A method for screening peanut seeds

By collecting peanut images from different perspectives and calculating their fullness and sharpness, we realize automatic screening of high-quality peanut seeds, solving the problem of inefficiency in the existing technology, and improving screening efficiency and accuracy.

CN114937077BActive Publication Date: 2024-11-01DAMING XINXIN SEED IND CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210431082.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-22
Publication Date
2024-11-01
Estimated Expiration
2042-04-22

AI Technical Summary

Technical Problem

The prior art is less efficient when screening high-quality peanuts as seeds, rely on manual labor and inconsistent judgment criteria.

Method used

By collecting peanut images from different perspectives, extracting edge lines and germ lines, calculating the fullness and sharpness of peanuts, combining multiple image features to determine whether they are suitable as seeds, and automatically screening using a mechanized system.

Benefits of technology

It realizes efficient screening without manual intervention, and improves the efficiency, accuracy and consistency of peanut seed screening.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114937077B_ABST
    Figure CN114937077B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing technology, and in particular to a method for screening peanut seeds. The method combines relevant electronic equipment to perform image recognition to obtain images of peanuts to be detected at different viewing angles, performs data processing on the images of peanuts to be detected at different viewing angles, obtains the sharpness and fullness of the embryo tip of the peanut to be detected, and judges whether the peanut to be detected can be used as a seed according to the sharpness and fullness of the embryo tip of the peanut to be detected. The present invention combines relevant electronic equipment to perform graphic recognition and relevant data processing, thereby realizing the screening of peanut seeds and improving the screening efficiency of peanut seeds.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a peanut seed screening method. Background Art

[0002] The importance of seeds to agricultural production is self-evident. In order to increase the yield rate of peanut production and improve its planting efficiency, it is necessary to accurately and quickly select peanuts suitable for seeds from different peanut varieties. Mechanized seed screening can only screen out damaged seeds and cannot select high-quality peanuts as seeds. The existing high-quality peanut seed screening mainly relies on manual labor, which is not only slow, but also has different judgment standards. Summary of the invention

[0003] In order to solve the problem of low efficiency in the existing method of screening high-quality peanuts as peanut seeds, the object of the present invention is to provide a method for screening peanut seeds, and the technical scheme adopted is as follows:

[0004] The present invention provides a peanut seed screening method, which comprises the following steps:

[0005] According to the images of the peanuts to be detected at different viewing angles, the edge lines of the peanuts to be detected at each viewing angle are obtained; according to the edge lines of the peanuts to be detected at each viewing angle, the germ lines of the peanuts to be detected at each viewing angle are obtained, wherein the germ lines are the line connecting the point on the edge line of the peanuts to be detected that is farthest from the embryo tip and the embryo tip; a perpendicular line to the germ line is drawn through each pixel point on the germ line of the peanuts to be detected at each viewing angle, and each pixel point pair is constructed according to the intersection of each perpendicular line and the edge line of the peanuts to be detected;

[0006] Calculate the fullness of the peanut to be detected according to the distance between each pixel point and the center pixel point on the edge line of the peanut to be detected at each viewing angle;

[0007] Calculate the sharpness of the peanut embryo tip to be detected according to the angle of the peanut embryo tip to be detected at each viewing angle;

[0008] Whether the peanut to be tested can be used as a seed is determined based on the fullness of the peanut to be tested and the sharpness of the tip of the peanut embryo to be tested.

[0009] Preferably, obtaining the germ line of the peanut to be detected at each viewing angle according to the edge line of the peanut to be detected at each viewing angle comprises:

[0010] Perform convex hull detection on the edge line of the peanut to be detected at each viewing angle to obtain the convex hull edge of the peanut to be detected at each viewing angle; connect each pixel point on the convex hull edge of the peanut to be detected at each viewing angle with the center point of the convex hull, use the intersection of the connecting line and the edge line of the peanut to be detected as the pixel point corresponding to each pixel point on the convex hull edge, and calculate the distance between each pixel point on the convex hull edge of the peanut to be detected at each viewing angle and the corresponding pixel point;

[0011] The distance between each pixel point on the edge of the convex hull of the peanut to be detected at each viewing angle and its corresponding pixel point is fitted into a curve, wherein the minimum point of the curve is the point at the top of the embryo;

[0012] The distance between each pixel point on the edge line of the peanut to be detected and the point at the top of the germ under each viewing angle is calculated, the pixel point corresponding to the maximum distance under each viewing angle is obtained, and the line connecting the pixel points corresponding to the maximum distance is recorded as the germ line of the peanut to be detected under each viewing angle.

[0013] Preferably, the method of calculating the fullness of the peanuts to be detected according to the distance between each pixel point and the center pixel point on the edge line of the peanuts to be detected at each viewing angle includes:

[0014] According to the distance between the center pixel points of each pixel point pair on the edge line of the peanut to be detected at each viewing angle, the average value of the slope change rate of the distance between the center pixel points of each pixel point pair between two mutation point pairs of the peanut to be detected at each viewing angle is obtained;

[0015] According to the difference between the distance between the center pixel points of each pixel point pair on the edge line of the peanut to be detected at each viewing angle and the maximum distance, the average value of the difference between the distance between the center pixel points of each pixel point pair between two mutation point pairs of the peanut to be detected at each viewing angle and the maximum distance is obtained; the maximum distance is the maximum value of the distance between the center pixel points of each pixel point pair;

[0016] The fullness of the peanuts to be tested is calculated according to the mean of the slope change rate and the mean of the maximum distance difference.

[0017] Preferably, the method of obtaining the average value of the slope change rate of the distance between the pixel points in each pixel pair between two mutation point pairs of the peanut to be detected at each viewing angle according to the distance between the pixel points in each pixel pair on the edge line of the peanut to be detected at each viewing angle comprises:

[0018] According to the distance between each pixel point on the edge line of the peanut to be detected and the center pixel point at each viewing angle, a distance model corresponding to the peanut to be detected at each viewing angle is constructed;

[0019] The two pixel pairs with the largest slope change rate in the distance model corresponding to the peanuts to be detected at each viewing angle are taken as two mutation point pairs, and the average of the slope change rates of the distances of the pixel points in each pixel pair between the two mutation point pairs of the peanuts to be detected at each viewing angle is calculated.

[0020] Preferably, the fullness of the peanuts to be tested is calculated using the following formula:

[0021]

[0022] Among them, F is the fullness of the peanut to be tested, n is the number of viewing angles, B is the average value of the change rate of the slope of the curve between two mutation point pairs of the peanut to be tested at any viewing angle, T is the average value of the difference between the distance between the pixel points in each pixel point pair and the maximum distance between the two mutation point pairs of the peanut to be tested at this viewing angle, and S is the uniformity index of the peanut to be tested.

[0023] Preferably, the step of calculating the sharpness of the peanut embryo tip to be detected according to the angle of the peanut embryo tip to be detected at each viewing angle includes:

[0024] Connect two pixel points of the mutation point close to the embryo tip on the edge line of the peanut to be detected at each viewing angle to the embryo tip, respectively, to obtain the first angle of the peanut embryo tip to be detected at each viewing angle, and obtain the average value of the first angle of the peanut embryo tip to be detected according to the first angle of the peanut embryo tip to be detected at each viewing angle;

[0025] Draw a minimum circumscribed ellipse of the edge line of the peanut to be detected at each viewing angle, obtain two pixel points on the minimum circumscribed ellipse of the edge line of the peanut to be detected at each viewing angle that are tangent to the edge line and closest to the embryo tip, connect the two pixel points with the embryo tip respectively, obtain a second angle of the peanut embryo tip to be detected at each viewing angle, and obtain a second angle average of the peanut embryo tip to be detected according to the second angle of the peanut embryo tip to be detected at each viewing angle;

[0026] The sharpness of the peanut embryo tip to be detected is obtained according to the first angle average of the peanut embryo tip to be detected and the second angle average of the peanut embryo tip to be detected.

[0027] Preferably, judging whether the peanuts to be tested can be used as seeds according to the fullness of the peanuts to be tested and the sharpness of the embryo tips of the peanuts to be tested comprises:

[0028] Calculate the fitness index of the peanuts to be tested according to the fullness of the peanuts to be tested and the sharpness of the embryo tips of the peanuts to be tested;

[0029] The fitness index of the peanuts to be tested is compared with the set threshold. When the fitness index of the peanuts to be tested is greater than the set threshold, it is determined that the peanuts to be tested can be used as seeds; when the fitness index of the peanuts to be tested is less than or equal to the set threshold, it is determined that the peanuts to be tested cannot be used as seeds.

[0030] The present invention has the following beneficial effects: the present invention obtains the edge lines of the peanuts to be detected at different viewing angles, calculates the fullness of the peanuts to be detected and the sharpness of the embryo tip according to the edge line features of the peanuts to be detected at each viewing angle, and judges whether the peanuts to be detected can be used as seeds according to the fullness of the peanuts to be detected and the sharpness of the embryo tip. The present invention no longer needs to rely on manual screening of the peanuts to be detected, and solves the problem of low efficiency of the existing manual screening of the peanuts to be detected. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A flowchart of a peanut seed screening method provided by one embodiment of the present invention;

[0033] Figure 2 A schematic diagram of the peanut mutation point pairs to be detected;

[0034] Figure 3 Schematic diagram of the tip angle of the peanut embryo to be tested. DETAILED DESCRIPTION

[0035] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, a peanut seed screening method proposed in accordance with the present invention is described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0036] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0037] The specific scheme of the peanut seed screening method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0038] A method for screening peanut seeds

[0039] The existing method of screening high-quality peanuts as peanut seeds has the problem of low efficiency. In order to solve the above problem, this embodiment proposes a method for screening peanut seeds, such as Figure 1 As shown, a peanut seed screening method of this embodiment includes the following steps:

[0040] Step S1, according to the images of the peanuts to be detected at different viewing angles, the edge lines of the peanuts to be detected at each viewing angle are obtained; according to the edge lines of the peanuts to be detected at each viewing angle, the germ lines of the peanuts to be detected at each viewing angle are obtained, and the germ lines are the line connecting the point on the edge line of the peanuts to be detected that is farthest from the embryo tip and the embryo tip; perpendicular lines to the germ lines are drawn through each pixel point on the germ line of the peanuts to be detected at each viewing angle, and each pixel point pair is constructed according to the intersection of each perpendicular line and the edge line of the peanuts to be detected.

[0041] Peanuts are three-dimensional. In this embodiment, three-dimensional features are judged by two-dimensional image features. Therefore, multiple images are used for judgment to avoid inaccurate judgment results caused by obtaining peanut images from only one perspective. In this embodiment, n cameras are arranged side by side above the roller conveyor, n≥2, and the value of n is set according to actual needs. The peanuts to be detected are placed on the conveyor, and the camera is used to collect the image of the peanuts to be detected directly below. The peanuts will roll with the rotation of the roller, and the field of view of the camera is unchanged. Due to the rolling of the peanuts, different cameras can collect peanut images under different perspectives. In this embodiment, the time when each camera starts to collect is different. The time when each camera starts to collect is set to the time when the peanuts begin to appear below. The shooting interval is set according to the speed of the conveyor. When the pth peanut rolls over, the pth image taken by all cameras is the image of the peanut. In this embodiment, the peanuts to be detected are shelled peanuts. For a peanut to be detected, n images are obtained. The canny operator is used to perform edge detection on the n images of the peanut to be detected to obtain the edge line corresponding to each image of the peanut to be detected.

[0042] Convex hull detection is performed on each edge line of the peanut to be detected to obtain the corresponding convex hull edge, and each pixel point on the convex hull edge line is connected to the center point of the convex hull. The intersection of the connection line and the edge line of the peanut to be detected is used as the pixel point corresponding to each pixel point on the convex hull, and the distance between each pixel point on the convex hull edge and its corresponding pixel point is calculated. Taking any pixel point on the convex hull edge as the starting point, each pixel point on the convex hull edge as the horizontal coordinate, and the distance between each pixel point on the convex hull edge and its corresponding pixel point as the vertical coordinate, a curve is fitted, and the minimum point of the curve is the point at the top of the embryo. The distance between each pixel point on the edge line of the peanut image to be detected and the point at the top of the embryo is calculated, and the pixel point corresponding to the maximum distance is the root point. In this embodiment, the line connecting the point at the top of the embryo and the root point is recorded as the embryo line, and the distance of this line segment is l max .

[0043] Draw a perpendicular line through a pixel on the germ line, and record the two intersection points with the edge line of the peanut to be detected as a pixel pair. Traverse all the pixels on the germ line to draw a perpendicular line to the germ line, and get k pixel pairs. These pixel pairs constitute a sequence M = (m1, m2, m3, ..., m k), m1 is the first pixel pair, m2 is the second pixel pair, m3 is the third pixel pair, m k is the kth pixel pair.

[0044] Step S2, calculating the fullness of the peanut to be detected according to the distance between each pixel point and the center pixel point on the edge line of the peanut to be detected at each viewing angle.

[0045] For peanut seeds, high-quality peanut seeds should be relatively full, and the top of its embryo tip should be relatively sharp. This embodiment determines whether the peanut to be tested can be used as a seed based on the fullness of the peanut to be tested and the sharpness of the embryo tip.

[0046] For one of the n edge line images of peanuts to be detected, this embodiment calculates the distance between two pixel points in k pixel pairs, and these distances constitute a sequence L=(l1,l2,l3,…,l k ), l1 is the distance between the two pixels in the first pixel pair, l2 is the distance between the two pixels in the second pixel pair, l3 is the distance between the two pixels in the third pixel pair, l k is the distance between two pixels in the kth pixel pair. A distance model is constructed based on the distance between two pixels in each pixel pair, wherein the embryo tip of the peanut to be tested is used as the starting point of the distance model, the number of each pixel pair is used as the horizontal coordinate of the distance model, and the distance between the pixels in each pixel pair is used as the vertical coordinate of the distance model. The two pixel pairs with the largest slope change in the distance model are recorded as mutation point pairs, such as Figure 2 As shown, Figure 2 1 is the first mutation point, denoted as a, and the other pixel in the pixel pair is Figure 2 As shown in 2, Figure 2 3 is the second mutation point, denoted as b, and the pixel points that form a pixel pair with this pixel point are as follows Figure 2 As shown in 4. Obtain the difference in the slopes of adjacent pixel pairs between two mutation point pairs in the distance model, and calculate the mean of the rate of change of the slope of the curve between the two mutation point pairs. The specific calculation formula is:

[0047]

[0048] Among them, B is the mean rate of change of the slope of the curve between two mutation point pairs, f′(x) is the slope of the x-th pixel pair between the two mutation point pairs in the distance model, and f(x-1) is the slope of the x-1th pixel pair in the distance model. The x-1th pixel pair is adjacent to the xth pixel pair.

[0049] This embodiment calculates the distance between two pixels in each pixel pair and the longest distance l max The difference between each pixel pair and the longest distance lmax The difference between max -c(x), where g(x) is the distance between the two pixels in the x-th pixel pair and the longest distance l max The difference between the two pixels in the x-th pixel pair is c(x), and c(x) is the distance between the two pixels in the x-th pixel pair. max The difference between the distance of the pixel points in each pixel point pair between the two mutation point pairs and the maximum distance l is calculated. max The specific calculation formula is:

[0050]

[0051] Where T is the average value of the difference between the distance between the pixels in each pixel pair and the maximum distance between the two mutation point pairs.

[0052] In order to calculate the fullness of the peanut to be detected, this embodiment judges the fullness of the peanut based on the area difference of the contour of the peanut to be detected in the n images collected, and calculates the area of ​​the contour of the peanut to be detected in these n images, which are recorded as S1, S2, S3, ..., S n , according to the contour area of ​​the peanut to be detected in these n images, calculate the uniformity index of the peanut to be detected. The smaller the uniformity index, the more uniform the area of ​​the peanut images collected from all sides. The specific calculation formula is:

[0053]

[0054] Among them, S is the uniformity index of the peanut to be detected, S1 is the area of ​​the contour in the first edge image of the peanut to be detected, S2 is the area of ​​the contour in the second edge image of the peanut to be detected, S3 is the area of ​​the contour in the third edge image of the peanut to be detected, and S n is the area of ​​the contour in the nth edge image of the peanut to be detected.

[0055] This embodiment obtains the fullness of the peanuts to be detected based on the mean of the rate of change of the slope of the curve between two mutation point pairs on each edge image of the peanuts to be detected and the mean of the difference between the distance between the pixels in each pixel pair between the two mutation point pairs and the maximum distance. The smaller the mean of the rate of change of the slope of the curve, the fuller the peanut; the smaller the mean of the difference between the distance between the pixels in each pixel pair and the maximum distance, the fuller the peanut; the smaller the area of ​​the contour image, the more likely the peanut is to be suitable as a seed. The specific formula for calculating the fullness of the peanuts to be detected is:

[0056]

[0057] Among them, F is the fullness of the peanuts to be tested. F is normalized and recorded as F0, so that the fullness of the peanuts to be tested is between 0 and 1. The fullness F0 after normalization is used as the final fullness of the peanuts to be tested.

[0058] At this point, the fullness of the peanuts to be tested is obtained.

[0059] Step S3, calculating the sharpness of the peanut embryo tip to be detected according to the angle of the peanut embryo tip to be detected at each viewing angle.

[0060] This embodiment obtains the sharpness of the peanut embryo tip based on the edge image of the peanut to be detected. The specific method is as follows:

[0061] For each edge image of the peanut to be tested, first, a pair of mutation points close to the embryo tip is selected, such as Figure 2 1. Figure 2 As shown in 2, two pixel points in the mutation point pair are connected with the point at the top of the embryo tip to obtain an angle α1, as shown in Figure 2 As shown in the figure, the mean value α of the angle on these n edge images is calculated. The smaller the mean value of the angle, the sharper the embryo tip is, and the more likely it is to be suitable as a seed. Then make the minimum circumscribed ellipse of the edge image of each peanut to be tested, obtain the two points closest to the embryo tip where the embryo tip and the edge image are tangent to the minimum circumscribed ellipse, and connect these two points with the embryo tip to obtain an angle β1, as shown in the figure. Figure 3 As shown, the mean β of the angle on these n edge images is calculated. The smaller the mean angle is, the sharper the embryo tip is, and the more likely it is to be suitable as a seed.

[0062] In this embodiment, the sharpness of the tip of the peanut embryo to be detected is calculated according to the angle mean α and the angle mean β. The specific formula is:

[0063] θ=αβ

[0064] Wherein, θ is the sharpness of the tip of the peanut embryo to be detected. In this embodiment, θ is normalized, and the sharpness after normalization is recorded as θ0. The smaller θ0 is, the sharper the tip of the peanut embryo is.

[0065] At this point, the sharpness of the tip of the peanut embryo to be tested is obtained.

[0066] Step S4, judging whether the peanut to be tested can be used as a seed according to the fullness of the peanut to be tested and the sharpness of the tip of the peanut embryo to be tested.

[0067] In this embodiment, the suitability index of the peanut to be tested is obtained according to the fullness of the peanut to be tested and the sharpness of the embryo tip. According to the suitability index of the peanut to be tested, it is determined whether the peanut to be tested can be used as a seed.

[0068] Specifically, the suitability index of the peanuts to be tested is calculated using the following formula:

[0069]

[0070] Wherein, P is the fitness index of the peanut to be tested, and e is a natural constant. The fuller the peanut to be tested is, and the sharper the embryo tip is, the greater the fitness index is, and the more suitable the peanut is as a seed.

[0071] In this embodiment, the suitability index threshold is set to 0.7. When the suitability index of the peanut to be detected is greater than the threshold, it is determined that the peanut can be used as a seed. In specific applications, the suitability index threshold is set according to actual needs. In this embodiment, a mechanical gripper is used to select peanuts that can be used as seeds for subsequent planting.

[0072] This embodiment obtains the edge lines of the peanuts to be detected at different viewing angles, calculates the fullness of the peanuts to be detected and the sharpness of the embryo tip according to the edge line features of the peanuts to be detected at each viewing angle, and determines whether the peanuts to be detected can be used as seeds according to the fullness of the peanuts to be detected and the sharpness of the embryo tip. This embodiment does not need to rely on manual screening of the peanuts to be detected, and solves the problem of low efficiency of the existing manual screening of the peanuts to be detected.

[0073] It should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for screening peanut seeds, It is characterized in that The method comprises the following steps: According to the images of the peanuts to be detected at different viewing angles, the edge lines of the peanuts to be detected at each viewing angle are obtained; according to the edge lines of the peanuts to be detected at each viewing angle, the germ lines of the peanuts to be detected at each viewing angle are obtained, wherein the germ lines are the line connecting the point on the edge line of the peanuts to be detected that is farthest from the embryo tip and the embryo tip; a perpendicular line to the germ line is drawn through each pixel point on the germ line of the peanuts to be detected at each viewing angle, and each pixel point pair is constructed according to the intersection of each perpendicular line and the edge line of the peanuts to be detected; Calculate the fullness of the peanut to be detected according to the distance between each pixel point and the center pixel point on the edge line of the peanut to be detected at each viewing angle; Calculate the sharpness of the peanut embryo tip to be detected according to the angle of the peanut embryo tip to be detected at each viewing angle; Judging whether the peanuts to be tested can be used as seeds according to the fullness of the peanuts to be tested and the sharpness of the embryo tips of the peanuts to be tested; The method of calculating the fullness of the peanuts to be detected according to the distance between each pixel point and the center pixel point on the edge line of the peanuts to be detected at each viewing angle includes: According to the distance between the center pixel points of each pixel point pair on the edge line of the peanut to be detected at each viewing angle, the average value of the slope change rate of the distance between the center pixel points of each pixel point pair between two mutation point pairs of the peanut to be detected at each viewing angle is obtained; According to the difference between the distance between the center pixel points of each pixel point pair on the edge line of the peanut to be detected at each viewing angle and the maximum distance, the average value of the difference between the distance between the center pixel points of each pixel point pair between two mutation point pairs of the peanut to be detected at each viewing angle and the maximum distance is obtained; the maximum distance is the maximum value of the distance between the center pixel points of each pixel point pair; Calculate the fullness of the peanuts to be tested according to the mean of the slope change rate and the mean of the maximum distance difference; The method of obtaining the average value of the slope change rate of the distance between the pixel points in each pixel pair between two mutation point pairs of the peanut to be detected at each viewing angle according to the distance between the pixel points in each pixel pair on the edge line of the peanut to be detected at each viewing angle comprises: According to the distance between each pixel point on the edge line of the peanut to be detected and the center pixel point at each viewing angle, a distance model corresponding to the peanut to be detected at each viewing angle is constructed; The two pixel point pairs with the largest slope change rate in the distance model corresponding to the peanuts to be detected at each viewing angle are taken as two mutation point pairs, and the average of the slope change rate of the distance between the pixel points in each pixel point pair between the two mutation point pairs of the peanuts to be detected at each viewing angle is calculated; The fullness of the peanuts to be tested is calculated using the following formula: Among them, F is the fullness of the peanut to be tested, n is the number of viewing angles, B is the average value of the change rate of the slope of the curve between two mutation point pairs of the peanut to be tested at any viewing angle, T is the average value of the difference between the distance between the pixel points in each pixel point pair and the maximum distance between the two mutation point pairs of the peanut to be tested at this viewing angle, and S is the uniformity index of the peanut to be tested; The step of calculating the sharpness of the peanut embryo tip to be detected according to the angle of the peanut embryo tip to be detected at each viewing angle includes: Connect two pixel points of the mutation point close to the embryo tip on the edge line of the peanut to be detected at each viewing angle to the embryo tip, respectively, to obtain the first angle of the peanut embryo tip to be detected at each viewing angle, and obtain the average value of the first angle of the peanut embryo tip to be detected according to the first angle of the peanut embryo tip to be detected at each viewing angle; Draw a minimum circumscribed ellipse of the edge line of the peanut to be detected at each viewing angle, obtain two pixel points on the minimum circumscribed ellipse of the edge line of the peanut to be detected at each viewing angle that are tangent to the edge line and closest to the embryo tip, connect the two pixel points with the embryo tip respectively, obtain a second angle of the peanut embryo tip to be detected at each viewing angle, and obtain a second angle average of the peanut embryo tip to be detected according to the second angle of the peanut embryo tip to be detected at each viewing angle; Obtaining the sharpness of the peanut embryo tip to be detected according to the first angle average of the peanut embryo tip to be detected and the second angle average of the peanut embryo tip to be detected; The distance model corresponding to the peanut to be detected at each viewing angle is constructed according to the distance between each pixel point and the center pixel point on the edge line of the peanut to be detected at each viewing angle, including: For the image of the peanut to be detected at any viewing angle: a distance model is constructed based on the distance between two pixels in each pixel pair, wherein the embryo tip of the peanut to be detected is used as the starting point of the distance model, the number of each pixel pair is used as the abscissa of the distance model, and the distance between the pixels in each pixel pair is used as the ordinate of the distance model, to construct the distance model corresponding to the peanut to be detected at this viewing angle; The uniformity index of the peanuts to be tested is calculated using the following formula: Among them, S is the uniformity index of the peanuts to be tested, S 1 is the area of ​​the contour in the first edge image of the peanut to be detected, S 2 is the area of ​​the contour of the second edge image of the peanut to be detected, S 3 is the area of ​​the contour in the third edge image of the peanut to be detected, S n is the area of ​​the contour in the nth edge image of the peanut to be detected, and n is the number of edge images of the peanut to be detected.

2. A peanut seed screening method according to claim 1, It is characterized in that The step of obtaining the germ line of the peanut to be detected at each viewing angle according to the edge line of the peanut to be detected at each viewing angle comprises: Perform convex hull detection on the edge line of the peanut to be detected at each viewing angle to obtain the convex hull edge of the peanut to be detected at each viewing angle; connect each pixel point on the convex hull edge of the peanut to be detected at each viewing angle with the center point of the convex hull, use the intersection of the connecting line and the edge line of the peanut to be detected as the pixel point corresponding to each pixel point on the convex hull edge, and calculate the distance between each pixel point on the convex hull edge of the peanut to be detected at each viewing angle and the corresponding pixel point; The distance between each pixel point on the edge of the convex hull of the peanut to be detected at each viewing angle and its corresponding pixel point is fitted into a curve, and the minimum point of the curve is the point of the embryo tip; The distance between each pixel point on the edge line of the peanut to be detected and the embryo tip point at each viewing angle is calculated, the pixel point corresponding to the maximum distance at each viewing angle is obtained, and the line connecting the pixel points corresponding to the maximum distance is recorded as the embryo line of the peanut to be detected at each viewing angle.

3. A peanut seed screening method according to claim 1, It is characterized in that The step of judging whether the peanut to be tested can be used as a seed according to the fullness of the peanut to be tested and the sharpness of the tip of the peanut embryo to be tested comprises: Calculate the fitness index of the peanuts to be tested according to the fullness of the peanuts to be tested and the sharpness of the embryo tips of the peanuts to be tested; The fitness index of the peanuts to be tested is compared with the set threshold. When the fitness index of the peanuts to be tested is greater than the set threshold, it is determined that the peanuts to be tested can be used as seeds; when the fitness index of the peanuts to be tested is less than or equal to the set threshold, it is determined that the peanuts to be tested cannot be used as seeds.