Intelligent tea garden service collaborative management system

Abstract

The present application relates to the technical field of intelligent tea garden management, and proposes an intelligent tea garden service collaborative management system, comprising: a tourist picking reservation module, used for receiving a reservation picking request of a tourist through a terminal, intelligently arranging a time and an area for the tourist to pick tea leaves independently; a picking behavior processing module, used for analyzing, processing and prompting a picking behavior of the tourist; a comprehensive management docking module, used for integrating all module data, generating a multi-dimensional report, and docking with a tea garden enterprise ERP system to generate a collaborative scheme for management personnel, planting personnel and customer service personnel, and completing data integration-role adaptation-decision landing collaborative management. By constructing a full-chain tea garden management system from data collection and integration, growth and variety analysis to visual decision-making, and with the help of multi-source data fusion and intelligent technology, the comprehensive improvement of yield increase, quality improvement, cost reduction and efficiency increase of the tea garden is realized, which not only enhances economic benefits, but also improves resource utilization efficiency and ecological sustainability.

Description

Technical Field

[0001] This invention relates to the field of smart tea garden management technology, and in particular to a smart tea garden service collaborative management system. Background Technology

[0002] While current smart tea garden technology has initially incorporated intelligent elements, it is still in the basic application stage. Most systems are only equipped with basic intelligent devices such as intelligent irrigation, pest and disease monitoring, and environmental data collection. The core functions are concentrated on basic management of tea tree planting, such as generating simple irrigation strategies based on soil moisture and light intensity, or using sensors to warn of pest and disease risks. These are sufficient to meet the basic needs of tea tree planting in smart tea gardens. For example, a tea garden tea picking location planning method disclosed in CN118195111B improves picking efficiency by updating the tea garden model and determining whether the tea trees at the current location have been picked based on the current location. This is also a common implementation method in current smart tea gardens.

[0003] With the rise of experiential consumption and personalized demands, more and more consumers are inclined to visit tea gardens to pick fresh tea leaves themselves and participate in the subsequent processing, aiming for a complete "from tea garden to teacup" experience. However, because tourists often use incorrect picking methods, it can significantly damage the tea trees, negatively impacting both the quality and quantity of tea produced. Furthermore, the inability to analyze varietal popularity in conjunction with tourist preferences prevents tea gardens from adjusting their planting structure and marketing strategies accordingly. This fails to meet consumers' experiential needs for self-picking and processing, and hinders the coordinated development of "tourist experience - tea garden management - market demand," creating a significant technological gap. Summary of the Invention

[0004] In view of the problems existing in the prior art, the purpose of this invention is to provide a smart tea garden service collaborative management system to solve the problems mentioned in the background art.

[0005] To solve the above problems, the present invention adopts the following technical solution: a smart tea garden service collaborative management system, comprising:

[0006] The visitor picking reservation module is used to receive visitor reservation requests for picking through the terminal and intelligently arrange the time and area for visitors to pick their own fruit.

[0007] The picking behavior processing module is used to analyze, process, and provide prompts for tourists' picking behavior;

[0008] The integrated management module is used to integrate data from all modules, generate multi-dimensional reports, and interface with the tea garden enterprise ERP system. It generates collaborative solutions for managers, planters, and customer service personnel, and completes collaborative management of data integration, role adaptation, and decision implementation.

[0009] The variety picking analysis module analyzes the tea plantation's tea growth and the actual popularity of varieties in the current year based on users' pre-picking requests and actual picking behavior.

[0010] The planting management optimization module, based on the tea plantation's growth status and the actual popularity of the varieties in the current year, and by collecting data on harvesting density, formulates current tea plantation management plans and planting plans for the vacant areas of the tea plantation for the following year.

[0011] Preferably, when a tourist submits a reservation request for tea picking through the terminal, they need to upload their identity information and the type of tea to be picked. The identity information includes the tourist's name and phone number. After receiving the reservation request, the tourist reservation module arranges the tourist's self-picking time interval, area, and route plan based on the current tea garden's reservation information, and sends it to the tourist's terminal. After the tourist confirms the specific picking time, the reservation request is sent to the customer service terminal one day before the picking time. The customer service personnel then contact and remind the tourist to enter the tea garden for picking, and send the customer a video of standard picking procedures and precautions.

[0012] Preferably, the picking behavior processing module analyzes, processes, and provides prompts for tourists' picking behavior, including the following steps:

[0013] S21. When tourists enter the tea garden to pick tea leaves, the behavior of tourists picking tea leaves is collected through smart devices and sensors in the tea garden, and the behavior data after preliminary processing is uploaded through 5G or Internet of Things.

[0014] S22. Extract key features from the behavioral data after preliminary processing. Based on the standard tea picking actions, input the key features into the trained AI model to screen for abnormal picking actions by tourists.

[0015] S23. Classify abnormal picking actions, trigger prompts in real time, and correct tourists' picking actions. The prompts include, but are not limited to, vibration prompts, voice prompts, and online guidance.

[0016] S24. Summarize the abnormal picking actions of tourists and the triggered prompts, analyze the effectiveness of different prompt methods, generate the abnormal picking actions that tourists are most likely to cause, generate picking precautions, and update them regularly.

[0017] Preferably, in step S21, collecting and initially processing tourists' picking behavior includes the following steps:

[0018] S211. The picking actions of tourists are collected through intelligent picking devices worn by tourists during picking. The intelligent devices include intelligent picking gloves, intelligent picking baskets, and intelligent picking scissors. The intelligent picking gloves are equipped with pressure sensors, replaceable miniature batteries, vibration motors, miniature cameras, and wireless transceivers to collect the force exerted by the tourist's hand during picking, the picking scene, and the transmission and reception of data signals. The handle of the intelligent picking basket and the blade of the intelligent picking scissors are equipped with six-axis motion sensors, GPS locators, wireless transceivers, and replaceable miniature batteries to collect angular velocity and acceleration, picking position, and transmit data signals during picking. The bottom of the intelligent picking basket is also equipped with weight sensors, data storage, voice broadcasters, and intercoms to count the amount of picking, temporarily store data, intelligently broadcast voice messages, and conduct voice calls. The handle of the intelligent picking scissors is equipped with pressure sensors to collect the cutting force.

[0019] S212. Link the intelligent picking equipment with a camera array deployed in the tea garden at 30m×30m intervals and pressure sensors deployed every 20m in the soil of the picking area. Use edge computing nodes to calibrate the time and location of tourists in the tea garden. After tourists pick, the intelligent picking equipment starts data collection. The data collection includes tourists' hand movements, picking amount, GPS positioning, and hand movement images during picking. The cameras in the tea garden collect picking images for assistance. The pressure sensors in the soil collect the degree of compaction in the area. The collected data is uploaded to the management system of the service center through 5G IoT. When both 5G and IoT signals are lost, it automatically switches to sending data to the data storage. Data is retransmitted after the signal is restored.

[0020] S213. After receiving the data signal, the system preprocesses the data signal, filters out invalid shaking caused by the intelligent harvesting equipment, calibrates the zero error of the sensor and corrects the GPS positioning drift, and backs up and uploads the processed data.

[0021] Preferably, in step S22, screening for abnormal picking behavior by tourists includes the following steps:

[0022] S221. After receiving the preprocessed data, extract key features and build an AI model. Put the key features into the AI ​​model to screen abnormal picking actions and analyze the comprehensive feature value of abnormal picking actions. The key features include the hand force curve, action trajectory coordinates, picking position height, and the frequency of cutting and folding actions.

[0023] S222. When abnormal picking actions are detected, the video footage 1 second before and after the abnormal picking action is selected from the collected video and analyzed frame by frame to further determine the accuracy of the abnormal picking action selection, verify the type of abnormal picking action, analyze the correlation between abnormal actions and tourist characteristics, establish a correlation analysis model, and generate a feature report of tourist picking actions, wherein the tourist characteristics include picking experience and preference for tea varieties.

[0024] S223. Store and upload the identified abnormal action types, frequencies, and correlation analysis results. Use an iterative optimization model to optimize the parameters of the AI ​​model for screening abnormal picking actions to reduce the false judgment rate.

[0025] Preferably, in step S223, prompting and correcting abnormal picking actions of tourists includes: configuring corresponding prompting rules based on the comprehensive feature value of abnormal picking actions obtained in steps S221 and S222 and the video footage of confirmed abnormal actions; after the prompt is triggered, collecting data on the tourist's adjustment actions through the intelligent picking device, and observing the tourist's behavior through the video footage to determine whether the prompt is responded to.

[0026] Preferably, the integrated management interface module performs integrated collaborative management including the following steps:

[0027] S31. Obtain visitor picking reservations, abnormal picking behaviors, and video action analysis, and perform data cleaning, standardization, and structuring to output pre-processed multi-module unified format data;

[0028] S32. Based on the preprocessed unified format data, generate multi-dimensional visualization reports by calling the report template engine according to the time dimension, behavior dimension, and role dimension.

[0029] S33. Convert the generated multidimensional report data into a format and transmit it in real time. Automatically generate a list of fresh tea leaves for processing in the next stage, and upload the details of picking and processing costs to the financial terminal for revenue accounting and cost amortization. At the same time, trigger the calculation of materials required for maintaining tea trees.

[0030] S34. Based on the transformed multidimensional report data, generate a three-way collaboration plan for managers, planting personnel, and customer service personnel, and perform task tracking and feedback optimization.

[0031] Preferably, in step S34, task tracking and feedback optimization are performed by collecting execution data between each role in real time, comparing the execution data with the expected goals of the collaborative solution, generating an execution deviation dataset, locating the deviation links, collecting suggestions from each role on improving the collaborative solution through the feedback entry point, and sending the execution deviation dataset back to step S31 to update the feature dimensions and quality of the preprocessed dataset, providing a basis for the next round of report generation and solution optimization.

[0032] Preferably, the variety harvesting analysis module analyzes the tea garden's tea conditions using the following steps:

[0033] S41. Collect data on tourist reservation requests and actual picking behavior, and integrate the data.

[0034] S42. Based on the integrated data, analyze the growth of tea trees from three aspects: growth progress, yield potential, and regional consistency.

[0035] S43. Quantitatively analyze the popularity of tea varieties in actual tea gardens in the current year from three dimensions: reservation fulfillment, popularity trend, and tourist preferences.

[0036] S44. Based on the analysis results, output an annual visualization report for the tea garden. The annual visualization report for the tea garden includes a heat map of the growth and health of each tea variety, an annual ranking of the popularity of tea varieties, and a report on variety planting and promotion strategies, which are used to support the implementation of tea garden planting structure adjustment and marketing decisions.

[0037] Preferably, in step S41, the data integration uses tourist identity information, reservation date, and reserved picking variety as association keys to match and merge reservation request data with actual picking behavior data, and marks them with tags for non-fulfillment, variety deviation, and regional deviation, ultimately forming a structured dataset of the correlation between tourist reservations and actual picking.

[0038] The intelligent tea garden service collaborative management system provided by this invention has the following beneficial effects:

[0039] 1. By constructing a full-chain tea garden management system that integrates data collection and integration, growth and variety analysis, and visualized decision-making, and by leveraging multi-source data fusion and intelligent technology, we can achieve comprehensive improvement in tea garden production, quality, cost reduction, and efficiency, thereby enhancing economic benefits, resource utilization efficiency, and ecological sustainability.

[0040] 2. By collecting and accurately preprocessing picking behavior data through multi-device linkage, a reliable data foundation is provided for the subsequent accurate screening of abnormal picking actions by tourists. This avoids the impact of data deviation on the judgment of abnormal actions. By establishing an AI model to calculate the comprehensive feature value of picking actions, abnormal picking actions are screened and the standardization of actions is classified. Then, the accuracy of the screened abnormal actions is verified by analyzing the preceding and following videos frame by frame. A model is established to explore its correlation with tourist characteristics and generate a report. Finally, the relevant data is stored and the AI ​​model weights are updated through iterative optimization to reduce the false judgment rate. This multi-stage verification and model optimization can more accurately identify abnormal picking actions by tourists and reduce the misjudgment of normal picking actions.

[0041] 3. By first acquiring and cleaning, standardizing, and structuring data on tourist picking reservations, abnormal picking behaviors, and video action analysis, and then generating multi-dimensional visual reports based on time, business, and role dimensions based on the pre-processed data, the report data format is converted and transmitted in real time to automatically generate a tea leaf processing list, synchronize financial accounting, and trigger material calculations. Finally, a three-party collaborative solution is generated based on the converted data, and task execution and feedback optimization are tracked. This achieves full-process data processing and multi-stage linkage, breaks down the barriers between tea garden business data and management decision-making and execution stages, avoids information gaps, makes operational decisions more aligned with actual needs, and improves management efficiency.

[0042] 4. By first collecting and integrating data on tourist reservation requests for tea picking and actual picking behavior, and then analyzing the growth of tea trees from the dimensions of growth progress, yield potential, and regional consistency based on the integrated data, the popularity of varieties is quantified from the dimensions of reservation fulfillment, popularity trend, and user preference. Finally, based on the analysis results, an annual visual report of the tea garden is generated to support the adjustment of planting structure and marketing decisions. This achieves the combination of tea tree growth status analysis and market demand preference mining, avoiding the blindness of tea gardens planning planting or promoting varieties based solely on experience. It makes planting adjustments and marketing strategies more in line with actual growth conditions and market demand, helping tea gardens achieve supply and demand matching. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A schematic diagram of the system modules of a smart tea garden service collaborative management system provided in this application;

[0045] Figure 2 This is a schematic diagram of the harvesting behavior processing flow of a smart tea garden service collaborative management system provided in this application. Detailed Implementation

[0046] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0047] like Figures 1-2 As shown, this embodiment proposes a smart tea garden service collaborative management system, characterized by comprising:

[0048] The visitor picking reservation module is used to receive visitor reservation requests for picking through the terminal and intelligently arrange the time and area for visitors to pick their own fruit.

[0049] The picking behavior processing module is used to analyze, process, and provide prompts for tourists' picking behavior;

[0050] The integrated management module is used to integrate data from all modules, generate multi-dimensional reports, and interface with the tea garden enterprise ERP system. It generates collaborative solutions for managers, planters, and customer service personnel, and completes collaborative management of data integration, role adaptation, and decision implementation.

[0051] The variety picking analysis module analyzes the tea plantation's tea growth and the actual popularity of varieties in the current year based on users' pre-picking requests and actual picking behavior.

[0052] The planting management optimization module, based on the tea plantation's growth status and the actual popularity of the varieties in the current year, and by collecting data on harvesting density, formulates current tea plantation management plans and planting plans for the vacant areas of the tea plantation for the following year.

[0053] Furthermore, when tourists submit a reservation request for tea picking through the terminal, they need to upload their identity information and the type of tea they wish to pick. The identity information includes the tourist's name and phone number. After receiving the reservation request, the tourist reservation module arranges the tourist's self-picking time range, area, and route plan based on the current tea garden's reservation information and sends it to the tourist's terminal. After the tourist confirms the specific picking time, the reservation request is sent to the customer service terminal one day before the picking time. The customer service personnel then contact and remind the tourist to enter the tea garden for picking and send the customer a video showing the standard picking procedures and precautions.

[0054] Furthermore, the picking behavior processing module analyzes, processes, and provides prompts for tourists' picking behavior, including the following steps:

[0055] S21. When tourists enter the tea garden to pick tea leaves, the behavior of tourists picking tea leaves is collected through smart devices and sensors in the tea garden, and the behavior data after preliminary processing is uploaded through 5G or Internet of Things.

[0056] S22. Extract key features from the behavioral data after preliminary processing. Based on the standard tea picking actions, input the key features into the trained AI model to screen for abnormal picking actions by tourists.

[0057] S23. Classify abnormal picking actions, trigger prompts in real time, and correct tourists' picking actions. The prompts include, but are not limited to, vibration prompts, voice prompts, and online guidance.

[0058] S24. Summarize the abnormal picking actions of tourists and the triggered prompts, analyze the effectiveness of different prompt methods, generate the abnormal picking actions that tourists are most likely to cause, generate picking precautions, and update them regularly.

[0059] Furthermore, by constructing a full-chain tea garden management system that integrates data collection and integration, growth and variety analysis, and visualized decision-making, and by leveraging multi-source data fusion and intelligent technology, we can achieve comprehensive improvements in tea garden production, quality, cost reduction, and efficiency, thereby enhancing economic benefits, resource utilization efficiency, and ecological sustainability.

[0060] Furthermore, in step S21, collecting and initially processing information on tourists' picking behavior includes the following steps:

[0061] S211. The picking actions of tourists are collected through intelligent picking devices worn by tourists during picking. The intelligent devices include intelligent picking gloves, intelligent picking baskets, and intelligent picking scissors. The intelligent picking gloves are equipped with pressure sensors, replaceable miniature batteries, vibration motors, miniature cameras, and wireless transceivers to collect the force exerted by the tourist's hand during picking, the picking scene, and the transmission and reception of data signals. The handle of the intelligent picking basket and the blade of the intelligent picking scissors are equipped with six-axis motion sensors, GPS locators, wireless transceivers, and replaceable miniature batteries to collect angular velocity and acceleration, picking position, and transmit data signals during picking. The bottom of the intelligent picking basket is also equipped with weight sensors, data storage, voice broadcasters, and intercoms to count the amount of picking, temporarily store data, intelligently broadcast voice messages, and conduct voice calls. The handle of the intelligent picking scissors is equipped with pressure sensors to collect the cutting force.

[0062] S212. Link the intelligent picking equipment with a camera array deployed in the tea garden at 30m×30m intervals and pressure sensors deployed every 20m in the soil of the picking area. Use edge computing nodes to calibrate the time and location of tourists in the tea garden. After tourists pick, the intelligent picking equipment starts data collection. The data collection includes tourists' hand movements, picking amount, GPS positioning, and hand movement images during picking. The cameras in the tea garden collect picking images for assistance. The pressure sensors in the soil collect the degree of compaction in the area. The collected data is uploaded to the management system of the service center through 5G IoT. When both 5G and IoT signals are lost, it automatically switches to sending data to the data storage. Data is retransmitted after the signal is restored.

[0063] S213. After receiving the data signal, the system preprocesses the data signal, filters out invalid shaking caused by the intelligent harvesting equipment, calibrates the zero error of the sensor and corrects the GPS positioning drift, and backs up and uploads the processed data.

[0064] Furthermore, by using multiple devices to collect and accurately preprocess picking behavior data, a reliable data foundation is provided for accurately screening abnormal picking actions by tourists, avoiding the impact of data deviation on the judgment of abnormal actions.

[0065] Furthermore, in step S22, screening for abnormal picking actions by tourists includes the following steps:

[0066] S221. After receiving the preprocessed data, extract key features and build an AI model. Put the key features into the AI ​​model to screen abnormal picking actions and analyze the comprehensive feature value of abnormal picking actions. The key features include the hand force curve, action trajectory coordinates, picking position height, and the frequency of cutting and folding actions.

[0067] Specifically, the formula for the AI ​​model used to screen for abnormal picking actions is: In the formula, for The comprehensive characteristic value of tourists' picking actions at any given moment; the higher the value, the greater the deviation of the action from the standard. This is the standard picking action, when The picking action was slightly abnormal. This was a severely abnormal picking action. for The characteristic value of the actual force exerted by the hand during harvesting, and , , for The pressure value after constant adjustment To improve the grip quality of the smart harvesting scissors, for The cutting and closing acceleration of the intelligent harvesting scissors at all times. It is the acceleration due to gravity. for The angle between the grip part of the smart harvesting scissors and the vertical direction. The motion trajectory feature values ​​of the intelligent harvesting scissors. , and They are respectively The tip of the smart harvesting scissors is always in , and Relative displacement in three directions, The height characteristic value of the picking location, for The height at which the fruit is picked at all times. To establish a safe harvesting height threshold for tea tree canopies, The feature values ​​are the frequency of shearing and folding actions. and for The number of cuts and folds at any given moment. To calculate the duration of a time interval, , , and These are learnable weights, summing to 1, with an initial value of [value missing]. , , , And among them It is always the maximum value among all weights;

[0068] S222. When abnormal picking actions are detected, the video footage 1 second before and after the abnormal picking action is selected from the collected video and analyzed frame by frame to further determine the accuracy of the abnormal picking action selection, verify the type of abnormal picking action, analyze the correlation between abnormal actions and tourist characteristics, establish a correlation analysis model, and generate a feature report of tourist picking actions, wherein the tourist characteristics include picking experience and preference for tea varieties.

[0069] Specifically, the formula for the correlation analysis model is: In the formula, For the results of the correlation analysis, A mutual information model for tourist characteristics and unusual behaviors. For tourist characteristics, based on picking experience And preferences for tea varieties Composition, harvesting experience And preferences for tea varieties The values ​​of are all in the range of [0-1], and , for and The joint probability, and Marginal probability;

[0070] S223. Store and upload the identified abnormal action types, frequencies, and correlation analysis results. Use an iterative optimization model to optimize the parameters of the AI ​​model for screening abnormal picking actions to reduce the false judgment rate.

[0071] Specifically, the iterative optimization model optimizes the parameters of the AI ​​model for screening abnormal picking actions by optimizing the four learnable weights in the AI ​​model formula. It calculates the average contribution of each feature value based on the misjudged samples and updates the four learnable weights in the model. The formula for the iterative optimization model is: In the formula, and These are the learnable weights before and after the update. For learning rate, For the misjudged sample The average of the features, It is the sum of the averages of the four characteristics. For the misjudged sample number The proportion of each feature.

[0072] Furthermore, in step S223, prompting and correcting abnormal picking actions of tourists includes: configuring corresponding prompting rules based on the comprehensive feature value of abnormal picking actions obtained in steps S221 and S222 and the video footage of confirmed abnormal actions; after the prompt is triggered, collecting data on the tourist's adjustment actions through the intelligent picking device, and observing the tourist's behavior through the video footage to determine whether the prompt is responded to.

[0073] Specifically, the prompting rules include four levels, level one: when Furthermore, when a tourist makes an abnormal picking action for the first time, the vibration motor in the smart picking glove is triggered to vibrate briefly for 0.5 seconds, Level Two: When Furthermore, if a tourist repeatedly makes abnormal picking actions, a long vibration is triggered, lasting 1 second, level three: when Furthermore, when a tourist makes an abnormal picking action for the first time, it triggers two consecutive long vibrations, each lasting 1 second, and simultaneously broadcasts the normal picking procedure for this abnormal picking action through the voice broadcaster in the smart picking basket. Level 4: When Furthermore, when tourists repeatedly perform abnormal picking actions, two consecutive long vibrations are triggered, each lasting 1 second, and a trigger prompt is sent to the customer service terminal, whereby the customer service personnel provide remote guidance on proper picking procedures.

[0074] Furthermore, by establishing an AI model to calculate the comprehensive feature value of picking actions, abnormal picking actions are screened and the standardization of the actions is classified. Then, the accuracy of the screened abnormal actions is verified by analyzing the preceding and following videos frame by frame. A model is established to explore its correlation with tourist characteristics and generate a report. Finally, the relevant data is stored and the AI ​​model weights are updated through iterative optimization to reduce the false judgment rate. This multi-stage verification and model optimization can more accurately identify abnormal picking actions of tourists, reduce the misjudgment of normal picking actions, and provide an accurate basis for subsequent targeted prompts and corrections.

[0075] Furthermore, the integrated management interface module performs integrated collaborative management in the following steps:

[0076] S31. Obtain visitor picking reservations, abnormal picking behaviors, and video action analysis, and perform data cleaning, standardization, and structuring to output pre-processed multi-module unified format data;

[0077] S32. Based on the preprocessed unified format data, generate multi-dimensional visualization reports by calling the report template engine according to the time dimension, behavior dimension, and role dimension.

[0078] Specifically, the time dimension generates a "Daily Report on Tourist Picking Behavior," which statistically analyzes the daily picking volume in each area and the tea varieties preferred by tourists that year. This is used to analyze the quality of tea leaves in tea gardens and the market's preference for tea varieties that year. The business dimension generates a "Report on Analysis of Abnormal Tourist Picking Actions," which includes the distribution of abnormal picking action types. This is used to statistically analyze abnormal picking actions frequently triggered by tourists, making it easier to educate tourists on proper picking actions and common mistakes before they pick tea, thus preventing damage to tea trees from repeated incorrect picking. The role dimension generates a "Overview of Tea Garden Operation Monitoring" and a "List of Frequently Asked Questions by Tourists," which statistically analyze tea-growing areas with frequent abnormal picking, facilitating necessary maintenance of tea trees in these areas by growers. At the same time, it summarizes the questions and standard answers asked by tourists regarding abnormal picking actions.

[0079] S33. Convert the generated multidimensional report data into a format and transmit it in real time. Automatically generate a list of fresh tea leaves for processing in the next stage, and upload the details of picking and processing costs to the financial terminal for revenue accounting and cost amortization. At the same time, trigger the calculation of materials required for maintaining tea trees.

[0080] S34. Based on the transformed multidimensional report data, generate a three-way collaboration plan for managers, planting personnel, and customer service personnel, and perform task tracking and feedback optimization.

[0081] Specifically, the tripartite collaboration solution includes a tea garden operation optimization decision-making scheme for managers, a refined field operation execution scheme for growers, and a scenario-based service collaboration scheme for tourists for customer service personnel. The tea garden operation optimization decision-making scheme includes dynamic resource allocation suggestions and medium- to long-term strategy adjustments, and is equipped with an ROI model simulation and prediction model to assist managers in making quick decisions. The formula for the supporting ROI model is: In the formula, A comprehensive decision-making index for tea garden operations. This refers to the number of smart harvesting tools deployed in the current cycle. The elasticity coefficient for tool deployment. This represents the tea harvest volume for the target region in the current cycle. This represents the tea harvest volume in the target area for the previous cycle. For the validity period of organic certification, The estimated annual yield of tea varieties, Net profit per unit of product before certification The premium rate after certification. To account for the total cost of organic certification, the field-specific operational execution plan is used to refine the daily field operation tasks of growers, including but not limited to mechanical soil loosening plans for different tea tree areas, plans to activate the intelligent irrigation system in the area, intelligent mechanical fertilization plans, and plans for key maintenance points of tea trees. The visitor scenario-based service collaboration plan is used to provide response scripts and guidance strategies for different scenarios, as well as explanations for correcting abnormal picking actions and suggestions for tea tasting in the area. At the same time, it summarizes frequently asked questions and provides standard answers and visual video materials.

[0082] Furthermore, in step S34, task tracking and feedback optimization are performed by collecting execution data between each role in real time, comparing the execution data with the expected goals of the collaborative solution, generating an execution deviation dataset, locating the deviation links, collecting suggestions from each role on improving the collaborative solution through the feedback entry point, and sending the execution deviation dataset back to step S31 to update the feature dimensions and quality of the preprocessed dataset, providing a basis for the next round of report generation and solution optimization.

[0083] Furthermore, by first acquiring and cleaning, standardizing, and structuring data on tourist picking reservations, abnormal picking behaviors, and video action analysis, and then generating multi-dimensional visual reports based on time, business, and role dimensions based on the pre-processed data, the report data format is converted and transmitted in real time to automatically generate a tea leaf processing list, synchronize financial accounting, and trigger material calculations. Finally, a three-party collaborative solution is generated based on the converted data, and task execution and feedback optimization are tracked. This achieves full-process data processing and multi-stage linkage, breaks down the barriers between tea garden business data and management decision-making and execution stages, avoids information gaps, makes operational decisions more aligned with actual needs, and improves management efficiency.

[0084] Furthermore, the variety harvesting analysis module analyzes the tea garden's tea conditions using the following steps:

[0085] S41. Collect data on tourist reservation requests and actual picking behavior, and integrate the data.

[0086] S42. Based on the integrated data, analyze the growth of tea trees from three aspects: growth progress, yield potential, and regional consistency.

[0087] Specifically, the growth progress analysis is based on the actual number of tea leaves picked, combined with the standard growth cycle of the variety, to calculate the deviation between the actual number of growing days and the theoretical cycle. The yield potential is first calculated using tea garden planting area, plant density, and historical highest yield data to determine the theoretical maximum harvestable yield for each variety and region. Then, the actual harvested yield is statistically analyzed to calculate the harvest achievement rate. Regional consistency is analyzed using a variance model based on the growth deviation and harvest achievement rate of the same variety in different harvesting regions. The variance model formula is: In the formula, For sample variance, For the sample size, For the first The observed values ​​of each formula, The sample mean is the variance. When the variance is significant, that is... This indicates that there are significant regional differences in the growth of this variety in tea gardens, and planting strategies need to be adjusted accordingly for the region.

[0088] S43. Quantitatively analyze the popularity of tea varieties in actual tea gardens in the current year from three dimensions: reservation fulfillment, popularity trend, and tourist preferences.

[0089] Specifically, the reservation fulfillment is calculated by statistically analyzing the actual harvest volume and reservation requests for each tea variety to determine the fulfillment rate, analyze the popularity of the tea variety, and thus obtain the deviation between the actual performance of the tea and tourists' expectations or supply shortages. The popularity trend is calculated by statistically analyzing the reservation and actual harvest volume of each variety on a weekly basis, calculating the month-on-month growth, and drawing a popularity trend curve to analyze the popularity of the tea variety. The reasons are also analyzed in conjunction with external factors, including market competitors' dynamics, harvest weather conditions, and the impact of planting areas on tea tree growth. The tourist preferences are analyzed by mining the correlation patterns between tourist groups and preferred types of tea based on tourist profile data, providing a basis for precise marketing and variety promotion. The tourist profile data includes tourist age, consumption frequency, and historical reservation records.

[0090] S44. Based on the analysis results, output an annual visualization report for the tea garden. The annual visualization report for the tea garden includes a heat map of the growth and health of each tea variety, an annual ranking of the popularity of tea varieties, and a report on variety planting and promotion strategies, which are used to support the implementation of tea garden planting structure adjustment and marketing decisions.

[0091] Furthermore, in step S41, data integration uses tourist identity information, reservation date, and reserved picking variety as association keys to match and merge reservation request data with actual picking behavior data, and marks non-fulfillment, variety deviation, and regional deviation with labels, ultimately forming a structured dataset of the correlation between tourist reservations and actual picking.

[0092] Furthermore, by first collecting and integrating data on tourist reservation requests for tea picking and actual picking behavior, and then analyzing the growth of tea trees from the dimensions of growth progress, yield potential, and regional consistency based on the integrated data, the popularity of varieties is quantified from the dimensions of reservation fulfillment, popularity trends, and user preferences. Finally, based on the analysis results, an annual visual report of the tea garden is generated to support the adjustment of planting structure and marketing decisions. This combines the analysis of tea tree growth status with the mining of market demand preferences, avoiding the blindness of tea gardens planning planting or promoting varieties based solely on experience. It makes planting adjustments and marketing strategies more in line with actual growth conditions and market demand, helping tea gardens achieve supply and demand matching.

[0093] Table 1 is a comparative report of the implementation of the intelligent tea garden management system before and after its implementation. It combines three varieties: Longjing 43, Fuding Dabai tea, and Biluochun tea, covering three areas: the core area, the old tea tree area, and the experimental area, as well as the two main harvesting seasons: spring (March-May) and autumn (September-November). The data collection period is two years before the implementation of the system (2021-2022) and two years after the implementation (2023-2024). All data in the table are seasonal averages or annual averages.

[0094] Table 1

[0095]

[0096] Based on the data in the table above, it can be seen that the present invention, through the deep integration of intelligent picking equipment and system model algorithm, has greatly improved the actual picking volume and reservation picking rate of all experimental tea varieties in all seasons. At the same time, it has comprehensively improved tourists' satisfaction with tea picking, while the overall planting cost has been greatly reduced and the equipment operating efficiency has been comprehensively improved. This shows that the system of the present invention is conducive to the collaborative management of smart tea gardens.

[0097] The above embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Although the invention has been described in detail with reference to the embodiments, those skilled in the art should understand that various combinations, modifications, or equivalent substitutions of the technical solutions of the invention do not depart from the spirit and scope of the invention and should be covered within the scope of the claims of the invention.

Claims

1.A smart tea garden service collaborative management system, characterized in that, The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking system and method. The application relates to a tea leaf picking S221, after receiving the pre-processed data, extracting key features, and establishing an AI model, putting the key features into the AI model to screen abnormal picking actions, analyzing the comprehensive characteristic values of abnormal picking actions, the key features including hand force curve, action trajectory coordinates, picking position height, and shear and fold action frequency; S222, when abnormal picking actions are screened, 1s of video frames before and after the abnormal picking action moment are screened from the collected video for frame-by-frame analysis to further judge the accuracy of abnormal picking action screening and verify the type of abnormal picking action, analyze the relevance of abnormal picking action and tourist characteristics, establish a relevance analysis model, and generate a feature report of tourist picking action, the tourist characteristics including picking experience and preference for tea varieties; S223, storing and uploading the identified abnormal picking action type, frequency and correlation analysis results, and using an iterative optimization model to optimize the parameters of the AI model for screening abnormal picking actions to reduce the misjudgment rate; S23, classifying abnormal picking actions, triggering real-time prompts to correct tourist picking actions, the prompts including vibration prompts, voice prompts and online guidance; S24, summarizing tourist abnormal picking actions and triggered prompts, analyzing the effectiveness of different prompt methods, and generating abnormal picking actions that tourists are most likely to cause, generating picking precautions, and updating regularly; The comprehensive management docking module is used for integrating all module data, generating multi-dimensional reports, and docking with the tea garden enterprise ERP system to generate a collaborative solution for management personnel, planting personnel and customer service personnel, and completing the collaborative management of data integration-role adaptation-decision landing; The variety picking analysis module analyzes the tea garden tea growth and the popularity of the actual variety of the year based on the user's reservation picking request and actual picking behavior; The planting management optimization module formulates the current tea garden management plan and the next year's tea garden idle area planting plan based on the tea garden tea growth and the popularity of the actual variety of the year, while collecting picking density. 2.The system according to claim 1, wherein, When the tourist makes a reservation picking request through the terminal, the tourist's identity information and the tea variety to be picked need to be uploaded, the identity information including the tourist's name and phone number. After receiving the reservation picking request information, the tourist picking reservation module arranges the tourist's self-collection time interval, area and route planning according to the current reservation picking information of the tea garden, and sends it to the tourist terminal. After the tourist confirms the specific picking time, the tourist reservation picking request is sent to the customer service personnel terminal one day before the picking time, and the customer service personnel contact and remind the tourist to enter the tea garden for picking, and send the video of the standard picking action specification and the video of the precautions to the customer. 3.The system of claim 1, wherein, In step S223, the abnormal picking action prompts and corrections for tourists include configuring corresponding prompt rules according to the abnormal picking action comprehensive characteristic values obtained in steps S221 and S222 and the video of the confirmed abnormal picking action. After triggering the prompt, the intelligent picking device collects the tourist's adjustment action data, and at the same time, the tourist's behavior is observed through the video to determine whether the prompt is responded. 4.The system of claim 1, wherein, The comprehensive management docking module includes the following steps: S31, obtaining tourist picking reservation, abnormal picking behavior, video action analysis, and performing data cleaning, standardization and structuring, and outputting preprocessed multi-module unified format data; S32, based on the preprocessed unified format data, calling a report template engine to generate multi-dimensional visual reports according to time dimension, behavior dimension and role dimension; S33, performing format conversion and real-time transmission on the generated multi-dimensional report data, automatically generating a tea leaf processing list for the next stage of the variety, and synchronously uploading the picking and processing cost details to a financial terminal for revenue accounting and cost amortization, while triggering the calculation of materials required for maintaining tea trees; S34, based on the converted multi-dimensional report data, generating a three-party collaborative scheme for management personnel, planting personnel and customer service personnel, and performing task tracking and feedback optimization. 5.The system of claim 4, wherein, In step S34, task tracking and feedback optimization is performed by collecting execution data between roles in real time, comparing the execution data with the expected target of the collaborative scheme, generating an execution deviation data set, locating the deviation link, collecting improvement suggestions for the collaborative scheme from each role through a feedback portal, and returning the execution deviation data set to step S31 to update the feature dimension and quality of the preprocessed data set, providing a basis for the next round of report generation and scheme optimization. 6.The system of claim 1, wherein, The tea garden tea variety analysis module includes the following steps: S41, collecting tourist reservation picking request data and actual picking behavior data, and performing data integration; S42, based on the integrated data, analyzing the growth of tea trees from three directions of growth progress, yield potential and regional consistency; S43, quantitatively analyzing the popularity of actual tea garden tea varieties in the current year from three dimensions of reservation performance, heat trend and tourist preference; S44, outputting a tea garden annual visual report according to the analysis results, which includes a tea garden variety growth and health heat map, an annual tea variety popularity ranking, and a variety planting and promotion strategy report, to support the landing of tea garden planting structure adjustment and marketing decision-making. 7.The system of claim 6, wherein, In step S41, data integration takes the tourist identity information, reservation date and reservation picking variety as the association key, matches and fuses the reservation request data and actual picking behavior data, marks the tags of non-performance, variety deviation and regional deviation, and finally forms a structured tourist reservation and actual picking correlation data set.

Images