Pruning method for different tree vigor of camellia oleifera tree based on machine vision
Automatically identify the maximum crown diameter and tree potential of the oil tea tree through machine vision technology, formulate corresponding pruning strategies, solve the problems of manual judgment error and high cost, and improve the efficiency and accuracy of oil tea tree pruning.
Patent Information
- Application Number
- CN202411233678.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-26
AI Technical Summary
The existing oil tea tree pruning technology relies on manual judgment of tree potential, which has errors and high costs, making it difficult to achieve "adapt to the tree", resulting in unstable oil tea production.
Using a machine vision-based method, the canopy information of the oil tea tree is obtained through a depth camera, combined with the YOLOv8-seg model and normal distribution law, the maximum diameter of the canopy is automatically calculated, and the pruning strategy is set according to different tree potentials, including light shearing and medium shearing. and re-cut.
It greatly reduces manpower and material costs, improves the feasibility of "adapting to trees" of pruning agronomics, and improves the efficiency and accuracy of pruning oil tea trees.
Smart Images

Figure CN119014229A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of fruit tree pruning, and more specifically, relates to a machine vision-based pruning method for tea oil trees with different tree vigors. Background Art
[0002] Camellia oleifera is a woody oil plant unique to my country. It has important economic and ecological value and is mainly distributed in southern my country. The fruit of the camellia oleifera can be used to extract oil. Camellia oleifera seed oil has high nutritional value and is known as "Oriental olive oil". In addition, the camellia oleifera also has ecological functions such as soil and water conservation, air purification, and soil structure improvement. However, due to the large number of varieties and mixed management, the yield of camellia oleifera is unstable. At present, the pruning of camellia oleifera still mainly relies on manual pruning. Although the pruning agronomy of camellia oleifera has been relatively perfected, such as CN108990658A and CN106818262A, both fully introduce the pruning agronomy of camellia oleifera from the seedling stage to the fruiting stage. However, due to the errors in the judgment of the tree body by the pruning workers, the "adapting to the tree" in the pruning agronomy cannot be achieved. Moreover, to meet the "adapting to the tree" in the pruning agronomy, a large amount of manpower and material resources are required, and the cost is too high, so there is an urgent need for a pruning method that can be applied to mechanized intelligent identification of different tree vigor, so as to reduce the management cost of the camellia oleifera industry and increase the income. Summary of the invention
[0003] The purpose of the present invention includes providing a machine vision-based pruning method for tea oil trees with different tree potentials to improve at least some of the above problems.
[0004] The technical problem solved by the present invention is achieved by adopting the following technical solutions: A machine vision-based pruning method for tea oil trees with different tree vigors, characterized in that it comprises the following steps: Step 1: Obtain the maximum diameter information of the oil tea tree crown; Step 2: Organize and analyze the maximum crown diameter information according to the normal distribution law, and set the crown diameter ranges of different tree potentials; Step 3: Develop corresponding pruning strategies based on the pruning rules for different oil tea trees.
[0005] Furthermore, the specific steps of step 1 are as follows: Step 1.1: Use a depth camera to obtain the image information and depth information of the tree crown directly above the oil tea tree, and remove the blurred images in the acquired images to establish the oil tea tree crown dataset; Step 1.2: Process the tea tree crown dataset in step 1.1 through the trained YOLOv8-seg model to obtain a tea tree crown detection result map, which includes the detection information of the tea tree crown contour and the two-dimensional image position information of the tea tree crown contour; Step 1.3: The oil tea tree crown detection result image obtained in step 1.2 is first subjected to morphological processing, and then the crown contour is detected to determine the two-dimensional image position information of the farthest two points in the crown contour at a horizontal distance; Step 1.4: The depth information obtained in step 1.1 is fused with the two-dimensional image position information of the farthest two points in the horizontal distance in the crown outline obtained in step 1.3 to calculate the maximum diameter information of the oil tea tree crown.
[0006] Furthermore, the specific steps of step 2 are as follows: Step 2.1: According to the method in step 1, measure and collect the maximum crown diameter information of all tea oil trees in the tea oil base to establish a tea oil tree crown diameter dataset; Step 2.2: According to the “σ principle” in the normal distribution law, the maximum crown diameter interval of the oil tea tree is divided into (0, μ-σ], (μ-σ, μ+σ] and (μ+σ, +∞), where μ represents the mean of the maximum crown diameter of the measured oil tea trees and σ represents the standard deviation, corresponding to the different growth conditions of the oil tea trees.
[0007] Furthermore, the specific steps of step 3 are as follows: Step 3.1: The oil tea trees are divided into weak trees, ordinary trees and strong trees according to their growth conditions, wherein the maximum diameter intervals of the crowns of the weak trees, ordinary trees and strong trees correspond to (0, μ-σ], (μ-σ, μ+σ] and (μ+σ, +∞), respectively; Step 3.2: According to the pruning agronomy of tea oil trees, the pruning intensity of the tree body is divided into light pruning, medium pruning and heavy pruning. The light pruning is to prune the foot branches, diseased and insect-infested branches, dead branches, damaged branches and weak fruiting branches in the tea oil trees. The medium pruning is to prune the drooping branches, inner cross branches, overlapping branches and long branches on the basis of light pruning. The heavy pruning is to prune the drooping branches, inner cross branches and overlapping branches on the basis of light pruning and medium pruning, so as to achieve the purpose of modifying the tree structure; Step 3.3: According to the pruning requirements of tea oil trees, the weak trees, ordinary trees and strong trees are respectively matched with heavy pruning, medium pruning and light pruning strategies to meet the pruning needs under different tree conditions.
[0008] Furthermore, the maximum diameter interval of the crown of the tea oil tree in step 2.2 is not limited to the above intervals, and the number and range of intervals can be adjusted according to corresponding circumstances.
[0009] Furthermore, in step 3.1, the classification of tea oil trees into weak trees, ordinary trees and strong trees is not limited to the above three categories, and the number of classifications can be adjusted according to corresponding circumstances. In step 3.2, the pruning intensity of the tree body is divided into light pruning, medium pruning and heavy pruning, which is not limited to the above three categories, and the number of pruning intensity types can be adjusted according to corresponding circumstances.
[0010] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The present invention can collect a large amount of maximum diameter data of oil tea tree crowns, use the normal distribution law to classify trees with different tree vigor conditions into intervals, and combine the deep learning model with the binocular camera. It can independently calculate the maximum diameter of the required oil tea tree crown, and then compare it with the crown diameter interval to determine the tree vigor condition and formulate corresponding pruning strategies, which greatly reduces the manpower and material costs and increases the feasibility of "adapting measures to each tree" in pruning agronomy. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0012] Figure 1 It is a flow chart of a method for pruning tea oil trees with different tree potentials based on machine vision in an embodiment of the present invention; DETAILED DESCRIPTION
[0013] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand and clarify, the present invention is further described below in conjunction with specific embodiments.
[0014] A machine vision-based pruning method for tea oil trees with different tree vigors, characterized in that it comprises the following steps: Step 1: Obtain the maximum diameter information of the oil tea tree crown; Step 2: Organize and analyze the maximum crown diameter information according to the normal distribution law, and set the crown diameter ranges of different tree potentials; Step 3: Develop corresponding pruning strategies based on the pruning rules for different oil tea trees.
[0015] The specific steps of step 1 are as follows: Step 1.1: Use a depth camera to obtain the image information and depth information of the tree crown directly above the oil tea tree, and remove the blurred images in the acquired images to establish the oil tea tree crown dataset; Step 1.2: Process the tea tree crown dataset in step 1.1 through the trained YOLOv8-seg model to obtain a tea tree crown detection result map, which includes the detection information of the tea tree crown contour and the two-dimensional image position information of the tea tree crown contour; Step 1.3: The oil tea tree crown detection result image obtained in step 1.2 is first subjected to morphological processing, and then the crown contour is detected to determine the two-dimensional image position information of the farthest two points in the crown contour at a horizontal distance; Step 1.4: The depth information obtained in step 1.1 is fused with the two-dimensional image position information of the farthest two points in the horizontal distance in the crown outline obtained in step 1.3 to calculate the maximum diameter information of the oil tea tree crown.
[0016] The specific steps of step 2 are as follows: Step 2.1: According to the method in step 1, measure and collect the maximum crown diameter information of all tea oil trees in the tea oil base to establish a tea oil tree crown diameter dataset; Step 2.2: According to the “σ principle” in the normal distribution law, the maximum crown diameter interval of the oil tea tree is divided into (0, μ-σ], (μ-σ, μ+σ] and (μ+σ, +∞), where μ represents the mean of the maximum crown diameter of the measured oil tea trees and σ represents the standard deviation, corresponding to the different growth conditions of the oil tea trees.
[0017] The specific steps of step 3 are as follows: Step 3.1: The oil tea trees are divided into weak trees, ordinary trees and strong trees according to their growth conditions, wherein the maximum diameter intervals of the crowns of the weak trees, ordinary trees and strong trees correspond to (0, μ-σ], (μ-σ, μ+σ] and (μ+σ, +∞), respectively; Step 3.2: According to the pruning agronomy of tea oil trees, the pruning intensity of the tree body is divided into light pruning, medium pruning and heavy pruning. The light pruning is to prune the foot branches, diseased and insect-infested branches, dead branches, damaged branches and weak fruiting branches in the tea oil trees. The medium pruning is to prune the drooping branches, inner cross branches, overlapping branches and long branches on the basis of light pruning. The heavy pruning is to prune the drooping branches, inner cross branches and overlapping branches on the basis of light pruning and medium pruning, so as to achieve the purpose of modifying the tree structure; Step 3.3: According to the pruning requirements of tea oil trees, the weak trees, ordinary trees and strong trees are respectively matched with heavy pruning, medium pruning and light pruning strategies to meet the pruning needs under different tree conditions.
[0018] 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 drive the essence of the corresponding technical solutions away from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A machine vision-based pruning method for tea oil trees with different tree potentials, characterized in that: The following steps are involved: Step 1: Obtain the maximum diameter information of the oil tea tree crown; Step 2: Organize and analyze the maximum crown diameter information according to the normal distribution law, and set the crown diameter ranges of different tree potentials; Step 3: Develop corresponding pruning strategies based on the pruning rules for different oil tea trees.
2. The machine vision-based pruning method for tea oil trees of different tree potentials according to claim 1, characterized in that: The specific steps of step 1 are as follows: Step 1.1: Use a depth camera to obtain the image information and depth information of the tree crown directly above the oil tea tree, and remove the blurred images in the acquired images to establish the oil tea tree crown dataset; Step 1.2: Process the tea tree crown dataset in step 1.1 through the trained YOLOv8-seg model to obtain a tea tree crown detection result map, which includes the detection information of the tea tree crown contour and the two-dimensional image position information of the tea tree crown contour; Step 1.3: The oil tea tree crown detection result image obtained in step 1.2 is first subjected to morphological processing, and then the crown contour is detected to determine the two-dimensional image position information of the farthest two points in the crown contour at a horizontal distance; Step 1.4: The depth information obtained in step 1.1 is fused with the two-dimensional image position information of the farthest two points in the horizontal distance in the crown outline obtained in step 1.3 to calculate the maximum diameter information of the oil tea tree crown.
3. The machine vision-based pruning method for tea oil trees of different tree potentials according to claim 1, characterized in that: The specific steps of step 2 are as follows: Step 2.1: According to the method in step 1, measure and collect the maximum crown diameter information of all tea oil trees in the tea oil base to establish a tea oil tree crown diameter dataset; Step 2.2: According to the "σ principle" in the normal distribution law, the maximum crown diameter interval of the oil tea tree is divided into (0, μ-σ], (μ-σ, μ+σ] and (μ+σ, +∞), where μ represents the mean of the maximum crown diameter of the measured oil tea trees and σ represents the standard deviation, corresponding to the different growth conditions of the oil tea trees.
4. The machine vision-based pruning method for tea oil trees of different tree potentials according to claim 1, characterized in that: The specific steps of step 3 are as follows: Step 3.1: The oil tea trees are divided into weak trees, ordinary trees and strong trees according to their growth conditions, wherein the maximum diameter intervals of the crowns of the weak trees, ordinary trees and strong trees correspond to (0, μ-σ], (μ-σ, μ+σ] and (μ+σ, +∞), respectively; Step 3.2: According to the pruning agronomy of tea oil trees, the pruning intensity of the tree body is divided into light pruning, medium pruning and heavy pruning. The light pruning is to prune the foot branches, diseased and insect-infested branches, dead branches, damaged branches and weak fruiting branches in the tea oil trees. The medium pruning is to prune the drooping branches, inner cross branches, overlapping branches and long branches on the basis of light pruning. The heavy pruning is to prune the drooping branches, inner cross branches and overlapping branches on the basis of light pruning and medium pruning, so as to achieve the purpose of modifying the tree structure; Step 3.3: According to the pruning requirements of tea oil trees, the weak trees, ordinary trees and strong trees are respectively matched with heavy pruning, medium pruning and light pruning strategies to meet the pruning needs under different tree conditions.
5. The machine vision-based pruning method for tea oil trees of different tree potentials according to claim 3, characterized in that: The maximum diameter interval of the crown of the tea oil tree in step 2.2 is not limited to the above intervals, and the number and range of intervals can be adjusted according to corresponding circumstances.
6. The machine vision-based pruning method for tea oil trees of different tree potentials according to claim 4, characterized in that: In the step 3.1, the oil tea trees are divided into weak trees, ordinary trees and strong trees, but not limited to the above three categories, and the number of categories can be adjusted according to the corresponding situation. In the step 3.2, the pruning intensity of the tree body is divided into light pruning, medium pruning and heavy pruning, but not limited to the above three categories, and the number of pruning intensity types can be adjusted according to the corresponding situation.
Citation Information
Patent Citations
Process for accurate pruning of camellia oleifera trees
CN106818262A
Camellia oleifera shaping trimming method
CN108990658A
