Method and system for monitoring growth situation of fruit tree

Abstract

The invention relates to the technical field of intelligent agriculture, in particular to a growth situation monitoring method and system for fruit trees, and the method comprises the following steps: collecting the image information of an orchard through an image collection device based on the position information of the orchard, analyzing and recognizing the position information of a plurality of fruit trees and obstacles in the orchard, and generating a three-dimensional model of the orchard. According to the invention, through image acquisition and three-dimensional model construction, the positions and surrounding environment information of a plurality of fruit trees are accurately identified, blind spots are reduced, resource allocation is optimized, a routing inspection path is optimized, the utilization rate of manpower and time resources is improved, and through real-time analysis of fruit tree state images and marking of key growth areas, the quality of the fruit trees is improved. The monitoring capability of the fruit tree health condition is improved, the early recognition capability of diseases and insect pests is enhanced, the timeliness and effectiveness of prevention and control measures are improved, the growth condition of the fruit tree is comprehensively evaluated by analyzing image information and physiological parameters, the diseases and insect pests are recognized in the early stage and intervened in time, and the quality and yield of fruits are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent agricultural technology, and in particular to a growth status monitoring method and system for fruit trees. Background Art

[0002] The field of smart agricultural technology focuses on applying information technology and automation technology to agricultural production, aiming to improve efficiency, reduce resource waste and increase crop yields. It uses the Internet of Things, big data analysis, machine learning, remote sensing technology and automated robots to achieve precise control of the agricultural environment and detailed monitoring of crop growth. It can also monitor soil moisture, temperature, and crop health in real time. It can also irrigate, fertilize, and control pests and diseases based on the data. It can also combine automated equipment for planting, spraying pesticides and harvesting, including drones and self-driving tractors, to reduce manpower requirements, reduce costs, and improve the accuracy and efficiency of operations.

[0003] Among them, the growth status monitoring method for fruit trees collects and analyzes image data of fruit trees, monitors growth status and health status, helps fruit farmers manage fruit trees in a scientific way, ensures healthy growth of fruit trees, and optimizes the overall yield and quality of the orchard. By continuously monitoring multiple growth parameters of fruit trees, including growth rate, morphology and growth cycle, cultivation strategies can be adjusted in real time, including irrigation plans, fertilizer application amounts and pest and disease control measures, to improve the level of refinement of orchard management, and predict and solve problems affecting fruit tree production in advance.

[0004] Traditional fruit tree growth status monitoring methods are insufficient in terms of detailed monitoring of each growth stage of fruit trees. They focus on macro-climate and soil condition monitoring, lack accurate identification and analysis of individual differences of each fruit tree, and limit the personalization and precision of management measures. The identification of fruit tree diseases and pests relies on periodic manual inspections or simple image recognition, lacks real-time and predictive capabilities, resulting in delayed response to disease and pest control, and failure to minimize losses. This makes it impossible for farmers to fully utilize existing data for refined management, affecting the overall economic benefits and sustainable development of the orchard. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a growth status monitoring method and system for fruit trees.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme, a method for monitoring the growth status of fruit trees, comprising the following steps:

[0007] S1: Based on the orchard location information, use image acquisition equipment to collect image information of the orchard, analyze and identify the location information of multiple fruit trees and obstacles in the orchard, and generate a three-dimensional model of the orchard;

[0008] S2: Based on the three-dimensional model of the orchard, the inspection path is adjusted in combination with obstacle information, the coverage rate and inspection efficiency are considered, the inspection path is calculated, and image information of multiple fruit trees is collected to generate an inspection path calculation result;

[0009] S3: Based on the inspection path calculation result, the fruit tree status image is analyzed, key structures in the image are identified, key growth areas are recorded, and image information is annotated to generate an annotation information data set;

[0010] S4: Analyze the physiological parameters of the fruit trees according to the labeled information data set, evaluate the health status of multiple fruit trees by comparing them with the physiological parameters of healthy fruit trees, identify abnormal growth patterns, and generate fruit tree health assessment information;

[0011] S5: Based on the fruit tree health assessment information, combined with the fruit tree image information, by comparing with known pest and disease feature images, identifying the pest and disease characteristics of the fruit tree, and analyzing the types of pests and diseases according to abnormal growth patterns to generate pest and disease risk prediction results;

[0012] S6: Based on the pest and disease risk prediction results, by analyzing the morphological characteristics of the fruit, evaluating the fruit quality and quality grade, recording the growth and health status information of the fruit trees and matching the management response information, the growth status monitoring information is generated.

[0013] As a further solution of the present invention, the three-dimensional model of the orchard includes an orchard terrain height map, fruit tree position marking information, and obstacle spatial layout information; the inspection path calculation results include an optimal path map, an estimated inspection time, and energy consumption estimation information; the annotation information data set includes trunk diameter and height measurement data, fruit quantity and maturity rating scores, and leaf health indicators; the fruit tree health assessment information includes chlorophyll concentration index, plant canopy nitrogen content information, and fruit size and maturity identification information; the pest and disease risk prediction results include a list of potential pest and disease types, infection possibility rating information, and disease response management matching information; the growth trend monitoring information includes fruit quality classification results, expected harvest, and fruit tree growth cycle analysis results.

[0014] As a further solution of the present invention, based on the orchard location information, an image acquisition device is used to collect image information of the orchard, and the location information of multiple fruit trees and obstacles in the orchard is analyzed and identified. The steps of generating a three-dimensional model of the orchard are specifically as follows:

[0015] S101: Based on the orchard location information, use an image acquisition device to collect image data inside the orchard, record information about the land, trees, and surrounding environment, and generate orchard image data;

[0016] S102: Based on the orchard image data, analyzing the image, identifying location information of multiple fruit trees and obstacles, including facilities and vegetation, and generating fruit tree location information;

[0017] S103: Based on the fruit tree location information, a three-dimensional terrain model of the target orchard is constructed, including the heights, volumes, and spatial layout of obstacles of multiple fruit trees, to generate a three-dimensional model of the orchard.

[0018] As a further solution of the present invention, based on the orchard three-dimensional model, the inspection path is adjusted in combination with obstacle information, the coverage rate and inspection efficiency are considered, the inspection path is calculated, and image information of multiple fruit trees is collected. The steps of generating the inspection path calculation result are specifically as follows:

[0019] S201: Based on the three-dimensional model of the orchard, analyzing the walkable area and obstacles in the orchard, identifying the position and size information of multiple obstacles, calculating the starting inspection point of the robot, and generating the starting point information of the inspection path;

[0020] S202: Based on the starting point information of the inspection path, considering the inspection efficiency and energy consumption, combined with the inspection coverage rate, calculating the orchard inspection path, and generating inspection parameter information;

[0021] S203: Based on the inspection parameter information, according to the terrain characteristics in the orchard, including ground slope and ground material, combined with the pest and disease detection needs of key fruit tree areas, the speed and stop points are adjusted to optimize energy consumption and inspection efficiency, and the inspection path calculation results are generated.

[0022] As a further solution of the present invention, based on the inspection path calculation result, the fruit tree status image is analyzed, and the key structures in the image are identified, the key growth areas are recorded and the image information is annotated. The steps of generating the annotated information data set are specifically as follows:

[0023] S301: Based on the inspection path calculation result, use an image acquisition device to collect images of multiple fruit trees in the orchard, record visual data including trunks, branches, leaves and fruits, and generate a fruit tree image data set;

[0024] S302: Based on the fruit tree image dataset, analyze and identify key growth areas in the image, identify types of multiple key areas, including trunks, branches, and leaves, and generate growth area annotation data;

[0025] S303: Based on the growth area annotation data, the coordinates and sizes of multiple structures are recorded, and labels and attribute information, including health status and color changes, are matched to generate an annotation information data set.

[0026] As a further solution of the present invention, the physiological parameters of the fruit trees are analyzed according to the labeled information data set, and the health conditions of multiple fruit trees are evaluated by comparing them with the physiological parameters of healthy fruit trees, and abnormal growth patterns are identified. The steps of generating fruit tree health assessment information are specifically as follows:

[0027] S401: Based on the annotation information data set, the chlorophyll content, canopy nitrogen of the fruit tree plant and fruit maturity are calculated according to the color depth and fruit size by using the fruit tree image data, and a fruit tree physiological parameter set is generated;

[0028] S402: Based on the fruit tree physiological parameter set, by analyzing the differences in physiological parameters among multiple fruit trees, identifying fruit trees with abnormal growth status, recording positions and coordinate information, and generating abnormal fruit tree record data;

[0029] S403: Based on the abnormal fruit tree record data, the health status of multiple fruit trees is evaluated, and risk analysis is performed in combination with the pest and disease database to predict the growth trend and health risk of the fruit trees and generate fruit tree health assessment information.

[0030] As a further solution of the present invention, the multivariate linear regression model is according to the formula:

[0031] y=β0+β1C+β2L+β3D+β4W+β5A

[0032] Calculate the chlorophyll content of fruit trees, where β0 is the intercept of the model, which is used to provide the expected output of chlorophyll content without considering any explanatory variables. C is the color depth, which is used to measure the depth of fruit color. It is obtained by analyzing the color histogram of the fruit in the image through image processing software and calculating its average color value. β1 is the weight coefficient of color depth, which is used to adjust the impact of color depth on the prediction results. L is the light intensity, which is used to measure the lighting conditions during image acquisition. It is determined by the brightness distribution statistics in the image. β2 is the weight coefficient of light intensity, which is used to adjust the impact of light intensity on the prediction results. D is the light direction, which is used to represent the light angle during image acquisition. The angle value determined by the shadow and illumination range, β3 is the weight coefficient of the illumination direction, which is used to adjust the influence of the illumination direction on the prediction results, W is the weather condition, which is used to indicate the meteorological conditions when the image is collected, and is classified as sunny, cloudy or rainy. β4 is the weight coefficient of the weather condition, which is used to adjust the influence of the weather condition on the prediction results. A is the leaf area index, which is used to measure the coverage area and density of the leaf surface. It is obtained by analyzing the coverage and density of vegetation in the image. β5 is the weight coefficient of the leaf area index, which is used to adjust the influence of the leaf area index on the prediction results. By predicting the chlorophyll content of fruit trees, the health status of fruit trees and the management of agricultural production can be evaluated, and effective monitoring of the growth trend of the orchard can be achieved.

[0033] As a further solution of the present invention, based on the fruit tree health assessment information, combined with the fruit tree image information, by comparing the known pest and disease feature images, identifying the pest and disease features of the fruit tree, and analyzing the types of pests and diseases according to the abnormal growth pattern, the steps of generating the pest and disease risk prediction results are specifically as follows:

[0034] S501: According to the fruit tree health assessment information, the abnormal growth signals of the fruit trees, including leaf color changes and abnormal fruit morphology, are identified and analyzed through the fruit tree image data to generate feature classification information;

[0035] S502: Based on the feature classification information, identify the pest and disease pattern of the target abnormal fruit tree, including leaf spot disease and fruit rot, and predict the pest and disease development trend to generate a pest and disease pattern recognition result;

[0036] S503: Based on the pest and disease pattern recognition results, evaluate and calculate the likelihood and impact range of various pests and diseases, predict disease risks in combination with the health status of fruit trees, and generate pest and disease risk prediction results.

[0037] As a further solution of the present invention, based on the pest risk prediction results, by analyzing the morphological characteristics of the fruit, evaluating the fruit quality and quality grade, recording the growth and health status information of the fruit tree and matching the management response information, the steps of generating the growth status monitoring information are specifically as follows:

[0038] S601: Based on the pest risk prediction result, analyze the size and color variation in the fruit image, measure the size and color saturation of the fruit, record the key quality parameters of the fruit, including the size, shape, and skin smoothness of the fruit, and generate a record of fruit morphological characteristics;

[0039] S602: Analyze the morphological characteristic information of the fruit according to the fruit morphological characteristic record, classify the fruit according to its quality and calculate the quality grades of multiple fruits to generate a fruit quality evaluation result;

[0040] S603: Based on the fruit quality assessment results and combined with the fruit tree growth status data, natural language processing technology is used to match management response information for the orchard, including irrigation and fertilization time adjustment information, fruit tree harvest time list, and pest and disease control response information, to generate growth status monitoring information.

[0041] A growth situation monitoring system for fruit trees, the growth situation monitoring system for fruit trees is used to execute the above-mentioned growth situation monitoring method for fruit trees, the system comprises:

[0042] The orchard terrain analysis module uses image acquisition equipment to collect image data of the orchard's terrain and vegetation distribution based on the orchard's location information, builds a three-dimensional model, and generates an orchard image data set;

[0043] The inspection path optimization module identifies obstacles and analyzes the terrain based on the orchard image data set, calculates the inspection path in combination with the coverage rate and inspection efficiency, and collects image information of multiple fruit trees to generate inspection parameter optimization data;

[0044] The image data analysis module analyzes the fruit tree image data based on the inspection parameter optimization data, identifies and annotates the trunk, branches, leaves and fruits in the image, and generates a fruit tree annotation data set;

[0045] The fruit tree status recognition module uses the fruit tree annotated data set to analyze the physiological parameters of the fruit trees, including chlorophyll content and fruit maturity, identify abnormal growth patterns, and generate health status information;

[0046] The disease risk assessment module identifies and predicts the risk of disease and pests based on the health status information and historical disease and pest data, evaluates the fruit quality according to the morphological characteristics of the fruit, matches the management response information, and generates growth status monitoring information.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are:

[0048] In the present invention, through image acquisition and three-dimensional model construction, the positions and surrounding environment information of multiple fruit trees can be accurately identified, blind spots can be reduced and resource allocation can be optimized, the inspection path can be optimized, and the utilization rate of human and time resources can be improved. By real-time analysis of fruit tree status images and marking of key growth areas, the monitoring ability of the health status of fruit trees can be improved, the early identification ability of diseases and pests can be enhanced, and the timeliness and effectiveness of prevention and control measures can be improved. By analyzing image information and physiological parameters, the growth status of fruit trees can be comprehensively evaluated, diseases and pests can be identified at an early stage and timely intervention can be made, thereby improving the quality and yield of fruits. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0050] Figure 2 This is a detailed flow chart of S1 of the present invention;

[0051] Figure 3 This is a detailed flow chart of S2 of the present invention;

[0052] Figure 4 This is a detailed flow chart of S3 of the present invention;

[0053] Figure 5 This is a detailed flow chart of S4 of the present invention;

[0054] Figure 6 This is a detailed flow chart of S5 of the present invention;

[0055] Figure 7 This is a detailed flow chart of S6 of the present invention;

[0056] Figure 8 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships 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, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0059] See also Figure 1 The present invention provides a technical solution, a method for monitoring the growth status of fruit trees, comprising the following steps:

[0060] S1: Based on the orchard location information, use image acquisition equipment to collect image information of the orchard, analyze and identify the location information of multiple fruit trees and obstacles in the orchard, and generate a three-dimensional model of the orchard;

[0061] S2: Based on the orchard 3D model, the inspection path is adjusted in combination with obstacle information, the coverage rate and inspection efficiency are considered, the inspection path is calculated, and image information of multiple fruit trees is collected to generate the inspection path calculation result;

[0062] S3: Based on the inspection path calculation results, the fruit tree status image is analyzed, the key structures in the image are identified, the key growth areas are recorded and the image information is annotated to generate an annotation information dataset;

[0063] S4: Analyze the physiological parameters of fruit trees based on the labeled information dataset, evaluate the health status of multiple fruit trees by comparing them with the physiological parameters of healthy fruit trees, identify abnormal growth patterns, and generate fruit tree health assessment information;

[0064] S5: Based on the fruit tree health assessment information and combined with the fruit tree image information, the characteristics of fruit tree diseases and pests are identified by comparing with known disease and pest characteristic images, and the types of diseases and pests are analyzed according to abnormal growth patterns to generate disease and pest risk prediction results;

[0065] S6: Based on the results of pest and disease risk prediction, by analyzing the morphological characteristics of the fruit, evaluating the fruit quality and quality grade, recording the growth and health status information of the fruit trees and matching the management response information, the growth status monitoring information is generated.

[0066] The three-dimensional model of the orchard includes an orchard terrain height map, fruit tree location marking information, and obstacle spatial layout information. The inspection path calculation results include the optimal path map, estimated inspection time, and energy consumption estimation information. The annotation information data set includes trunk diameter and height measurement data, fruit quantity and maturity rating scores, and leaf health indicators. Fruit tree health assessment information includes chlorophyll concentration index, plant canopy nitrogen content information, and fruit size and maturity identification information. Pest and disease risk prediction results include a list of potential pest and disease types, infection possibility rating information, and disease response management matching information. Growth trend monitoring information includes fruit quality classification results, expected harvest, and fruit tree growth cycle analysis results.

[0067] See also Figure 2Based on the orchard location information, the image acquisition equipment is used to collect the image information of the orchard, and the location information of multiple fruit trees and obstacles in the orchard is analyzed and identified. The steps to generate the three-dimensional model of the orchard are as follows:

[0068] S101: Based on the orchard location information, use an image acquisition device to collect image data inside the orchard, record information about the land, trees, and surrounding environment, and generate orchard image data;

[0069] In sub-step S101, based on the orchard location information, image acquisition equipment is used to collect image data inside the orchard, record information about the land, trees, and surrounding environment, and obtain multi-angle visual data through a high-resolution camera system. The data includes texture details of the land, leaf arrangement and color depth of trees, and objects in the surrounding environment, including fences and various vegetation. Each image data is annotated with a timestamp and GPS coordinates to ensure the accuracy and traceability of each data collection. The collected image data undergoes preliminary digital processing, including image cropping, contrast adjustment, and color balance, to improve the accuracy and efficiency of the analysis and generate orchard image data.

[0070] S102: Based on the orchard image data, analyze the image, identify the location information of multiple fruit trees and obstacles, including facilities and vegetation, and generate fruit tree location information;

[0071] In sub-step S102, based on the orchard image data, the image is analyzed, and image recognition technology is used to identify key features in the image, including various fruit trees and surrounding obstacles such as horticultural facilities and other vegetation. Edge detection and pattern recognition algorithms are used to accurately capture the morphology of the fruit trees, including determining the outline of the trees through edge detection and using pattern recognition to distinguish between tree species and other plants or buildings. Each identified object is assigned a unique identifier and its location information is recorded in the image, providing a basis for building a more complete orchard database and generating fruit tree location information.

[0072] S103: Based on the fruit tree location information, a three-dimensional terrain model of the target orchard is constructed, including the heights, volumes, and spatial layout of obstacles of multiple fruit trees, to generate a three-dimensional model of the orchard;

[0073] In sub-step S103, according to the position information of the fruit tree, according to the formula V = π × r 2 ×h, calculate the spatial volume of the fruit tree model and generate a three-dimensional model of the orchard.

[0074] In the formula, r represents half of the average crown diameter of the fruit tree, and h represents the height of the fruit tree. The measurement errors of the crown diameter and height of the fruit tree are taken into account in the calculation process, and the average value of multiple measurements is used for calculation. In addition, in order to more accurately describe the orchard terrain, the potential impact of terrain undulation on fruit tree growth is taken into account. By analyzing the correlation between terrain data and fruit tree growth data, the parameter settings in the three-dimensional model are adjusted to ensure that the model can truly reflect the terrain and vegetation distribution status of the orchard.

[0075] Detailed explanation of the formula and the process of formula calculation and derivation:

[0076] In the process of calculating the spatial volume of the fruit tree model, the crown diameter r is taken as the average of the longest and shortest crown diameters. Assuming that the longest crown diameter of a fruit tree is 4 meters and the shortest crown diameter is 3 meters, then r = (4 + 3) / 2 = 3.5 meters. The height h of the fruit tree is usually represented by the vertical distance from the ground to the highest leaf. Assuming that the measured value is 5 meters, the value is substituted into the volume calculation formula V = π × 3.5 2 ×5 calculation, the spatial volume of the fruit tree model is V = π × 12.25 × 5 ≈ 192.475 cubic meters. The results show that the spatial volume of the fruit tree model obtained by calculation is used for ecological research and orchard management analysis, and provides data support for the orchard three-dimensional model.

[0077] See also Figure 3 ,Based on the orchard 3D model, the inspection path is adjusted in combination with obstacle information, the coverage rate and inspection efficiency are considered, the inspection path is calculated, and the image information of multiple fruit trees is collected. The specific steps for generating the inspection path calculation result are as follows:

[0078] S201: Based on the three-dimensional model of the orchard, analyze the walkable area and obstacles in the orchard, identify the location and size information of multiple obstacles, calculate the starting inspection point of the robot, and generate the starting point information of the inspection path;

[0079] In step S201, based on the orchard 3D model, spatial analysis technology is used to calculate the distance formula. Calculate the shortest path from the starting point to each obstacle and generate the starting point information of the inspection path.

[0080] Where x1, y1 and x2, y2 represent the coordinate positions of the starting point and the obstacle respectively. The calculation process takes into account the size and position of the obstacle to ensure the accuracy of path planning and the safe action of the robot.

[0081] Detailed explanation of the formula and the process of formula calculation and derivation:

[0082] In this calculation process, assuming that the starting point coordinates are (0,0) and the obstacle is located at (3,4), the path length d is: The calculation results help determine the obstacle areas that the robot should avoid, optimize the layout of the inspection route, and ensure efficiency and safety.

[0083] S202: Based on the starting point information of the inspection path, considering the inspection efficiency and energy consumption, combined with the inspection coverage, the orchard inspection path is calculated to generate inspection parameter information;

[0084] In sub-step S202, based on the starting point information of the inspection path, the shortest path algorithm is used in the path planning process to receive distance data from the inspection starting point to each fruit tree, taking into account the robot's battery life, load capacity, and location information of various obstacles in the orchard. The inspection frequency is adjusted according to the type of fruit trees in the orchard, the growth stage, and historical pest and disease records to ensure that risk areas are monitored preferentially and frequently. In path planning, the escape route in an emergency and the robot's recharging path are taken into account to ensure that the robot can complete the task safely and efficiently under any circumstances and generate inspection parameter information.

[0085] S203: Based on the inspection parameter information, according to the terrain characteristics in the orchard, including ground slope and ground material, combined with the pest and disease detection requirements of key fruit tree areas, the speed and stop points are adjusted to optimize energy consumption and inspection efficiency, and the inspection path calculation results are generated;

[0086] In sub-step S203, based on the inspection parameter information, the terrain features are analyzed, including the use of a topographic surveyor and a soil analyzer to measure the ground slope and classify the ground material. The ground slope data helps determine the robot's travel ability and possible travel speed. The ground material data affects the friction coefficient between the robot's tires and the ground, affecting the driving efficiency. The needs of pest and disease detection are considered in key fruit tree areas, and the inspection frequency and dwell time are calculated and adjusted. Energy consumption and inspection efficiency requirements are considered, and a geographic information system and machine learning algorithm are used to calculate the optimal inspection path. The path calculation results affect the inspection robot's operating protocol and power management strategy to ensure coverage of key areas.

[0087] See also Figure 4 ,Based on the inspection path calculation results, the fruit tree status image is analyzed, by identifying the key structures in the image, recording the key growth areas and annotating the image information, the steps to generate the annotated information data set are as follows:

[0088] S301: Based on the inspection path calculation result, an image acquisition device is used to collect images of multiple fruit trees in the orchard, and visual data including trunks, branches, leaves and fruits are recorded to generate a fruit tree image dataset;

[0089] In sub-step S301, based on the inspection path calculation results, an image acquisition device is used to collect images of multiple fruit trees in the orchard, and visual data including trunks, branches, leaves and fruits are recorded to construct a comprehensive fruit tree image dataset. The dataset includes the original pixel information of the image, the timestamp and geographic location information of each image, which is used for image processing and data analysis. During the image data acquisition process, the quality of the acquired images and the consistency of the data are ensured by paying attention to the lighting conditions and the angle of the camera. For multiple parts of the trunks, branches, leaves and fruits, differentiated image acquisition strategies are adopted, including the use of high-resolution cameras to capture details for the trunks and branches, and the use of macro lenses to obtain more comprehensive visual information for leaves and fruits. The environmental information of each sample is recorded, including temperature and humidity, for image processing and analysis.

[0090] S302: Based on the fruit tree image dataset, analyze and identify key growth areas in the image, identify the types of multiple key areas, including trunks, branches, and leaves, and generate growth area annotation data;

[0091] In sub-step S302, based on the fruit tree image dataset, in the process of analyzing and identifying the key growth areas in the image, according to the formula Calculate the average brightness A of the key growth area.

[0092] In the formula, a i represents the brightness value of the i-th pixel, w i Represents the weight of the pixel in the image processing algorithm, and n represents the total number of pixels in the image.

[0093] Detailed explanation of the formula and the process of formula calculation and derivation:

[0094] Assume that there are 5 pixels in the image, and the brightness value and weight of each pixel are as follows: (a1 = 120, w1 = 1.0), (a2 = 135, w2 = 0.8), (a3 = 125, w3 = 1.0), (a4 = 130, w4 =

[0095] 0.7), (a5=140,w5=0.9), substitute the parameters into the formula:

[0096]

[0097] The results represent the average brightness of the key growth areas in the image, indicating the average visual characteristics of the target area and are used for image analysis and growth condition assessment.

[0098] S303: Based on the growth area annotation data, the coordinates and sizes of multiple structures are recorded, and the labels and attribute information, including health status and color changes, are matched to generate an annotation information data set;

[0099] In sub-step S303, based on the growth area annotation data, the coordinates and dimensions of multiple structures are recorded, and the labels and attribute information are matched. The recorded coordinate and dimension data are obtained through a positioning system and a dimension measurement tool to ensure the accuracy and repeatability of the data. For each annotated growth area, the position information and corresponding dimension parameters in the image are recorded, including length, width, and area. The health status and color changes of each area are judged and annotated through an image analysis algorithm, including the use of a color recognition algorithm to identify and record the health status of the leaves, including whether there are yellowing or spots and other pathological characteristics. The data is integrated in the annotation information data set, which includes space, dimension information, health status, and biological attributes, providing a basis for data analysis and decision-making.

[0100] See also Figure 5 ,According to the labeled information dataset, the physiological parameters of the fruit trees are analyzed, and by comparing them with the physiological parameters of healthy fruit trees, the health status of multiple fruit trees is evaluated, and abnormal growth patterns are identified. The specific steps for generating fruit tree health assessment information are as follows:

[0101] S401: Based on the annotated information dataset, the fruit tree image data is used to calculate the chlorophyll content, canopy nitrogen of the fruit tree plant and fruit maturity according to the color depth and fruit size, and generate a fruit tree physiological parameter set;

[0102] The multiple linear regression model is based on the formula:

[0103] y=β0+β1C+β2L+β3D+β4W+β5A

[0104] Calculate the chlorophyll content of fruit trees, where β0 is the intercept of the model, which is used to provide the expected output of chlorophyll content without considering any explanatory variables. C is the color depth, which is used to measure the depth of fruit color. It is obtained by analyzing the color histogram of the fruit in the image through image processing software and calculating its average color value. β1 is the weight coefficient of color depth, which is used to adjust the impact of color depth on the prediction results. L is the light intensity, which is used to measure the lighting conditions during image acquisition. It is determined by the brightness distribution statistics in the image. β2 is the weight coefficient of light intensity, which is used to adjust the impact of light intensity on the prediction results. D is the light direction, which is used to represent the light angle during image acquisition. The angle value determined by the shadow and illumination range, β3 is the weight coefficient of the illumination direction, which is used to adjust the influence of the illumination direction on the prediction results, W is the weather condition, which is used to indicate the meteorological conditions when the image is collected, and is classified as sunny, cloudy or rainy. β4 is the weight coefficient of the weather condition, which is used to adjust the influence of the weather condition on the prediction results. A is the leaf area index, which is used to measure the coverage area and density of the leaf surface. It is obtained by analyzing the coverage and density of vegetation in the image. β5 is the weight coefficient of the leaf area index, which is used to adjust the influence of the leaf area index on the prediction results. By predicting the chlorophyll content of fruit trees, the health status of fruit trees and the management of agricultural production can be evaluated, and effective monitoring of the growth trend of the orchard can be achieved.

[0105] formula:

[0106] y=β0+β1C+β2L+β3D+β4W+β5A

[0107] Detailed explanation of the formula and the process of formula calculation and derivation:

[0108] The physiological parameters of fruit trees were predicted by multiple linear regression models;

[0109] Model parameters and data acquisition instructions:

[0110] C is the color depth, assuming it is 120, which reflects the maturity and health of the fruit.

[0111] L is the light intensity, assumed to be 500 hausk, reflecting the lighting conditions during image acquisition.

[0112] D is the illumination direction, assuming it is 45°, reflecting the illumination angle during image acquisition.

[0113] W is the weather condition. Assuming it is sunny, it is coded as 0, which reflects the weather conditions when the image was collected.

[0114] A is the leaf area index, assuming it is 3.0, which reflects the coverage and density of the leaf surface.

[0115] β0 is the intercept, which is assumed to be 1.0 and provides the expected output of chlorophyll content without considering any explanatory variables.

[0116] β1 is the coefficient of color depth, assuming it is 0.01, which adjusts the effect of color depth on the prediction results.

[0117] β2 is the coefficient of light intensity, assuming it is 0.002, which adjusts the effect of light intensity on the prediction results.

[0118] β3 is the coefficient of the illumination direction, assuming it is 0.05, which adjusts the effect of illumination direction on the prediction results.

[0119] β4 is the coefficient of weather conditions, which is assumed to be 0.3, to adjust the impact of weather conditions on the forecast results.

[0120] β5 is the coefficient of leaf area index, which is assumed to be 0.5, and adjusts the impact of leaf area index on the prediction results.

[0121] Substitute the parameters into the formula for calculation:

[0122] y=1.0+(0.01×120)+(0.002×500)

[0123] +(0.05×45)+(0.3×0)+(0.5×3.0)

[0124] y=1.0+1.2+1.0+2.25+0+1.5

[0125] y=6.95

[0126] The result y=6.95 indicates that under the given conditions, the predicted chlorophyll content is 6.95 units, indicating that under sunny days and moderate light conditions, the chlorophyll content of the fruit trees is good, reflecting a healthy growth state. The numerical value is used to evaluate the nutrient status of the fruit trees and provide a scientific basis for determining the management strategies of the fruit trees, including irrigation, fertilization, and pest and disease control, to optimize orchard management and improve fruit quality.

[0127] S402: Based on the fruit tree physiological parameter set, by analyzing the differences in physiological parameters among multiple fruit trees, identifying fruit trees with abnormal growth status, recording the location and coordinate information, and generating abnormal fruit tree record data;

[0128] In sub-step S402, based on the set of fruit tree physiological parameters, statistical analysis is performed to identify fruit trees with abnormal growth status, and the mean and standard deviation of each physiological parameter are calculated, including statistical data of chlorophyll content and fruit maturity index. An abnormal threshold is set as the mean plus or minus twice the standard deviation to define the normal and abnormal range of fruit tree growth status. Fruit trees that deviate from the threshold are marked, and their location and coordinate information are recorded to form abnormal fruit tree record data. Combined with time series analysis, the changes in physiological parameters across seasons are evaluated to evaluate the health status of each fruit tree and provide data support for orchard management.

[0129] S403: Based on the abnormal fruit tree record data, the health status of multiple fruit trees is evaluated, and risk analysis is performed in combination with the pest and disease database to predict the growth trend and health risk of the fruit trees and generate fruit tree health assessment information;

[0130] In sub-step S403, the abnormal fruit tree record data is used to assess the health status of the marked fruit trees. In combination with the geographic information system, the location of each abnormal fruit tree is marked on the digital map of the orchard. The information in the pest and disease database is used to analyze the pest and disease risks of the abnormal fruit trees, including the types of pests and diseases, associated symptoms and infection cycles. By comparing the database information with the physiological parameters of the fruit trees, the type of disease and health impact of each fruit tree are predicted. Machine learning models, including random forests or neural networks, are applied to analyze abnormal data and identify the growth trends and potential health risks of fruit trees. The generated fruit tree health assessment information guides the farm to take intervention measures, including timely fertilization or spraying of pesticides, and the formulation of long-term management and disease prevention and control strategies.

[0131] See also Figure 6 Based on the fruit tree health assessment information and combined with the fruit tree image information, the characteristics of fruit tree diseases and pests are identified by comparing the known characteristic images of diseases and pests, and the types of diseases and pests are analyzed according to the abnormal growth patterns. The specific steps for generating the disease and pest risk prediction results are as follows:

[0132] S501: Based on the fruit tree health assessment information, the abnormal growth signals of the fruit trees, including leaf color changes and abnormal fruit morphology, are identified and analyzed through the fruit tree image data to generate feature classification information;

[0133] In sub-step S501, based on the fruit tree health assessment information, the fruit tree image data is used to identify and analyze abnormal growth signals of the fruit trees, including the application of image processing technology and pattern recognition algorithms, and feature extraction of leaf color changes and fruit morphology abnormalities in the image, including the use of edge detection and color analysis technology to locate and quantify color and morphology changes. Feature vectors are constructed through image features, and clustering algorithms are applied to classify images into different abnormality types. The generated feature classification information provides visual evidence to support pest and disease identification and management decisions.

[0134] S502: Based on the feature classification information, identify the pest and disease patterns of the target abnormal fruit trees, including leaf spot disease and fruit rot, and predict the pest and disease development trend to generate pest and disease pattern recognition results;

[0135] In sub-step S502, based on the feature classification information, the pest and disease pattern of the target abnormal fruit tree is identified, and multiple linear regression analysis is used to predict the development trend of pests and diseases based on the image features in the feature classification information and the real-time climate and seasonal conditions, using the formula P=α+β1X1+β2X2+β3X3 to generate a pest and disease pattern recognition result.

[0136] Where P represents the probability of the development of pests and diseases, α is the intercept term, β1, β2, and β3 are regression coefficients, X1 represents the characteristic score of leaf spot disease, X2 represents the characteristic score of fruit rot disease, and X3 represents the score of current climate conditions.

[0137] Detailed explanation of the formula and the process of formula calculation and derivation:

[0138] Assuming that the leaf spot score X1 = 0.8, the fruit rot score X2 = 0.5, and the climate condition score X3 = 0.3 are obtained from the feature classification information, and the regression coefficients α = 0.05, β1 = 0.6, β2 = 0.4, and β3 = 0.3 are known, the probability P of the development of pests and diseases is calculated as: P = 0.05 + 0.6 × 0.8 + 0.4 × 0.5 + 0.3 × 0.3 =

[0139] 0.05+0.48+0.2+0.09=0.82 This means that based on the current characteristics and conditions, the predicted probability of pest and disease development is 82%, providing important decision support for orchard management.

[0140] S503: Based on the results of pest and disease pattern recognition, evaluate and calculate the probability and impact range of various pests and diseases, predict the disease risk in combination with the health status of the fruit trees, and generate pest and disease risk prediction results;

[0141] In sub-step S503, based on the results of pest and disease pattern recognition, the possibility and impact range of various pests and diseases are evaluated and calculated, and the disease risk is predicted in combination with the health status of fruit trees, including the application of statistical models and risk assessment algorithms to quantify the types and severity of pests and diseases. Multiple types of data are integrated through an integrated model, including historical disease records, fruit tree physiological parameters, currently identified disease types, and climatic conditions, to predict the development trend and impact range of pests and diseases in the future. The generated pest and disease risk prediction results help orchard managers take targeted prevention and control measures, optimize resource allocation, and improve the overall health management efficiency of the orchard.

[0142] See also Figure 7Based on the results of pest and disease risk prediction, the steps of generating growth status monitoring information are as follows:

[0143] S601: Based on the results of pest and disease risk prediction, analyze the size and color variation in the fruit image, measure the size and color saturation of the fruit, record the key quality parameters of the fruit, including the size, shape, and skin smoothness of the fruit, and generate a record of the fruit morphological characteristics;

[0144] In sub-step S601, based on the results of pest and disease risk prediction, fruit tree image data is used to analyze the size and color variation of the fruit. The image analysis technology is used to measure the size and color saturation of the fruit. Edge detection and color segmentation are used to identify the fruit boundary and color features. The image features are used to calculate the size of the fruit, including diameter, area, and color saturation. By converting the color space and measuring the size, the size data and color data of each fruit are recorded to form the key quality parameters of the fruit, including the size, shape, and skin smoothness of the fruit. The parameters provide a basis for quality assessment and classification. The recorded data is used to construct a record of fruit morphological characteristics. The record reflects the morphological characteristics of each fruit and provides key information for quality control and processing.

[0145] S602: Analyze the morphological characteristic information of the fruit according to the fruit morphological characteristic record, classify the fruit according to its quality and calculate the quality grades of multiple fruits to generate a fruit quality evaluation result;

[0146] In sub-step S602, the morphological characteristic information of the fruit is analyzed according to the fruit morphological characteristic records, and the quality grade of the fruit is calculated according to the size and color saturation of the fruit using a multivariate linear regression model. The formula Q=β0+β1·D+β2·S is applied to generate the fruit quality evaluation result.

[0147] Where Q represents the quality grade of the fruit, β0, β1, and β2 are regression coefficients, D represents the fruit diameter, and S represents the color saturation.

[0148] Detailed explanation of the formula and the process of formula calculation and derivation:

[0149] Assume that the fruit diameter D = 8 cm, the color saturation S = 0.75, and the regression coefficient β0 =

[0150] 1.0, β1=0.05, β2=0.8, the fruit quality grade Q is calculated as: Q=1.0+0.05×8+0.8×0.75=1.0+0.4+0.6=2.0. The calculation results show that based on the evaluation of diameter and color saturation, the fruit quality grade is 2.0, which is used for fruit classification and market positioning.

[0151] S603: Based on the fruit quality evaluation results and combined with the fruit tree growth status data, natural language processing technology is used to match management response information for the orchard, including irrigation and fertilization time adjustment information, fruit tree harvest time list, and pest and disease control response information, to generate growth status monitoring information;

[0152] In sub-step S603, based on the fruit quality assessment results and combined with the fruit tree growth status data, natural language processing technology is used to match management response information for the orchard, including irrigation and fertilization time adjustment information, fruit tree harvest time list, and pest and disease control response information. This includes using text analysis tools to extract key information, such as growth rate and fruit quality, and using pattern recognition algorithms to compare target data with historical management measures to generate growth status monitoring information. The information is used to formulate targeted agricultural management strategies, optimize resource allocation, and provide managers with responsive management information through data visualization tools, providing the orchard with comprehensive growth and health monitoring information.

[0153] See also Figure 8 A growth situation monitoring system for fruit trees is provided. The growth situation monitoring system for fruit trees is used to execute the above-mentioned growth situation monitoring method for fruit trees. The system comprises:

[0154] The orchard terrain analysis module uses image acquisition equipment to collect image data of the orchard's terrain and vegetation distribution based on the orchard's location information, builds a three-dimensional model, and generates an orchard image data set;

[0155] The inspection path optimization module identifies obstacles and analyzes the terrain based on the orchard image dataset, calculates the inspection path based on coverage and inspection efficiency, and collects image information of multiple fruit trees to generate inspection parameter optimization data;

[0156] The image data analysis module analyzes the fruit tree image data based on the inspection parameter optimization data, identifies and annotates the trunks, branches, leaves and fruits in the images, and generates a fruit tree annotation dataset;

[0157] The fruit tree status recognition module uses the fruit tree annotation dataset to analyze the physiological parameters of the fruit trees, including chlorophyll content and fruit maturity, identify abnormal growth patterns, and generate health status information;

[0158] The disease risk assessment module identifies and predicts disease and pest risks based on health status information and historical disease and pest data, evaluates fruit quality based on the morphological characteristics of the fruit, matches management response information, and generates growth status monitoring information.

[0159] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for monitoring the growth status of fruit trees, characterized in that: The following steps are involved: Based on the orchard location information, image acquisition equipment is used to collect image information of the orchard, analyze and identify the location information of multiple fruit trees and obstacles in the orchard, and generate a three-dimensional model of the orchard; Based on the orchard three-dimensional model, the inspection path is adjusted in combination with obstacle information, the coverage rate and inspection efficiency are considered, the inspection path is calculated, and image information of multiple fruit trees is collected to generate an inspection path calculation result; Based on the inspection path calculation results, the fruit tree status image is analyzed, and the key structures in the image are identified, the key growth areas are recorded, and the image information is annotated to generate an annotated information data set; Analyze the physiological parameters of the fruit trees according to the labeled information data set, evaluate the health status of multiple fruit trees by comparing them with the physiological parameters of healthy fruit trees, identify abnormal growth patterns, and generate fruit tree health assessment information; Based on the fruit tree health assessment information, combined with the fruit tree image information, by comparing known pest and disease feature images, identifying the pest and disease features of the fruit trees, and analyzing the types of pests and diseases according to abnormal growth patterns, generating pest and disease risk prediction results; Based on the pest and disease risk prediction results, the morphological characteristics of the fruit are analyzed, the fruit quality and quality grade are evaluated, the growth and health status information of the fruit trees is recorded and matched with the management response information, and the growth status monitoring information is generated.

2. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: The orchard three-dimensional model includes an orchard terrain height map, fruit tree location marking information, and obstacle spatial layout information; the inspection path calculation results include an optimal path map, estimated inspection time, and energy consumption estimation information; the annotation information data set includes trunk diameter and height measurement data, fruit quantity and maturity rating scores, and leaf health indicators; the fruit tree health assessment information includes chlorophyll concentration index, plant canopy nitrogen content information, and fruit size and maturity identification information; the pest and disease risk prediction results include a list of potential pest and disease types, infection possibility rating information, and disease response management matching information; the growth trend monitoring information includes fruit quality classification results, expected harvest, and fruit tree growth cycle analysis results.

3. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: Based on the orchard location information, the image information of the orchard is collected by using image acquisition equipment, and the location information of multiple fruit trees and obstacles in the orchard is analyzed and identified. The steps to generate the 3D model of the orchard are as follows: Based on the orchard location information, use image acquisition equipment to collect image data inside the orchard, record information about the land, trees, and surrounding environment, and generate orchard image data; Based on the orchard image data, analyzing the image, identifying location information of multiple fruit trees and obstacles, including facilities and vegetation, and generating fruit tree location information; Based on the fruit tree location information, a three-dimensional terrain model of the target orchard is constructed, including the heights, volumes, and spatial layout of obstacles of multiple fruit trees, to generate a three-dimensional model of the orchard.

4. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: Based on the orchard three-dimensional model, the inspection path is adjusted in combination with obstacle information, the coverage rate and inspection efficiency are considered, the inspection path is calculated, and image information of multiple fruit trees is collected to generate the inspection path calculation result. Specifically, the steps are as follows: Based on the three-dimensional model of the orchard, the walkable area and obstacles in the orchard are analyzed, the location and size information of multiple obstacles are identified, the starting inspection point of the robot is calculated, and the starting point information of the inspection path is generated; Based on the starting point information of the inspection path, considering the inspection efficiency and energy consumption, combined with the inspection coverage, the orchard inspection path is calculated to generate inspection parameter information; Based on the inspection parameter information, according to the terrain characteristics in the orchard, including ground slope and ground material, combined with the pest and disease detection needs of key fruit tree areas, the speed and stop points are adjusted to optimize energy consumption and inspection efficiency, and the inspection path calculation results are generated.

5. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: Based on the inspection path calculation results, the fruit tree status image is analyzed, and the key structures in the image are identified, the key growth areas are recorded, and the image information is annotated. The steps of generating the annotated information data set are as follows: Based on the inspection path calculation result, an image acquisition device is used to collect images of multiple fruit trees in the orchard, and visual data including trunks, branches, leaves and fruits are recorded to generate a fruit tree image data set; Based on the fruit tree image dataset, analyzing and identifying key growth areas in the image, identifying the types of multiple key areas, including trunks, branches, and leaves, and generating growth area annotation data; Based on the growth area annotation data, the coordinates and dimensions of multiple structures are recorded, and the labels and attribute information, including health status and color changes, are matched to generate an annotation information data set.

6. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: The steps of analyzing the physiological parameters of the fruit trees according to the labeled information data set, comparing the physiological parameters with those of healthy fruit trees, evaluating the health status of multiple fruit trees, identifying abnormal growth patterns, and generating fruit tree health assessment information are as follows: Based on the labeled information data set, the chlorophyll content, canopy nitrogen of the fruit tree plant and fruit maturity are calculated according to the color depth and fruit size by using the fruit tree image data and a multivariate linear regression method to generate a fruit tree physiological parameter set; Based on the fruit tree physiological parameter set, by analyzing the differences in physiological parameters among multiple fruit trees, fruit trees with abnormal growth status are identified, their positions and coordinate information are recorded, and abnormal fruit tree record data is generated; Based on the abnormal fruit tree record data, the health status of multiple fruit trees is evaluated, and risk analysis is performed in combination with the pest and disease database to predict the growth trend and health risk of the fruit trees and generate fruit tree health assessment information.

7. The method for monitoring the growth status of fruit trees according to claim 6, characterized in that: The multivariate linear regression model is based on the formula: y=β0+β1C+β2L+β3D+β4W+β5A Calculate the chlorophyll content of fruit trees, where β0 is the intercept of the model, which is used to provide the expected output of chlorophyll content without considering any explanatory variables. C is the color depth, which is used to measure the depth of fruit color. It is obtained by analyzing the color histogram of the fruit in the image through image processing software and calculating its average color value. β1 is the weight coefficient of color depth, which is used to adjust the impact of color depth on the prediction results. L is the light intensity, which is used to measure the lighting conditions during image acquisition. It is determined by the brightness distribution statistics in the image. β2 is the weight coefficient of light intensity, which is used to adjust the impact of light intensity on the prediction results. D is the light direction, which is used to represent the light angle during image acquisition. The angle value determined by the shadow and illumination range, β3 is the weight coefficient of the illumination direction, which is used to adjust the influence of the illumination direction on the prediction results, W is the weather condition, which is used to indicate the meteorological conditions when the image is collected, and is classified as sunny, cloudy or rainy. β4 is the weight coefficient of the weather condition, which is used to adjust the influence of the weather condition on the prediction results. A is the leaf area index, which is used to measure the coverage area and density of the leaf surface. It is obtained by analyzing the coverage and density of vegetation in the image. β5 is the weight coefficient of the leaf area index, which is used to adjust the influence of the leaf area index on the prediction results. By predicting the chlorophyll content of fruit trees, the health status of fruit trees and the management of agricultural production can be evaluated, and effective monitoring of the growth trend of the orchard can be achieved.

8. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: Based on the fruit tree health assessment information, combined with the fruit tree image information, by comparing the known pest and disease feature images, identifying the pest and disease features of the fruit trees, and analyzing the types of pests and diseases according to the abnormal growth patterns, the steps of generating the pest and disease risk prediction results are specifically as follows: According to the fruit tree health assessment information, the abnormal growth signals of the fruit trees, including leaf color changes and abnormal fruit morphology, are identified and analyzed through the fruit tree image data to generate feature classification information; Based on the feature classification information, identify the pest and disease patterns of the target abnormal fruit trees, including leaf spot disease and fruit rot, and predict the pest and disease development trend to generate pest and disease pattern recognition results; Based on the pest and disease pattern recognition results, the likelihood and impact range of various pests and diseases are evaluated and calculated, and the disease risk is predicted in combination with the health status of the fruit trees to generate a pest and disease risk prediction result.

9. The method for monitoring the growth status of fruit trees according to claim 1, characterized in that: Based on the pest and disease risk prediction results, by analyzing the morphological characteristics of the fruit, evaluating the fruit quality and quality grade, recording the growth and health status information of the fruit tree and matching the management response information, the steps of generating the growth status monitoring information are as follows: Based on the pest and disease risk prediction results, analyzing the size and color variation in the fruit image, measuring the size and color saturation of the fruit, recording the key quality parameters of the fruit, including the size, shape, and skin smoothness of the fruit, and generating a record of the fruit morphological characteristics; Analyzing the morphological characteristic information of the fruits according to the fruit morphological characteristic records, classifying the fruits according to their quality and calculating the quality grades of multiple fruits to generate a fruit quality evaluation result; Based on the fruit quality assessment results and combined with the fruit tree growth status data, natural language processing technology is used to match management response information for the orchard, including irrigation and fertilization time adjustment information, fruit tree harvest time list, and pest and disease control response information, to generate growth status monitoring information.

10. A growth status monitoring system for fruit trees, characterized in that: According to the method for monitoring the growth status of fruit trees according to any one of claims 1 to 9, the system comprises: The orchard terrain analysis module uses image acquisition equipment to collect image data of the orchard's terrain and vegetation distribution based on the orchard's location information, builds a three-dimensional model, and generates an orchard image data set; The inspection path optimization module identifies obstacles and analyzes the terrain based on the orchard image data set, calculates the inspection path in combination with the coverage rate and inspection efficiency, and collects image information of multiple fruit trees to generate inspection parameter optimization data; The image data analysis module analyzes the fruit tree image data based on the inspection parameter optimization data, identifies and annotates the trunk, branches, leaves and fruits in the image, and generates a fruit tree annotation data set; The fruit tree status recognition module uses the fruit tree annotated data set to analyze the physiological parameters of the fruit trees, including chlorophyll content and fruit maturity, identify abnormal growth patterns, and generate health status information; The disease risk assessment module identifies and predicts the risk of disease and pests based on the health status information and historical disease and pest data, evaluates the fruit quality according to the morphological characteristics of the fruit, matches the management response information, and generates growth status monitoring information.

Images