various parameters, a mapping relationship between seed morphology and physiological characteristics is established. For example, by analyzing the relationship between the void ratio and volume parameters, the degree of development and fullness of the seed can be evaluated; by comparing the ratio of surface area to volume, the morphological regularity of the seed can be determined.

[0155] In the data organization process, this embodiment designs a hierarchical data structure. The parameter table not only stores the original measurement parameters, but also contains derived characteristics and evaluation indicators. This structural design facilitates subsequent data retrieval and analysis and supports multi-angle germplasm resource evaluation.

[0156] The quality control of feature data is a link that this embodiment pays special attention to. By setting the parameter value range and logical relationship constraints, abnormal data is marked and processed. At the same time, a data consistency check mechanism is established to ensure the reliability of the feature data stored in the database.

[0157] The technical solution of this embodiment solves the problem that the internal structure characteristics are difficult to quantify in the traditional seed evaluation method. Through precise gap feature analysis and multi-dimensional parameter integration, objective evaluation of peanut seed quality is achieved. This solution is particularly suitable for the screening and quality evaluation of germplasm resources, providing reliable data support for breeding work.

[0158] The physiological characteristic evaluation system established in this embodiment provides a new technical means for peanut variety improvement. Through the comprehensive analysis of the internal and external characteristics of the seeds, it can not only accurately evaluate the development status of the seeds, but also provide a quantitative selection basis for variety selection, thereby improving the scientificity and efficiency of breeding work.

[0159] By establishing a standardized feature database, this embodiment provides a basic platform for the management and utilization of peanut germplasm resources. The complete feature data supports breeding experts in variety comparison and selection, and provides data support for the protection and utilization of germplasm resources.

[0160] In order to achieve high-precision automated measurement of peanut seed physiological parameters, the present application provides an embodiment of a peanut seed physiological parameter measurement device based on CT images for realizing all or part of the contents of the peanut seed physiological parameter measurement method based on CT images, see Figure 8The peanut seed physiological parameter measurement device based on CT images specifically includes the following contents:

[0161] The point cloud data construction module 10 is used to collect a CT tomographic image sequence of peanut kernels, obtain a binary image of the fruit region and the pod region of the peanut kernels through an image segmentation algorithm, convert the pixel coordinates of the binary image into a three-dimensional coordinate system of the physical world in combination with the CT device calibration parameters, calculate the depth coordinate value in the three-dimensional coordinate system according to the CT scanning layer number information, and generate spatial point cloud data including the fruit region and the pod region;

[0162] A three-dimensional data analysis module 20 is used to perform three-dimensional reconstruction processing on the spatial point cloud data, construct a surface mesh model of the fruit region and the pod region using a triangular mesh subdivision algorithm, extract spatial contour information of the fruit region and the pod region based on the surface mesh model, calculate the major axis vector and the minor axis vector of the fruit region and the pod region, determine the intersection of the major axis vector and the three-dimensional model as a length parameter, determine the intersection of the minor axis vector and the three-dimensional model as a width parameter, determine a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determine the intersection of the normal vector and the three-dimensional model as a height parameter;

[0163] The physiological parameter measurement module 30 is used to calculate the surface area parameters of the fruit area and the pod area based on the surface mesh model, obtain the volume parameters of the fruit area and the pod area by performing depth dimension integration operation on the fruit area and the pod area of ​​each layer of the CT tomography image sequence, calculate the gap ratio parameter according to the difference between the fruit area volume parameter and the pod area volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and gap ratio parameter into the peanut seed physiological characteristic database.

[0164] From the above description, it can be seen that the peanut seed physiological parameter measurement device based on CT images provided in the embodiment of the present application can obtain binary images of the fruit and pod area by collecting the CT tomography sequence of peanut seeds and combining the image segmentation algorithm. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and the triangular mesh partitioning algorithm is used to achieve accurate three-dimensional reconstruction. The length, width and height parameters are obtained by spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated by combining the surface mesh model, and the volume and void ratio are obtained by depth dimension integration. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

[0165] From the hardware level, in order to achieve high-precision automatic measurement of peanut seed physiological parameters, the present application provides an embodiment of an electronic device for implementing all or part of the content of the peanut seed physiological parameter measurement method based on CT images, and the electronic device specifically includes the following content:

[0166] A processor, a memory, a communications interface and a bus; wherein the processor, the memory and the communications interface communicate with each other through the bus; the communications interface is used to realize information transmission between the peanut seed physiological parameter measurement device based on CT images and related devices such as core business systems, user terminals and related databases; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the peanut seed physiological parameter measurement method based on CT images and the embodiment of the peanut seed physiological parameter measurement device based on CT images in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0167] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0168] In practical applications, part of the method for measuring physiological parameters of peanut kernels based on CT images can be executed on the electronic device side as described above, or all operations can be completed in the client device. The selection can be made based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0169] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0170] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0171] In one embodiment, the function of the peanut seed physiological parameter measurement method based on CT images can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control:

[0172] Step S101: collecting a sequence of CT tomographic scan images of peanut kernels, obtaining binary images of the fruit region and pod region of the peanut kernels through an image segmentation algorithm, converting the pixel coordinates of the binary images into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, calculating the depth coordinate value in the three-dimensional coordinate system according to CT scan layer number information, and generating spatial point cloud data including the fruit region and pod region;

[0173] Step S102: performing three-dimensional reconstruction processing on the spatial point cloud data, constructing a surface mesh model of the fruit region and the pod region using a triangular meshing algorithm, extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection point of the major axis vector with the three-dimensional model as a length parameter, determining the intersection point of the minor axis vector with the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection point of the normal vector with the three-dimensional model as a height parameter;

[0174] Step S103: Calculate the surface area parameters of the fruit region and the pod region based on the surface mesh model, obtain the volume parameters of the fruit region and the pod region by performing depth dimension integration operation on the fruit region and the pod region areas of each layer of the CT tomography image sequence, calculate the void ratio parameter according to the difference between the fruit region volume parameter and the pod region volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and void ratio parameter into the peanut seed physiological characteristic database.

[0175] From the above description, it can be seen that the electronic device provided in the embodiment of the present application acquires a binary image of the fruit and pod area by collecting a CT tomography sequence of peanut seeds and combining an image segmentation algorithm. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and a triangular mesh partitioning algorithm is used to achieve accurate three-dimensional reconstruction. The length, width and height parameters are obtained by spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated by combining a surface mesh model, and the volume and void ratio are obtained by integrating the depth dimension. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

[0176] In another embodiment, the peanut seed physiological parameter measurement device based on CT images can be configured separately from the central processing unit 9100. For example, the peanut seed physiological parameter measurement device based on CT images can be configured as a chip connected to the central processing unit 9100, and the function of the peanut seed physiological parameter measurement method based on CT images can be implemented through the control of the central processing unit.

[0177] like Fig. 9 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0178] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0179] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0180] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0181] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0182] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0183] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0184] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0185] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the method for measuring physiological parameters of peanut kernels based on CT images in the above-mentioned embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the method for measuring physiological parameters of peanut kernels based on CT images in the above-mentioned embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0186] Step S101: collecting a sequence of CT tomographic scan images of peanut kernels, obtaining binary images of the fruit region and pod region of the peanut kernels through an image segmentation algorithm, converting the pixel coordinates of the binary images into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, calculating the depth coordinate value in the three-dimensional coordinate system according to CT scan layer number information, and generating spatial point cloud data including the fruit region and pod region;

[0187] Step S102: performing three-dimensional reconstruction processing on the spatial point cloud data, constructing a surface mesh model of the fruit region and the pod region using a triangular meshing algorithm, extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection point of the major axis vector with the three-dimensional model as a length parameter, determining the intersection point of the minor axis vector with the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection point of the normal vector with the three-dimensional model as a height parameter;

[0188] Step S103: Calculate the surface area parameters of the fruit region and the pod region based on the surface mesh model, obtain the volume parameters of the fruit region and the pod region by performing depth dimension integration operation on the fruit region and the pod region areas of each layer of the CT tomography image sequence, calculate the void ratio parameter according to the difference between the fruit region volume parameter and the pod region volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and void ratio parameter into the peanut seed physiological characteristic database.

[0189] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application acquires binary images of the fruit and pod regions by collecting a CT tomography sequence of peanut seeds and combining an image segmentation algorithm. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and a triangular mesh partitioning algorithm is used to achieve accurate three-dimensional reconstruction. The length, width, and height parameters are obtained by spatial analysis of the major axis vector, minor axis vector, and normal vector, the surface area is calculated by combining a surface mesh model, and the volume and void ratio are obtained by integrating the depth dimension. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

[0190] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the method for measuring physiological parameters of peanut kernels based on CT images in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the method for measuring physiological parameters of peanut kernels based on CT images are implemented. For example, the computer program / instruction implements the following steps:

[0191] Step S101: collecting a sequence of CT tomographic scan images of peanut kernels, obtaining binary images of the fruit region and pod region of the peanut kernels through an image segmentation algorithm, converting the pixel coordinates of the binary images into a three-dimensional coordinate system of the physical world in combination with CT equipment calibration parameters, calculating the depth coordinate value in the three-dimensional coordinate system according to CT scan layer number information, and generating spatial point cloud data including the fruit region and pod region;

[0192] Step S102: performing three-dimensional reconstruction processing on the spatial point cloud data, constructing a surface mesh model of the fruit region and the pod region using a triangular meshing algorithm, extracting spatial contour information of the fruit region and the pod region based on the surface mesh model, calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection point of the major axis vector with the three-dimensional model as a length parameter, determining the intersection point of the minor axis vector with the three-dimensional model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection point of the normal vector with the three-dimensional model as a height parameter;

[0193] Step S103: Calculate the surface area parameters of the fruit region and the pod region based on the surface mesh model, obtain the volume parameters of the fruit region and the pod region by performing depth dimension integration operation on the fruit region and the pod region areas of each layer of the CT tomography image sequence, calculate the void ratio parameter according to the difference between the fruit region volume parameter and the pod region volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and void ratio parameter into the peanut seed physiological characteristic database.

[0194] From the above description, it can be seen that the computer program product provided in the embodiment of the present application acquires binary images of the fruit and pod area by collecting CT tomography sequences of peanut seeds and combining image segmentation algorithms. The pixel coordinates are innovatively converted into three-dimensional physical coordinates, and accurate three-dimensional reconstruction is achieved by using a triangular mesh partitioning algorithm. The length, width and height parameters are obtained by spatial analysis of the major axis vector, minor axis vector and normal vector, the surface area is calculated by combining the surface mesh model, and the volume and void ratio are obtained by depth dimension integration. This application breaks through the limitations of traditional manual measurement and realizes high-precision automated measurement of peanut seed physiological parameters.

[0195] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0196] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0197] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0198] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0199] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for measuring physiological parameters of peanut seeds based on CT images, characterized in that: The method comprises: Collecting a CT tomographic image sequence of peanut kernels, obtaining a binary image of a fruit region and a pod region of the peanut kernels by an image segmentation algorithm, and converting the pixel coordinates of the binary image into a three-dimensional coordinate system of the physical world in combination with a CT device calibration parameter, specifically comprising: the CT device performs a multi-angle tomographic scan on the peanut kernels by a rotating X-ray tube, converting the scanning signal into a digital image sequence, extracting the edge contours of the fruit region and the pod region from the digital image sequence by using a threshold segmentation and a regional growth algorithm, and generating a binary image set containing labels of the fruit region and the pod region; reading the geometric calibration parameters and imaging calibration parameters of the CT device, performing a spatial coordinate system transformation on the pixel coordinates in the binary image set according to the calibration parameter matrix, generating fruit region coordinate data and pod region coordinate data in the three-dimensional coordinate system of the physical world, and writing the coordinate data into a spatial data buffer; Calculating the depth coordinate value in the three-dimensional coordinate system according to the CT scanning layer number information, generating spatial point cloud data including the fruit area and the pod area, specifically comprising: reading the layer spacing parameter and the scanning layer number information of the CT tomography image sequence, performing depth dimension mapping on the coordinate data in the spatial data buffer area, calculating the depth value of the coordinate point of the fruit area and the pod area in each layer of the image, combining the depth coordinate value with the original two-dimensional coordinate data to generate three-dimensional coordinate data; classifying and arranging the three-dimensional coordinate data according to the label information of the fruit area and the pod area, constructing a point cloud data structure including the spatial position and the regional attributes, writing the point cloud data into the three-dimensional data storage area, and establishing a point cloud data index table; The three-dimensional reconstruction process is performed on the spatial point cloud data, a surface mesh model of the fruit region and the pod region is constructed by using a triangulated mesh algorithm, and spatial contour information of the fruit region and the pod region is extracted based on the surface mesh model, specifically including: reading the point cloud data in the three-dimensional data storage area, performing noise filtering and spatial smoothing on the point cloud data, constructing an initial mesh model based on a Delaunay triangulation algorithm, and generating a fine surface mesh model of the fruit region and the pod region by mesh boundary optimization and a surface fitting algorithm; executing a boundary extraction algorithm on the surface mesh model, calculating the curvature and normal vector distribution of mesh vertices, identifying characteristic edges and contour lines of the mesh surface, and storing the extracted spatial contour data in a geometric feature data table according to the region type; Calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection of the major axis vector and the surface mesh model as a length parameter, determining the intersection of the minor axis vector and the surface mesh model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection of the normal vector and the surface mesh model as a height parameter; The surface area parameters of the fruit region and the pod region are calculated based on the surface mesh model, and the volume parameters of the fruit region and the pod region are obtained by performing depth dimension integration operation on the fruit region and the pod region areas of each layer of the CT tomography image sequence. The void ratio parameter is calculated according to the difference between the fruit region volume parameter and the pod region volume parameter, and the length parameter, width parameter, height parameter, surface area parameter, volume parameter and void ratio parameter are written into a peanut seed physiological characteristic database.

2. The method for measuring physiological parameters of peanut seeds based on CT images according to claim 1, characterized in that: The step of calculating the major axis vector and the minor axis vector of the fruit region and the pod region, determining the intersection of the major axis vector and the surface mesh model as a length parameter, determining the intersection of the minor axis vector and the surface mesh model as a width parameter, determining a normal vector according to a cross product result of the major axis vector and the minor axis vector, and determining the intersection of the normal vector and the surface mesh model as a height parameter comprises: Performing principal direction analysis on the contour data in the geometric feature data table, calculating the covariance matrix of the contour point set, obtaining the eigenvector corresponding to the maximum eigenvalue as the major axis vector and the eigenvector corresponding to the minimum eigenvalue as the minor axis vector through eigenvalue decomposition, and calculating the coordinates of the intersection of the major axis vector and the minor axis vector with the surface mesh model; A cross multiplication operation is performed based on the major axis vector and the minor axis vector to obtain a normal vector, the normal vector is normalized and then intersected with the surface mesh model, the Euclidean distance between the intersection points is calculated to obtain a height parameter, the major axis vector intersection distance is used as a length parameter, the minor axis vector intersection distance is used as a width parameter, and the parameter data is written into the size feature table.

3. The method for measuring physiological parameters of peanut seeds based on CT images according to claim 2, characterized in that: The surface area parameters of the fruit area and the pod area are calculated based on the curved mesh model, and the volume parameters of the fruit area and the pod area are obtained by performing a depth dimension integration operation on the fruit area and the pod area of ​​each layer of the CT tomography image sequence, including: Reading the triangular face data of the surface mesh model, calculating the area of ​​each triangular face and performing cumulative summation to obtain surface area parameters of the fruit region and the pod region respectively, performing connected domain labeling on the binary regions in the CT tomography image sequence, and calculating the pixel area of ​​the fruit region and the pod region in each layer of the image; The pixel area is multiplied by the pixel resolution of the CT device to obtain the actual area value, and a depth dimension integral function is constructed according to the CT scanning layer spacing parameters. An integral operation is performed on the fruit area and pod area of ​​each layer of the image to obtain the volume parameters of the fruit area and pod area, and the surface area parameters and volume parameters are written into the morphological feature table.

4. The method for measuring physiological parameters of peanut seeds based on CT images according to claim 3, characterized in that: The method comprises: calculating the gap ratio parameter according to the difference between the fruit area volume parameter and the pod area volume parameter, and writing the length parameter, width parameter, height parameter, surface area parameter, volume parameter and gap ratio parameter into a peanut seed physiological characteristic database, including: Reading the fruit region volume parameter and the pod region volume parameter in the morphological feature table, calculating the difference between the two volume parameters to obtain the cavity volume, dividing the cavity volume by the pod region volume parameter to calculate the void ratio parameter, and generating a void feature record according to the calculation result; Length parameters, width parameters, height parameters, surface area parameters, and volume parameters are extracted from the size feature table and the morphological feature table, combined with the gap ratio parameters in the gap feature record to construct a physiological feature data structure, and the physiological feature data is written into the parameter table of the peanut seed feature database.

5. A device for measuring physiological parameters of peanut seeds based on CT images, characterized in that: The device comprises: Point cloud data building blocks for: Collecting a CT tomographic image sequence of peanut kernels, obtaining a binary image of a fruit region and a pod region of the peanut kernels by an image segmentation algorithm, and converting the pixel coordinates of the binary image into a three-dimensional coordinate system of the physical world in combination with a CT device calibration parameter, specifically comprising: the CT device performs a multi-angle tomographic scan on the peanut kernels by a rotating X-ray tube, converting the scanning signal into a digital image sequence, extracting the edge contours of the fruit region and the pod region from the digital image sequence by using a threshold segmentation and a regional growth algorithm, and generating a binary image set containing labels of the fruit region and the pod region; reading the geometric calibration parameters and imaging calibration parameters of the CT device, performing a spatial coordinate system transformation on the pixel coordinates in the binary image set according to the calibration parameter matrix, generating fruit region coordinate data and pod region coordinate data in the three-dimensional coordinate system of the physical world, and writing the coordinate data into a spatial data buffer; Calculating the depth coordinate value in the three-dimensional coordinate system according to the CT scanning layer number information, generating spatial point cloud data including the fruit area and the pod area, specifically comprising: reading the layer spacing parameter and the scanning layer number information of the CT tomography image sequence, performing depth dimension mapping on the coordinate data in the spatial data buffer area, calculating the depth value of the coordinate point of the fruit area and the pod area in each layer of the image, combining the depth coordinate value with the original two-dimensional coordinate data to generate three-dimensional coordinate data; classifying and arranging the three-dimensional coordinate data according to the label information of the fruit area and the pod area, constructing a point cloud data structure including the spatial position and the regional attributes, writing the point cloud data into the three-dimensional data storage area, and establishing a point cloud data index table; The three-dimensional data analysis module is used to perform three-dimensional reconstruction processing on the spatial point cloud data, use a triangulated meshing algorithm to construct a surface mesh model of the fruit area and the pod area, and extract the spatial contour information of the fruit area and the pod area based on the surface mesh model, specifically including: reading the point cloud data in the three-dimensional data storage area, performing noise filtering and spatial smoothing processing on the point cloud data, constructing an initial mesh model based on the Delaunay triangulation algorithm, and generating a fine surface mesh model of the fruit area and the pod area through mesh boundary optimization and surface fitting algorithm; The model executes a boundary extraction algorithm, calculates the curvature and normal vector distribution of mesh vertices, identifies characteristic edges and contour lines of mesh surfaces, and stores the extracted spatial contour data in a geometric feature data table according to the region type; calculates the major axis vector and the minor axis vector of the fruit region and the pod region, determines the intersection point of the major axis vector with the surface mesh model as a length parameter, determines the intersection point of the minor axis vector with the surface mesh model as a width parameter, determines the normal vector according to the cross product result of the major axis vector and the minor axis vector, and determines the intersection point of the normal vector with the surface mesh model as a height parameter; The physiological parameter measurement module is used to calculate the surface area parameters of the fruit area and the pod area based on the surface mesh model, obtain the volume parameters of the fruit area and the pod area by performing depth dimension integration operation on the fruit area and the pod area area of ​​each layer of the CT tomography image sequence, calculate the gap ratio parameter according to the difference between the fruit area volume parameter and the pod area volume parameter, and write the length parameter, width parameter, height parameter, surface area parameter, volume parameter and gap ratio parameter into the peanut seed physiological characteristic database.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for measuring physiological parameters of peanut kernels based on CT images according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for measuring physiological parameters of peanut kernels based on CT images as described in any one of claims 1 to 4 are implemented.

Images