Method for automatically extracting position of safflower point cloud picking point in blooming period
By identifying the three-dimensional point cloud of red flowers, calculating the coordinates of filaments and fruit spheres, and determining the picking points, the time-consuming and wrong picking problems of red flowers are solved, and the automatic extraction of red flowers is realized, providing data support for the picking robot and improving picking. efficiency.
Patent Information
- Application Number
- CN202410936966.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-15
AI Technical Summary
The current safflower picking is time-consuming and labor-intensive during the blooming process. The mechanized picking has problems such as mis-picking, missed picking and miscellaneous debris, and lacks technical support for automated picking.
By identifying the three-dimensional point cloud of red flowers, semantic segmentation is performed to obtain the filament and fruit ball point cloud, calculate the filament center of mass and fruit ball center coordinates, and determine the picking points with unit vectors to realize automatic extraction of the safflower picking points.
Automatic positioning of safflower picking points is realized, providing data support for subsequent picking robots, improving picking efficiency, and reducing missed picking and debris picking.
Smart Images

Figure CN118967787A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of plant safflower picking, and in particular to a method for automatically extracting picking point positions of safflower point clouds in bloom. Background Art
[0002] In the process of safflower picking, the summer picking of safflower in the blooming period (July-August) is a very time-consuming and labor-intensive process. By accurately determining the safflower picking time to achieve the purpose of economic optimization, the yield of safflower in the later period can be increased. However, safflower picking is currently mainly done manually, which has high labor intensity and high labor costs, and mechanized whole-plant geometric picking is prone to wrong picking, missed picking and impurities. In order to improve the efficiency of safflower picking in the blooming period and reduce the wrong picking, missed picking and picking of impurities in the safflower process, the development of a safflower picking robot in the blooming period is of great practical application value. An important step in the development of a safflower picking robot is to automatically extract the picking point position parameters based on the identification of safflower, and provide data support for the subsequent automatic picking of safflower. Summary of the invention
[0003] In order to solve the problems and needs existing in the background technology, the present invention proposes a method for automatically extracting the picking point positions of safflower point clouds in bloom.
[0004] The present invention solves the technical problem by adopting the following technical solution:
[0005] A method for automatically extracting the location of picking points of safflower point clouds in full bloom period comprises the following steps:
[0006] Step S1, obtaining a single safflower three-dimensional point cloud for the identified safflower, and performing semantic segmentation on the safflower three-dimensional point cloud to obtain a safflower filament point cloud and a safflower fruit ball point cloud;
[0007] Step S2, using the obtained safflower filament point cloud, calculating the mass center of the safflower filament, and obtaining the three-dimensional coordinate value of the mass center of the safflower filament; for the safflower fruit ball point cloud, using the random sample consensus algorithm to perform spatial sphere fitting on the safflower fruit ball, and obtaining the coordinates of the center of the safflower fruit ball fitting spatial sphere and the radius of the fitting spatial sphere;
[0008] Step S3, using the obtained 3D coordinate value of the mass center of the safflower filament and the coordinates of the center of the safflower fruit ball fitting space ball, calculate the spatial unit vector pointing from the center of the safflower fruit ball fitting space ball to the mass center of the safflower filament;
[0009] Step S4, using the coordinates of the center of the safflower ball fitting space ball, calculate the radius distance of the safflower ball fitting space ball moved along the unit vector of the center of mass space, and determine the three-dimensional coordinate value of the point as the safflower picking point in the blooming period.
[0010] Furthermore, the specific method of step S1 is: using a visual perception system built based on a Canon EOSR50 camera to obtain a high-quality three-dimensional point cloud of a single safflower in full bloom, and using a deep learning method to perform semantic segmentation on the three-dimensional point cloud of the safflower to obtain a single safflower filament point cloud and a safflower fruit ball point cloud.
[0011] Furthermore, the specific method of step S2 is:
[0012] Step S21, using the obtained safflower filament point cloud, calculate the three-dimensional coordinate value P1 of the mass center of the safflower filament, the three-dimensional coordinate value P1 (x1, y1, z1) of the mass center of the safflower filament is:
[0013]
[0014]
[0015]
[0016] Among them, n is the serial number of the point, and N is the number of points;
[0017] Step S22, using the random sample consensus algorithm to fit the fruit ball, and obtain the coordinates of the center of the safflower fruit ball fitting space ball P2 (x2, y2, z2) and the radius of the fitting space ball R i :
[0018] (x i -x2) 2 +(y i -y2) 2 +(z i -z2) 2 =R i 2 (R i >0).
[0019] Furthermore, the specific method of step S3 is:
[0020] Calculate the space vector by fitting the space ball center coordinates of the safflower filament mass center and the safflower fruit ball Use this space vector to calculate the unit vector
[0021]
[0022]
[0023] Furthermore, the specific method of step S4 is:
[0024] Use the fruit ball to fit the center coordinates P2 of the space ball and calculate the center coordinates P2 along the unit vector Moving fruit ball fitting space ball radius R iDistance, determine the three-dimensional coordinate value of the point as the safflower picking point P in the blooming period:
[0025]
[0026] Beneficial effects of the present invention:
[0027] The present invention realizes automatic extraction and positioning of safflower picking points based on the identification of the three-dimensional point cloud of safflower plants in full bloom, thus providing a basis for subsequent automatic picking by a picking robot. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 It is the overall flow chart of the present invention.
[0029] Figure 2 It is a schematic diagram of the three-dimensional point cloud of safflower of the present invention.
[0030] Figure 3 It is a schematic diagram of the method of separating safflower filaments and safflower fruit balls according to the present invention.
[0031] Figure 4 It is a schematic diagram of the coordinates of the centroid of safflower silk of the present invention.
[0032] Figure 5 It is a schematic diagram of fitting the safflower ball of the present invention.
[0033] Figure 6 It is a schematic diagram of the direction vector of the present invention.
[0034] Figure 7 The figure is a schematic diagram of the coordinates of the safflower picking points of the present invention.
[0035] In the figure: 3-1 is the filament of safflower; 3-2 is the fruit ball of safflower. DETAILED DESCRIPTION
[0036] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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.
[0037] Reference Figure 1-7 The present invention provides a method for automatically extracting the location of picking points of safflower point clouds in full bloom, comprising the following steps:
[0038] Step S1, obtaining a single safflower three-dimensional point cloud for the identified safflower, and performing semantic segmentation on the safflower three-dimensional point cloud to obtain a safflower filament point cloud and a safflower fruit ball point cloud;
[0039] Step S2, since two point clouds cannot be operated at the same time, the safflower filament point cloud and the safflower fruit ball point cloud are solved separately using the idea of segmented solution. Therefore, the obtained safflower filament point cloud is used to calculate the safflower filament mass center and obtain the 3D coordinate value of the safflower filament mass center; for the safflower fruit ball point cloud, the random sample consensus algorithm is used to fit the safflower fruit ball to obtain the coordinates of the safflower fruit ball fitting space ball center and the radius of the fitting space ball;
[0040] Step S3, using the obtained 3D coordinate value of the mass center of the safflower filament and the coordinates of the center of the safflower fruit ball fitting space ball, calculate the spatial unit vector pointing from the center of the safflower fruit ball fitting space ball to the mass center of the safflower filament;
[0041] Step S4, using the coordinates of the center of the safflower ball fitting space ball, calculate the radius distance of the safflower ball fitting space ball moved along the unit vector of the center of mass space, and determine the three-dimensional coordinate value of the point as the safflower picking point in the blooming period.
[0042] Furthermore, the specific method of step S1 is: using a visual perception system built based on a Canon EOSR50 camera to obtain a high-quality three-dimensional point cloud of a single safflower in full bloom, and using a deep learning method to perform semantic segmentation on the three-dimensional point cloud of the safflower to obtain a single safflower filament point cloud and a safflower fruit ball point cloud.
[0043] Furthermore, the specific method of step S2 is:
[0044] Step S21, using the obtained safflower filament point cloud, calculate the three-dimensional coordinate value P1 of the mass center of the safflower filament, the three-dimensional coordinate value P1 (x1, y1, z1) of the mass center of the safflower filament is:
[0045]
[0046]
[0047]
[0048] Among them, n is the serial number of the point, and N is the number of points;
[0049] Step S22, using the random sample consensus algorithm to fit the fruit ball, and obtain the coordinates of the center of the safflower fruit ball fitting space ball P2 (x2, y2, z2) and the radius of the fitting space ball R i :
[0050] (x i -x2) 2 +(y i -y2) 2 +(z i -z2) 2 =R i2 (R i >0).
[0051] Furthermore, the specific method of step S3 is:
[0052] Calculate the space vector by fitting the space ball center coordinates of the safflower filament mass center and the safflower fruit ball Use this space vector to calculate the unit vector
[0053]
[0054]
[0055] Furthermore, the specific method of step S4 is:
[0056] Use the fruit ball to fit the center coordinates P2 of the space ball and calculate the center coordinates P2 along the unit vector Moving fruit ball fitting space ball radius R i Distance, determine the three-dimensional coordinate value of the point as the safflower picking point P in the blooming period:
[0057]
[0058] The present invention realizes automatic extraction and positioning of safflower picking points based on the identification of the three-dimensional point cloud of safflower plants in full bloom, thus providing a basis for subsequent automatic picking by a picking robot.
[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for automatically extracting the location of picking points of safflower point cloud in full bloom, It is characterized in that The steps include: Step S1, obtaining a single safflower three-dimensional point cloud for the identified safflower, and performing semantic segmentation on the safflower three-dimensional point cloud to obtain a safflower filament point cloud and a safflower fruit ball point cloud; Step S2, using the obtained safflower filament point cloud, calculating the mass center of the safflower filament, and obtaining the three-dimensional coordinate value of the mass center of the safflower filament; for the safflower fruit ball point cloud, using the random sample consensus algorithm to perform spatial sphere fitting on the safflower fruit ball, and obtaining the coordinates of the center of the safflower fruit ball fitting spatial sphere and the radius of the fitting spatial sphere; Step S3, using the obtained 3D coordinate value of the mass center of the safflower filament and the coordinates of the center of the safflower fruit ball fitting space ball, calculate the spatial unit vector pointing from the center of the safflower fruit ball fitting space ball to the mass center of the safflower filament; Step S4, using the coordinates of the center of the safflower ball fitting space ball, calculate the radius distance of the safflower ball fitting space ball moved along the unit vector of the center of mass space, and determine the three-dimensional coordinate value of the point as the safflower picking point in the blooming period.
2. According to the method for automatically extracting the location of the picking points of safflower point cloud in full bloom period according to claim 1, It is characterized in that The specific method of step S1 is: using a visual perception system built based on a Canon EOS R50 camera to obtain a high-quality three-dimensional point cloud of a single safflower in full bloom, and using a deep learning method to perform semantic segmentation on the three-dimensional point cloud of the safflower to obtain a single safflower filament point cloud and a safflower fruit ball point cloud.
3. According to claim 2, a method for automatically extracting the location of picking points of safflower point cloud in full bloom, It is characterized in that The specific method of step S2 is: Step S21, using the obtained safflower filament point cloud, calculate the 3D coordinate value P of the centroid of the safflower filament 1 , the three-dimensional coordinate value of the mass center of safflower silk P 1 (x 1 ,y 1 , z 1 )for: Among them, n is the serial number of the point, and N is the number of points; Step S22, using the random sample consensus algorithm to fit the fruit ball, and obtain the coordinates P of the center of the safflower fruit ball in the fitting space 2 (x 2 ,y 2 , z 2 ) and the radius R of the fitted space sphere i : (x i -x 2 ) 2 +(y i -y 2 ) 2 +(z i -z 2 ) 2 =R i 2 (R i >0)。 4. According to claim 3, a method for automatically extracting the location of picking points of safflower point cloud in full bloom, It is characterized in that The specific method of step S3 is: Calculate the space vector by fitting the space ball center coordinates of the safflower filament mass center and the safflower fruit ball Use this space vector to calculate the unit vector 5. According to claim 4, a method for automatically extracting the location of picking points of safflower point cloud in full bloom, It is characterized in that The specific method of step S4 is: Use the fruit ball to fit the center coordinates P of the space ball 2 , calculate the coordinates of the sphere center P 2 Along the unit vector Moving fruit ball fitting space ball radius R i Distance, determine the three-dimensional coordinate value of the point as the safflower picking point P in the blooming period:
