Seedling stage crop strip-shaped composite planting seedling strip identification method
The field point cloud data is obtained through lidar, the target area of the seedling belt is identified and the navigation route is calculated, which solves the problem of unstable identification of traditional machine vision under the changes in the field environment, and realizes automatic identification and mechanical navigation of soybean and corn seedling belts.
Patent Information
- Application Number
- CN202411080100.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-12-06
AI Technical Summary
The unstable identification of seedlings under field environment changes (such as light and shadow) in traditional machine vision leads to waste of resources and increased costs, which cannot meet market demand.
Lidar is used to obtain point cloud data, and through coordinate conversion and filter noise removal, determine the target area of the seedling belt, extract the width and target points of the seedling belt, distinguish different seedling belts, calculate navigation lines, and plan agricultural machinery paths.
Automatic identification of soybean and corn seedling belts is realized, providing accurate independent operation navigation guidance for belt-shaped composite planting machinery, and improving identification accuracy and efficiency.
Smart Images

Figure CN119073176A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of intelligent agriculture, and in particular relates to a method for identifying seedling belts of strip-shaped composite planting of crops at the seedling stage. Background Art
[0002] Corn and soybeans are both food and economic crops, but they are traditionally produced using manual labor or semi-automatic manual machinery. Currently, labor costs are rising, and existing production levels can no longer meet market demand.
[0003] In the existing technology, machine vision is used to identify seedling strips for early single crop planting. However, due to its own performance limitations, machine vision is easily affected by changes in the field environment, such as light, shadow, etc. The detection results are unstable, which will cause waste of resources and increase costs.
[0004] Therefore, it is urgent to develop a new method for identifying seedling strips of crop strip composite planting in the seedling stage to solve the above problems. Summary of the invention
[0005] The purpose of the invention is to provide a method for identifying seedling strips of strip-shaped composite planting of crops at the seedling stage.
[0006] In order to solve the above technical problems, the present invention provides a method for identifying seedling belts of strip-shaped composite planting of crops in the seedling stage, which includes: acquiring point cloud data of an identification area; determining a seedling belt target area in the identification area to extract point cloud data of the seedling belt target area; extracting various types of seedling belts from the point cloud data of the seedling belt target area, and extracting target points of various types of seedling belts; respectively distinguishing the target points corresponding to various types of seedling belts to obtain seedling belt navigation lines.
[0007] Specifically, the method for obtaining point cloud data of the identification area includes: obtaining point cloud data of the identification area through a laser radar; traversing each point in the point cloud data of the identification area, converting the point cloud data in the radar coordinate system to the world coordinate system through a coordinate transformation matrix, and segmenting ground point cloud data and non-ground point cloud data; filtering out noise points in the non-ground point cloud data.
[0008] Specifically, the method for determining the seedling belt target area in the identification area to extract point cloud data of the seedling belt target area includes: setting the length and width of the seedling belt target area, the length of the seedling belt target area is along the x-axis direction of the world coordinate system, and the width of the seedling belt target area is along the y-axis direction of the world coordinate system, so that the center point of the seedling belt target area is located on the x-axis of the world coordinate system; traversing each point in the non-ground point cloud data, when the projection point of any point on the coordinate plane of the world coordinate system is located in the seedling belt target area, the point cloud data corresponding to the point is the point cloud data of the seedling belt target area.
[0009] Specifically, the method for extracting various types of seedling belts from the point cloud data of the seedling belt target area and extracting the target points of each type of seedling belt includes: determining an initial horizontal target area in the seedling belt target area; extracting point cloud data within the initial horizontal target area to determine the width of the seedling belt within the initial horizontal target area; selecting the seedling belt target area according to the width of the seedling belt to extract the first type of seedling belt and the second type of seedling belt; comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with a set belt number threshold, and when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, the target points of each type of seedling belt are extracted.
[0010] Specifically, the method for determining the initial horizontal target area in the seedling belt target area includes: setting the length and width of the initial horizontal target area in the seedling belt target area, the length of the initial horizontal target area is along the x-axis direction of the world coordinate system, and the width of the initial horizontal target area is along the y-axis direction of the world coordinate system, so that the center point of the initial horizontal target area is located on the x-axis of the world coordinate system.
[0011] Specifically, the method of extracting point cloud data within the initial horizontal target area to determine the width of the seedling bandwidth within the initial horizontal target area includes: traversing each point in the non-ground point cloud data in the initial horizontal target area, when the projection point of any point on the coordinate plane of the world coordinate system is located within the initial horizontal target area, the point cloud data corresponding to the point is the point cloud data of the initial horizontal target area; clustering the point cloud data within the initial horizontal target area, and determining the width of the first type of seedling bandwidth and the width of the second type of seedling bandwidth within the initial horizontal target area based on the maximum distance between the point cloud data within the initial horizontal target area in each cluster.
[0012] Specifically, the method of selecting the target area of the seedling belt according to the width of the seedling belt to extract the first type of seedling belt and the second type of seedling belt includes: in the target area of the seedling belt, setting the corresponding frame selection length and frame selection width according to the width of the first type of seedling belt and the width of the second type of seedling belt, respectively, to extract the first type of seedling belt and the second type of seedling belt, respectively.
[0013] Specifically, the method of comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with a set belt number threshold, and when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, extracting the target points of each type of seedling belt includes: setting the belt number threshold; comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with the belt number threshold; when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is less than the belt number threshold, redetermining the seedling belt target area, or when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, extracting the target points of each type of seedling belt.
[0014] Specifically, the method for distinguishing the target points corresponding to each type of seedling belt to obtain the seedling belt navigation line includes: determining the position of each type of seedling belt according to the target points of each type of seedling belt; calculating the distances between adjacent seedling belts in turn to set the row spacing threshold; comparing the distances between adjacent seedling belts with the row spacing threshold to distinguish the various types of seedling belts and the target points corresponding to each type of seedling belt; clustering and straight line fitting of the target points of each type of seedling belt to obtain the belt lines corresponding to each type of seedling belt; obtaining the navigation lines corresponding to each type of seedling belt according to the belt lines corresponding to each type of seedling belt; obtaining the seedling belt navigation lines according to the navigation lines corresponding to each type of seedling belt.
[0015] Specifically, the method for identifying seedling belts of strip-shaped composite planting of crops in the seedling stage also includes: planning the path of agricultural machinery through the seedling belt navigation line.
[0016] The beneficial effect of the present invention is that the present invention can realize automatic identification of soybean seedling strips and corn seedling strips in the seedling stage, and provide accurate navigation guidance for the autonomous operation of soybean and corn strip composite planting machinery.
[0017] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practicing the present invention.
[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0020] Figure 1 It is a work flow chart of the method for identifying the seedling belt of the crop strip composite planting in the seedling stage of the present invention;
[0021] Figure 2 is a workflow diagram of point cloud data processing of the present invention;
[0022] Figure 3 It is a schematic diagram of the preprocessing of point cloud data of the seedling belt target area of the present invention;
[0023] Figure 4 is a schematic diagram of extracting an initial horizontal target area according to the present invention;
[0024] Figure 5 It is a schematic diagram of extracting various seedling strips of the present invention;
[0025] Figure 6 is a schematic diagram of extracting target points of various seedling belts according to the present invention;
[0026] Figure 7 It is a schematic diagram of extracting the guide line of the seedling belt of the present invention. DETAILED DESCRIPTION
[0027] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0028] Embodiment 1, in this embodiment, as Figures 1 to 7 As shown, the present embodiment provides a method for identifying seedling belts of strip-shaped composite planting of crops in the seedling stage, which includes: acquiring point cloud data of an identification area; determining a seedling belt target area in the identification area to extract point cloud data of the seedling belt target area; extracting various types of seedling belts from the point cloud data of the seedling belt target area, and extracting target points of various types of seedling belts; respectively distinguishing the target points corresponding to various types of seedling belts to obtain seedling belt navigation lines.
[0029] Specifically, the various types of seedling strips include at least two types of seedling strips, soybean seedling strips and corn seedling strips.
[0030] In this embodiment, this embodiment can realize automatic identification of soybean seedling strips and corn seedling strips in the seedling stage, and provide accurate navigation guidance for the autonomous operation of soybean-corn strip composite planting machinery.
[0031] In this embodiment, the method for obtaining point cloud data of the identification area includes: obtaining point cloud data of the identification area through a laser radar; traversing each point in the point cloud data of the identification area, converting the point cloud data in the radar coordinate system to the world coordinate system through a coordinate transformation matrix, and segmenting ground point cloud data and non-ground point cloud data; filtering out noise points in the non-ground point cloud data.
[0032] Specifically, a seedling strip collection system for strip-shaped composite planting of soybeans and corn in the seedling stage was built, and the point cloud data of strip-shaped composite planting of soybeans and corn in the seedling stage was acquired by lidar, and the point cloud data was preprocessed by coordinate transformation, filtering and noise reduction.
[0033] Specifically, traverse each point in the point cloud data and use the coordinate transformation matrix to transform the point cloud data in the radar coordinate system XYZ to the world coordinate system X w Y w Z w The point cloud coordinate transformation formula is as follows:
[0034]
[0035] Where θ0 is the tilt angle of the laser radar (unit: °), x B ,y B 、z B is the point cloud coordinate in the radar coordinate system (unit: m), x A ,y A 、z A is the point cloud coordinate in the world coordinate system (unit: m).
[0036] Specifically, the RANSAC algorithm is used to segment the ground point cloud data and the non-ground point cloud data.
[0037] Specifically, in the preprocessing of point cloud data, statistical filtering is used to remove noise from non-ground point cloud data.
[0038] In this embodiment, the method for determining the seedling belt target area in the identification area to extract point cloud data of the seedling belt target area includes: setting the length and width of the seedling belt target area, the length of the seedling belt target area is along the x-axis direction of the world coordinate system, and the width of the seedling belt target area is along the y-axis direction of the world coordinate system, so that the center point of the seedling belt target area is located on the x-axis of the world coordinate system; traversing each point in the non-ground point cloud data, when the projection point of any point on the coordinate plane of the world coordinate system is located in the seedling belt target area, the point cloud data corresponding to the point is the point cloud data of the seedling belt target area.
[0039] Specifically, according to the soybean-corn strip composite planting pattern, the seedling belt target area is determined, and the point cloud data of the seedling belt target area is extracted using the vertical projection method.
[0040] Determination of seedling belt target area: The seedling belt target area is a region located in the world coordinate system X w Y w The length L of the target area of the seedling belt is set to 20c1, the width W is set to 10w1, and the length L is along the X w Axis direction, length W along the Y w The center point of the target area of the seedling belt is located in the X-axis direction. w Axis, which is on the Y w The coordinate on the axis is 15c1, c1 is the plant spacing of the corn seedling strip in the soybean-corn strip composite planting pattern, and w1 is the row spacing of the soybean seedling strip in the soybean-corn strip composite planting pattern.
[0041] In one embodiment, L = 2m, W = 3m, c1 = 0.1m, w1 = 0.3m.
[0042] In this embodiment, the method of extracting various types of seedling belts from the point cloud data of the seedling belt target area and extracting the target points of each type of seedling belt includes: determining an initial horizontal target area in the seedling belt target area; extracting point cloud data within the initial horizontal target area to determine the width of the seedling belt within the initial horizontal target area; selecting the seedling belt target area according to the width of the seedling belt to extract the first type of seedling belt and the second type of seedling belt; comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with a set belt number threshold, and when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, the target points of each type of seedling belt are extracted.
[0043] Specifically, a wide bounding box selection algorithm is used to extract soybean seedling strips and corn seedling strips.
[0044] In this embodiment, the method for determining the initial horizontal target area in the seedling belt target area includes: setting the length and width of the initial horizontal target area in the seedling belt target area, the length of the initial horizontal target area is along the x-axis direction of the world coordinate system, and the width of the initial horizontal target area is along the y-axis direction of the world coordinate system, so that the center point of the initial horizontal target area is located on the x-axis of the world coordinate system.
[0045] Determine the initial horizontal target area: In the seedling belt target area, set the width ΔW of the target area to 10w1, the length ΔL to 1.2c1, and the width ΔW along the world coordinate system Y w Axis direction, length ΔL along the world coordinate system X w The center point of the initial horizontal target area is located in the world coordinate system X w axis, which is in the world coordinate system Y w The coordinate value on the axis is 7c1, where ΔL is the spacing of corn seedling strips in the soybean-corn strip composite planting pattern, and the results are as follows Figure 4 shown.
[0046] In one embodiment, ΔW=3m, ΔL=0.12m.
[0047] In this embodiment, the method of extracting point cloud data within the initial horizontal target area to determine the width of the seedling bandwidth within the initial horizontal target area includes: traversing each point in the non-ground point cloud data in the initial horizontal target area, when the projection point of any point on the coordinate plane of the world coordinate system is located within the initial horizontal target area, the point cloud data corresponding to the point is the point cloud data of the initial horizontal target area; clustering the point cloud data within the initial horizontal target area, and determining the width of the first type of seedling bandwidth and the width of the second type of seedling bandwidth within the initial horizontal target area based on the maximum distance between the point cloud data within the initial horizontal target area in each cluster.
[0048] Determine the width of the seedling bandwidth within the initial horizontal target area: Use the k-means clustering algorithm to cluster the point cloud data within the initial horizontal target area, and determine the width of the soybean seedling bandwidth and the width of the corn seedling bandwidth within the initial horizontal target area based on the maximum distance between the point cloud data within the initial horizontal target area within each cluster.
[0049] In this embodiment, the method of selecting the target area of the seedling belt according to the width of the seedling belt to extract the first type of seedling belt and the second type of seedling belt includes: in the target area of the seedling belt, setting the corresponding frame selection length and frame selection width according to the width of the first type of seedling belt and the width of the second type of seedling belt, respectively, to extract the first type of seedling belt and the second type of seedling belt, respectively.
[0050] Specifically, within the seedling target area, a wide bounding box selection algorithm is used to enlarge the length of the initial horizontal target area along the world coordinate system X w The frame selection is performed in the axial direction, the frame selection length L1 is set to 20c1, and the width is the width of the seedling belt in the initial horizontal target area; according to the frame selection results, the soybean seedling belt and corn seedling belt located in the rectangular frame are extracted, and the results are as follows Figure 5 shown.
[0051] In one embodiment, L1 = 2m.
[0052] In this embodiment, the method of comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with the set belt number threshold, and when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, extracting the target points of each type of seedling belt includes: setting the belt number threshold; comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with the belt number threshold; when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is less than the belt number threshold, redetermining the seedling belt target area, or when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, extracting the target points of each type of seedling belt.
[0053] Target point extraction: According to the soybean-corn strip composite planting pattern, the strip number threshold is set. When the number of extracted seedling strips is greater than the strip number threshold, the target points of the soybean seedling strip and corn seedling strip are extracted. The results are as follows: Figure 6 shown.
[0054] In one embodiment, the band number threshold is 7.
[0055] In this embodiment, the method for distinguishing the target points corresponding to each type of seedling belt to obtain the seedling belt navigation line includes: determining the position of each type of seedling belt according to the target points of each type of seedling belt; calculating the distances between adjacent seedling belts in turn to set the row spacing threshold; comparing the distances between adjacent seedling belts with the row spacing threshold to distinguish the various types of seedling belts and the target points corresponding to each type of seedling belt; clustering and straight line fitting of the target points of each type of seedling belt to obtain the belt lines corresponding to each type of seedling belt; obtaining the navigation lines corresponding to each type of seedling belt according to the belt lines corresponding to each type of seedling belt; obtaining the seedling belt navigation lines according to the navigation lines corresponding to each type of seedling belt.
[0056] Specifically, according to the obtained target points of the soybean seedling belt and the corn seedling belt, the target points of the soybean seedling belt and the corn seedling belt are calculated in the world coordinate system Y w The mean value on the axis is used as the mean value of the soybean and corn seedling strips in the world coordinate system Y w The position of the soybean seedling belt and the corn seedling belt are obtained along the world coordinate system Y w The axes are numbered from small to large: 1, 2, 3, ..., n, and the distance D(i, i+1) between adjacent soybean seedling strips and corn seedling strips is calculated, and the maximum distance is used as the row spacing threshold; the calculation formula of the row spacing threshold T is as follows:
[0057] In the formula, y i and i+1 Represents the position coordinates of continuous soybean seedling strips and corn seedling strips.
[0058] Specifically, traverse each seedling belt in the world coordinate system Y w For points on the axis, when D(i, i+1)≥T, the seedling strips numbered i and (i+1) are both identified as corn seedling strips, and the points within the corn seedling strips are all used as target points of the corn seedling strips, while those that are not identified are classified as target points of the soybean seedling strips.
[0059] Extraction of seedling belt navigation line: The target points of soybean seedling belt and corn seedling belt are clustered using Euclidean clustering algorithm, and the target points are fitted with straight lines by the least squares method to obtain the seedling belt navigation line. The results are shown in the figure. Figure 7 shown.
[0060] Specifically, the average slope of all soybean seedling belts and corn seedling belts is calculated, and the corn seedling belt navigation line Y1 and the soybean seedling belt navigation line Y2 are extracted; Y1: y1 = k1x1 + b1, Y2: y2 = k2x2 + b2; the average slope m0 and the intersection coordinates P (x0, y0) of the corn seedling belt navigation line Y1 and the soybean seedling belt navigation line Y2 are calculated to obtain the navigation line Y. The calculation formula is as follows:
[0061] In this embodiment, the method for identifying seedling belts of strip-shaped composite planting of crops in the seedling stage further includes: planning the path of the agricultural machinery through the seedling belt navigation line.
[0062] Specifically, the central axis of the agricultural machinery chassis is used as the X-axis and is aligned with the world coordinate system X w The axes are parallel, and the navigation angle θ and the lateral deviation d are calculated. The navigation angle θ is the angle between the guidance route and the center axis of the trolley chassis, and the lateral deviation d refers to the vertical distance from the center of the agricultural machinery to the navigation line, that is, the shortest distance perpendicular to the navigation line. The calculation formulas for the navigation angle θ and the lateral deviation d are as follows: Where (x1, y1) is the coordinate of the center of the agricultural machinery in the world coordinate system.
[0063] In summary, the present invention can realize automatic identification of soybean seedling strips and corn seedling strips in the seedling stage, and provide accurate navigation guidance for the autonomous operation of soybean-corn strip composite planting machinery.
[0064] The various devices selected in this application (components whose specific structures are not described) are all universal standard parts or components known to technical personnel in this field, and their structures and principles can be known to technical personnel through technical manuals or through conventional experimental methods.
[0065] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.
[0066] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.
Claims
1. A method for identifying seedling belts of crop strips in composite planting at the seedling stage, characterized in that: include: Obtain point cloud data of the recognition area; Determine the seedling belt target area in the identification area to extract point cloud data of the seedling belt target area; Extract various seedling belts from the point cloud data of the seedling belt target area, and extract the target points of each seedling belt; Distinguish the target points corresponding to each type of seedling belt to obtain the seedling belt navigation line; The method for obtaining point cloud data of the identification area includes: Obtain point cloud data of the identification area through laser radar; Traverse each point in the point cloud data of the recognition area, transform the point cloud data in the radar coordinate system into the world coordinate system through the coordinate transformation matrix, and separate the ground point cloud data and non-ground point cloud data; Filter out noise in non-ground point cloud data; The method for determining the seedling belt target area in the identification area to extract point cloud data of the seedling belt target area includes: The length and width of the seedling belt target area are set. The length of the seedling belt target area is along the x-axis direction of the world coordinate system, and the width of the seedling belt target area is along the y-axis direction of the world coordinate system, so that the center point of the seedling belt target area is located on the x-axis of the world coordinate system; Traverse each point in the non-ground point cloud data, when the projection point of any point on the coordinate plane of the world coordinate system is located in the seedling belt target area, the point cloud data corresponding to the point is the point cloud data of the seedling belt target area; The method of extracting various types of seedling belts from the point cloud data of the seedling belt target area and extracting the target points of various types of seedling belts includes: Determine the initial horizontal target area within the seedling strip target area; Extract point cloud data within the initial horizontal target area to determine the width of the seedling belt within the initial horizontal target area; Select the target area of the seedling belt according to the width of the seedling belt to extract the first type of seedling belt and the second type of seedling belt; The total number of seedling belts of the first type and the second type is compared with the set belt number threshold, and when the total number of seedling belts of the first type and the second type is not less than the set belt number threshold, the target points of each type of seedling belt are extracted.
2. The method for identifying seedling strips of crop strip composite planting in the seedling stage according to claim 1, characterized in that: The method for determining the initial horizontal target area in the seedling belt target area comprises: The length and width of the initial horizontal target area are set in the seedling belt target area. The length of the initial horizontal target area is along the x-axis direction of the world coordinate system, and the width of the initial horizontal target area is along the y-axis direction of the world coordinate system, so that the center point of the initial horizontal target area is located on the x-axis of the world coordinate system.
3. The method for identifying seedling strips of crop strip composite planting in the seedling stage according to claim 2, characterized in that: The method for extracting point cloud data within the initial horizontal target area to determine the width of the seedling band within the initial horizontal target area includes: Traverse each point in the non-ground point cloud data in the initial horizontal target area. When the projection point of any point on the coordinate plane of the world coordinate system is within the initial horizontal target area, the point cloud data corresponding to the point is the point cloud data of the initial horizontal target area. The point cloud data within the initial horizontal target area are clustered, and the bandwidth of the first type of seedlings and the bandwidth of the second type of seedlings within the initial horizontal target area are determined according to the maximum distance between the point cloud data within the initial horizontal target area in each cluster.
4. The method for identifying seedling strips of crop strip composite planting in the seedling stage according to claim 3, characterized in that: The method of selecting the seedling belt target area according to the width of the seedling belt to extract the first type of seedling belt and the second type of seedling belt comprises: In the target area of the seedling belt, the corresponding frame selection length and frame selection width are set according to the width of the first type of seedling belt and the second type of seedling belt, so as to extract the first type of seedling belt and the second type of seedling belt respectively.
5. The method for identifying seedling strips of crop strip composite planting in the seedling stage according to claim 4, characterized in that: The method of comparing the total number of seedling belts of the first type of seedling belt and the second type of seedling belt with a set belt number threshold, and when the total number of seedling belts of the first type of seedling belt and the second type of seedling belt is not less than the set belt number threshold, extracting target points of each type of seedling belt includes: Set the band count threshold; The total number of seedling belts of the first type and the second type is compared with the belt number threshold; When the total number of seedling belts of the first type and the second type is less than the belt number threshold, the target area of the seedling belt is re-determined, or When the total number of seedling belts of the first type and the second type is not less than the set belt number threshold, the target points of each type of seedling belt are extracted.
6. The method for identifying seedling strips of crop strip composite planting in the seedling stage according to claim 5, characterized in that: The method of respectively distinguishing the target points corresponding to various seedling belts to obtain the seedling belt navigation line includes: Determine the location of each type of seedling belt according to its target point; The distances between adjacent seedling strips are calculated in sequence to set the row spacing threshold; Compare the distance between adjacent seedling belts with the row spacing threshold to distinguish each type of seedling belt and the target points corresponding to each type of seedling belt; Clustering and straight line fitting of target points of various seedling belts are carried out to obtain the belt lines corresponding to various seedling belts; Obtain the navigation lines corresponding to each type of seedling belt according to the belt lines corresponding to each type of seedling belt; Obtain the seedling belt navigation line according to the navigation lines corresponding to each type of seedling belt.
7. The method for identifying seedling strips of crop strip composite planting in the seedling stage according to claim 1, characterized in that: Also includes Plan the path of agricultural machinery using the seedling belt navigation lines.
Citation Information
Patent Citations
Orchard inter-row path extraction method based on laser radar
CN111983637A
Field seedling zone navigation line detection method based on laser radar point cloud
CN113376614A
Agricultural robot navigation line extraction method based on 3D laser radar navigation
CN113867339A
Light intelligent orchard edge navigation method
CN114485667A
Vehicle classification method and system
CN115439648A
