Coconut falling and collecting device
By integrating pressure sensors and industrial cameras in the coconut fall collection device, real-time monitoring and prediction of coconut falls, the problems of device overload and coconut damage are solved, and an efficient and safe coconut collection process is achieved.
Patent Information
- Application Number
- CN202510044846.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-10
- Publication Date
- 2025-05-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing coconut fall collection device cannot carry weight for prediction, causing the device to collapse during overload and cause coconut damage.
A coconut fall collection device is designed, integrating pressure sensors, industrial cameras, GPS locators and data processing modules. By monitoring pressure data and image data in real time, it predicts coconut fall conditions and triggers emergency processing procedures when overloaded.
Real-time monitoring and prediction of the coconut fall collection process is achieved, avoiding device overload and coconut damage, and improving collection efficiency and safety.
Smart Images
Figure CN119924083A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of coconut falling collection, in particular to a coconut falling collection device. Background Art
[0002] The coconut falling collection device is an innovative device designed to improve the efficiency of coconut picking. It is usually installed under the coconut tree to catch the naturally fallen coconuts, thereby preventing the coconuts from being damaged during the falling process, protecting the safety of pedestrians in the forest, and reducing the labor intensity of the pickers. This device not only improves the efficiency of coconut collection, but also ensures the safety of pickers and pedestrians. It is an indispensable part of the modern coconut industry. Through scientific design and careful installation, the coconut falling collection device can effectively collect fallen coconuts, which can be used for the selection and breeding of excellent coconut varieties, prediction of fruiting rules, and preparation of basic data and materials for the control of picking time, contributing to the improvement of coconut fruit quality and the sustainable development of the industry.
[0003] In the process of collecting coconuts, the existing coconut falling collection device cannot predict the bearing weight. When the coconut falling collection device cannot bear the total weight of the coconuts, the coconut falling collection device will collapse, causing the coconuts to be damaged after falling to the ground. In order to solve this technical pain point, the present invention provides a coconut falling collection device. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present invention provides a coconut falling collection device to solve the problem that the existing coconut falling collection device cannot bear the weight for prediction. When the coconut falling collection device cannot bear the total weight of the coconuts, the coconut falling collection device will collapse, causing the coconuts to be damaged after falling to the ground.
[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:
[0006] The invention discloses a coconut falling collection device, comprising: a first fastening hoop, a second fastening hoop, a first-layer support rod, a lateral support rod, a second-layer support rod, a top connecting rod, a pressure sensor, an external equipment mounting platform, an industrial camera and a GPS locator, a first solar panel bracket, a second solar panel bracket, a third solar panel bracket, a solar panel, a rubber-coated wire mesh bag, support rod welding fixing screws, and a coconut tree trunk rubber protection.
[0007] The first layer of support rods are installed on the side wall of the first fastening hoop, the second layer of support rods are installed on the side wall of the second fastening hoop, a transverse support rod is installed at the connection between the first layer of support rods and the second layer of support rods, an interception net is installed between the transverse support rods, the second layer of support rods and the top connecting rods, the interception net is used to receive coconuts, the second fastening hoop is connected to the bottom of the second layer of support rods, and the second fastening hoop is used to support the transverse support rods.
[0008] An external equipment installation platform is installed on the side wall of the first fastening hoop, and an industrial camera and a GPS locator are installed on the top of the external equipment installation platform. The industrial camera is used to collect image data at the transverse support rod, and the interception net is located at the transverse support rod and is installed with a pressure sensor and an environmental sensor;
[0009] The outer wall of the second layer of support rods is installed with a first solar panel bracket, a second solar panel bracket and a third solar panel bracket, the first solar panel bracket is connected with the second solar panel bracket and the third solar panel bracket, and a solar panel is installed on the third solar panel bracket;
[0010] The first fastening hoop and the second fastening hoop are used to fasten the coconut tree. A control device, a power supply and an inverter are also installed under the first fastening hoop. One end of the inverter is connected to the first solar panel and the second solar panel, and the other end of the inverter is connected to the power supply. The power supply is connected to the GPS locator, the control device, the industrial camera and the pressure sensor. The control device establishes a communication connection with the pressure sensor and the industrial camera. The pressure sensor and the industrial camera transmit the collected data to the control device. After analyzing the data, the control device generates coconut falling collection information and transmits the coconut falling collection information to the background control terminal and the user mobile terminal.
[0011] Furthermore, the coconut falling collection device described in the present invention, the control device comprises:
[0012] Data acquisition module, which acquires the falling time data, falling quantity data, coconut status data, environmental parameter data, historical coconut falling image data, device status data and geographic location information data;
[0013] The data processing module divides the falling time data, the falling quantity data, the coconut status data, the historical coconut falling image data, the environmental parameter data, the device status data and the geographical location information data into a test set and a verification set, and uses the test set and the verification set to train the neural network model to obtain a coconut falling collection prediction model;
[0014] The prediction module receives the real-time image data collected by the industrial camera, substitutes the real-time image data into the coconut image analysis model, and outputs the coconut quantity monitoring results;
[0015] The detection module obtains data collected by the pressure sensor when the coconut falls onto the interception net and touches the pressure sensor, and compares the data collected by the pressure sensor with the preset safety value of the coconut falling collection device. If the data collected by the pressure sensor is greater than the preset safety value of the coconut falling collection device, coconut falling collection information is generated;
[0016] The data synchronization module substitutes the coconut quantity monitoring results and the coconut falling collection information into the coconut falling collection prediction model to obtain the coconut falling collection prediction results, and sends the coconut falling collection prediction results to the background control terminal and the user's mobile terminal.
[0017] Furthermore, in the coconut falling collection device described in the present invention, the data acquisition module comprises:
[0018] Obtaining the falling time data, by installing a timer on the coconut falling collection device or using the internal clock of the control device to record the time when each coconut falls, when the coconut touches the interception net or the pressure sensor, the timer automatically records the current time as the falling time;
[0019] Obtain the data on the number of coconuts dropped. Using pressure sensors and image recognition technology, each time a coconut falls and is successfully caught, the counter automatically increases by one, thereby accumulating the number of coconuts dropped.
[0020] Obtain coconut status data, collect images of fallen coconuts through industrial cameras, and use preset image recognition algorithms to analyze the appearance, color, size characteristics, and weight of coconuts, so as to determine the maturity and quality of coconuts;
[0021] Obtain environmental parameter data. Environmental sensors include anemometers, wind vanes, thermometers, hygrometers, and light intensity meters, which collect environmental parameter data at the installation location in real time.
[0022] Obtain historical coconut falling image data. Industrial cameras are used for real-time monitoring and also store historical coconut falling image data.
[0023] Furthermore, in the coconut falling collection device described in the present invention, the data processing module comprises:
[0024] Pre-processing the acquired falling time data, falling quantity data, coconut status data, environmental parameter data, historical coconut falling image data, device status data and geographic location information data;
[0025] The preprocessed data set is randomly divided into a test set and a validation set. The test set is used to train the neural network model, and the validation set is used to evaluate the performance of the model and adjust the model parameters.
[0026] Select a neural network model for training, use the test set to train the model, use the validation set to evaluate the trained model, and obtain a trained coconut falling collection prediction model.
[0027] Furthermore, in the coconut falling collection device described in the present invention, the prediction module comprises:
[0028] Receive real-time image data from industrial cameras. The real-time image data is transmitted in the form of video streams or image frames. The real-time image data also includes dynamic information about the environment around the coconut fall collection device. The image data can also be used to collect the growth process of the fruit, trace back and record the male and female sources of pollination and the ripening process of excellent fruit varieties, and can be used as a reference for selecting parent sources with good taste and sweetness.
[0029] Preprocess the received real-time image data, and load the pre-trained coconut image analysis model into the prediction module. The coconut image analysis model is built based on deep learning or traditional image processing algorithms to identify and count coconuts in the image.
[0030] The pre-processed image data is input into the coconut image analysis model, and the coconut image analysis model analyzes the image frame by frame or region by region to identify the coconuts in the image;
[0031] Using the counting function of the coconut image analysis model, the number of coconuts in each frame or each area is counted, and the identified coconut targets are marked, tracked, and counted;
[0032] The results of image recognition and counting are organized into coconut quantity monitoring results, which include the total number of coconuts, the number of newly added coconuts, and the coconut location information in the current frame or current time period.
[0033] Furthermore, in the coconut falling collection device described in the present invention, the detection module comprises:
[0034] Before the detection module is started or the coconut falling collection device starts working, the pressure sensor is initialized and set to a preset safety value. The preset safety value is used as the maximum pressure or weight to be borne. If the preset safety value is exceeded, it is considered to be an unsafe state;
[0035] The detection module continuously monitors the output data of the pressure sensor. When the coconut falls onto the interception net, the pressure sensor senses the pressure changes and converts these changes into electrical signals for transmission;
[0036] Process the collected pressure data in real time, and compare the real-time monitored pressure data with the preset safety value. If the pressure data is less than or equal to the preset safety value, the detection module continues to monitor without taking further action. If the pressure data is greater than the preset safety value, proceed to the next step;
[0037] When it is detected that the pressure data exceeds the preset safety value, the detection module immediately generates coconut falling collection information, which includes the current time, pressure value, degree of exceeding the safety value, and alarm signal;
[0038] The coconut falling collection information is used to trigger notification to users, initiate emergency procedures, and adjust the working status of the coconut falling collection device.
[0039] Furthermore, in the coconut falling collection device described in the present invention, the data synchronization module comprises:
[0040] The data synchronization module first receives the coconut quantity monitoring result from the prediction module and the coconut falling collection information from the detection module, and integrates the received coconut quantity monitoring result and coconut falling collection information to match the timestamps of the coconut quantity monitoring result and the coconut falling collection information;
[0041] Perform necessary preprocessing on the integrated data and substitute the preprocessed data into the coconut falling collection prediction model. The coconut falling collection prediction model is trained based on historical data and machine learning algorithms to predict the future trend of coconut falling collection.
[0042] The prediction results of coconut falling collection are generated by calculating the coconut falling collection prediction model, and the coconut falling collection prediction model includes the prediction of coconut quantity, collection efficiency and device status;
[0043] The generated coconut falling collection prediction results are sent to the background control terminal and the user's mobile terminal.
[0044] Beneficial effects of the present invention:
[0045] The coconut falling collection device of the present invention realizes comprehensive monitoring, real-time prediction and efficient management of the coconut falling collection process by integrating a prediction module, a detection module and a data synchronization module, bringing remarkable beneficial effects. The prediction module can identify, count and track coconuts in real time, provides accurate target positioning for the collection process, reduces blind searches and omissions, and thus improves the collection efficiency.
[0046] The coconut fall collection prediction model can predict future trends based on historical data and current conditions, helping users optimize collection strategies and arrange resources and time reasonably. The detection module can timely detect and warn potential safety hazards such as overload and equipment failure by continuously monitoring the data of the pressure sensor, effectively preventing the occurrence of safety accidents. The data synchronization module improves the timeliness of prediction and detection information, allowing users to quickly respond to safety warnings and take necessary measures to ensure the safety of personnel and equipment.
[0047] The present invention realizes the digital and visual management of the coconut falling collection process by integrating multi-module data. Users can check the collection progress, equipment status and prediction results at any time through the background control terminal and the user mobile terminal. The application of the coconut falling collection prediction model makes management decisions more scientific and accurate, reduces the dependence on human experience, and improves the intelligent level of management.
[0048] By accurately predicting the number of coconuts and collection efficiency, users can more reasonably allocate human and material resources to avoid resource waste and over-investment. The prediction of device status helps to timely perform equipment maintenance and upkeep, extend equipment life, and reduce operation and maintenance costs.
[0049] The coconut falling collection device of the present invention improves collection efficiency and safety, reduces collection costs, improves the dynamic management technology of fruit quality, and helps to enhance the competitiveness and sustainable development capabilities of the coconut industry. Intelligent management means also provide strong support for the transformation and upgrading of the coconut industry, and promote the modernization and intelligent development of the industry.
[0050] In summary, the coconut falling collection device described in the present invention shows significant beneficial effects in improving efficiency, enhancing safety, improving the level of intelligent management, optimizing resource utilization and promoting the sustainable development of the coconut industry, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.
[0052] Figure 1 The present invention provides a schematic structural diagram of a coconut falling collection device.
[0053] Figure 2 A schematic diagram of a control device module in a coconut falling collection device provided by the present invention.
[0054] Figure 3 A schematic diagram of an application scenario of a coconut falling collection device provided by the present invention.
[0055] Figure 4 The present invention provides a schematic top view of a coconut falling collection device.
[0056] Figure 5 The present invention provides a schematic diagram of the three-dimensional structure of a coconut falling collection device.
[0057] Figure 6This is a first partial schematic diagram of a coconut falling collection device provided by the present invention.
[0058] Figure 7 This is a second partial schematic diagram of a coconut falling collection device provided by the present invention.
[0059] Figure 8 This is a third partial schematic diagram of a coconut falling collection device provided by the present invention.
[0060] Figure numerals: 1. first fastening hoop, 2. second fastening hoop, 3. first layer support rod, 4. lateral support rod, 5. second layer support rod, 6. top connecting rod, 7. pressure sensor, 8. external equipment mounting table, 9. coconut tree, 10. industrial camera, 11. first bracket of solar panel, 12. second bracket of solar panel, 13. third bracket of solar panel, 14. solar panel, 15. rubber-coated wire mesh bag, 16. support rod welding fixing screws, 17. rubber protection of coconut tree trunk, 20. coconut, 21. person. DETAILED DESCRIPTION
[0061] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.
[0062] In order to better understand the purpose of the present invention, the present invention is described in further detail below.
[0063] See also Figures 1 to 8 The present invention provides a coconut falling collection device, comprising: a first fastening hoop 1, a second fastening hoop 2, a first-layer support rod 3, a lateral support rod 4, a second-layer support rod 5, a top connecting rod 6, a pressure sensor 7, an external equipment mounting platform 8, an industrial camera and a GPS locator 10, a first solar panel bracket 11, a second solar panel bracket 12, a third solar panel bracket 13, a solar panel 14, a rubber-coated wire mesh bag 15, support rod welding fixing screws 16, and a coconut trunk rubber protection 17.
[0064] A first-layer support rod 3 is installed on the side wall of the first fastening hoop 1, a second-layer support rod 5 is installed on the side wall of the second fastening hoop 2, a transverse support rod 4 is installed at the connection between the first-layer support rod 3 and the second-layer support rod 5, an interception net is installed between the transverse support rod 4, the second-layer support rod 5 and the top connecting rod 6, and the interception net is used to receive coconuts. The second fastening hoop 2 is connected to the bottom of the second-layer support rod 5, and the second fastening hoop 2 is used to support the second-layer support rod 5.
[0065] An external equipment mounting platform 8 is installed on the side wall of the first fastening hoop 1, and an industrial camera and a GPS locator 10 are installed on the top of the external equipment mounting platform 8. The industrial camera 10 is used to collect image data at the transverse support rod 4, and the interception net is located at the transverse support rod 4 and is installed with a pressure sensor 7 and an environmental sensor;
[0066] The outer wall of the second support rod 5 is installed with a first solar panel bracket 11, a second solar panel bracket 12 and a third solar panel bracket 13. The first solar panel bracket 11 is connected with the second solar panel bracket 12 and the third solar panel bracket 13. The third solar panel bracket 13 is installed with a solar panel 14.
[0067] The first fastening hoop 1 and the second fastening hoop 2 are used to fasten the coconut tree. A control device, a power supply and an inverter are also installed below the first fastening hoop 1. One end of the inverter is connected to the solar panel 14, and the other end of the inverter is connected to the power supply. The power supply is connected to the GPS locator, the control device, the industrial camera 10 and the pressure sensor 7. The control device establishes a communication connection with the pressure sensor 7 and the industrial camera 10. The pressure sensor 7 and the industrial camera 10 transmit the collected data to the control device. After analyzing the data, the control device generates coconut falling collection information and transmits the coconut falling collection information to the background control terminal and the user mobile terminal.
[0068] The solar panel 14 absorbs sunlight, converts light energy into electrical energy through an inverter and transmits it to a power source, which supplies power to the electrical device of the present invention.
[0069] Fastening clamps: The first fastening clamp 1 and the second fastening clamp 2 are used to fasten the coconut tree to ensure that the device is stable and immovable.
[0070] Support rods and connecting rods: The first-layer support rods 3, the lateral support rods 4, the second-layer support rods 5, and the top connecting rods 6 together constitute the skeleton of the device, supporting and connecting various components.
[0071] Interception net: installed between the lateral support rod 4, the second layer support rod 5 and the top connecting rod 6, used to receive the fallen coconuts.
[0072] Pressure sensor 7: installed on the interception net at the lateral support rod 4, to monitor in real time the pressure generated by the coconut falling onto the interception net.
[0073] External device mounting platform 8: used for mounting external devices such as industrial cameras 10 and GPS locators.
[0074] Industrial camera 10: installed on the top of the external equipment mounting platform 8, used for collecting image data at the transverse support rod 4.
[0075] Working principle: The pressure sensor 7 monitors the pressure data on the interception net in real time to reflect the weight and quantity of coconuts. The industrial camera 10 collects image data at the lateral support rod 4 to identify the number, position and other information of coconuts. The pressure sensor 7 and the industrial camera 10 transmit the collected data to the control device. The control device also receives additional data from other sensors (such as environmental sensors) to more fully understand the environment and status of the coconut falling collection process.
[0076] The control device analyzes and processes the received data, including real-time monitoring of pressure data, recognition and counting of image data, etc. By comparing the preset safety value, the control device can determine whether the coconut falling collection device is overloaded or has other abnormal conditions. Based on the data analysis results, the control device generates coconut falling collection information, including the number of coconuts, location, weight, and device status. The information is transmitted to the background control terminal and the user's mobile terminal, so that the user can understand the situation of the coconut falling collection process at any time and make corresponding decisions.
[0077] In summary, the coconut falling collection device provided by the present invention realizes efficient and safe collection of coconuts through the coordinated work of various components, and its working principle and operation process are clear and concise, which improves the smooth progress of the coconut falling collection process and the convenient operation of the user.
[0078] Specifically, the coconut falling collection device described in the present invention, the control device comprises:
[0079] Data acquisition module, which acquires the falling time data, falling quantity data, coconut status data, environmental parameter data, historical coconut falling image data, device status data and geographic location information data;
[0080] The data processing module divides the falling time data, the falling quantity data, the coconut status data, the historical coconut falling image data, the environmental parameter data, the device status data and the geographical location information data into a test set and a verification set, and uses the test set and the verification set to train the neural network model to obtain a coconut falling collection prediction model;
[0081] The prediction module receives the real-time image data collected by the industrial camera 10, substitutes the real-time image data into the coconut image analysis model, and outputs the coconut quantity monitoring result;
[0082] The detection module obtains data collected by the pressure sensor when the coconut falls onto the interception net and touches the pressure sensor 7, and compares the data collected by the pressure sensor with the preset safety value of the coconut falling collection device. If the data collected by the pressure sensor is greater than the preset safety value of the coconut falling collection device, coconut falling collection information is generated;
[0083] The data synchronization module substitutes the coconut quantity monitoring results and the coconut falling collection information into the coconut falling collection prediction model to obtain the coconut falling collection prediction results, and sends the coconut falling collection prediction results to the background control terminal and the user's mobile terminal.
[0084] The control device integrates multiple modules to achieve comprehensive monitoring, prediction and control of the coconut falling collection process. It first collects various data related to the coconut falling collection through the data acquisition module, and then uses the data processing module to process and analyze these data to obtain a prediction model for the coconut falling collection. Then, the number of coconuts and the status of the device are monitored in real time through the prediction module and the detection module, and finally the coconut falling collection information is sent to the background control terminal and the user mobile terminal through the data synchronization module.
[0085] Data acquisition module: responsible for acquiring various data related to coconut falling collection, including falling time, falling quantity, coconut status, environmental parameters, historical coconut falling images, device status and geographical location information, etc. The data acquisition module acquires and stores these data in real time through interfaces with various sensors (such as pressure sensors, environmental sensors, etc.) and cameras (such as industrial cameras).
[0086] Data processing module: Process and analyze the acquired data to obtain a prediction model for the collection of coconut falls. The data processing module first divides the collected data into a test set and a validation set, and then uses these data to train the neural network model, and finally obtains a model that can predict the collection of coconut falls.
[0087] Prediction module: Use the coconut image analysis model to process real-time image data and output the coconut quantity monitoring results. The prediction module receives real-time image data from industrial cameras, substitutes these data into the coconut image analysis model for processing, and finally obtains the monitoring results such as the number and location of coconuts.
[0088] Detection module: monitors the status of the coconut falling collection device in real time. When the data collected by the pressure sensor exceeds the preset safety value, the coconut falling collection information is generated. The detection module monitors the output data of the pressure sensor and compares it with the preset safety value. When the data exceeds the safety value, the coconut falling collection information containing the current time, pressure value, degree of exceeding the safety value, and alarm signal is immediately generated.
[0089] Data synchronization module: After integrating the coconut quantity monitoring results and the coconut falling collection information, they are substituted into the coconut falling collection prediction model for calculation to obtain the prediction results of coconut falling collection, and these results are sent to the background control terminal and the user's mobile terminal.
[0090] The data synchronization module first integrates the data from the prediction module and the detection module, and then substitutes these data into the coconut fall collection prediction model for calculation. Finally, the generated prediction results are sent to the background control terminal and the user's mobile terminal through the communication network, so that the user can understand the status and prediction of the coconut fall collection device at any time.
[0091] In summary, the control device in the coconut falling collection device of the present invention realizes comprehensive monitoring, prediction and control of the coconut falling collection process by integrating multiple functional modules, and the various modules work together to ensure the safety of the coconut falling collection process.
[0092] Specifically, the coconut falling collection device described in the present invention, the data acquisition module includes:
[0093] Obtaining the falling time data, by installing a timer on the coconut falling collection device or using the internal clock of the control device to record the time when each coconut falls, when the coconut touches the interception net or the pressure sensor, the timer automatically records the current time as the falling time;
[0094] Obtain the data on the number of coconuts dropped. Using pressure sensors and image recognition technology, each time a coconut falls and is successfully caught, the counter automatically increases by one, thereby accumulating the number of coconuts dropped.
[0095] Obtain coconut status data, collect images of fallen coconuts through industrial cameras, and use preset image recognition algorithms to analyze the appearance, color, and size characteristics of coconuts to determine the maturity of coconuts;
[0096] Obtain environmental parameter data. Environmental sensors include anemometers, wind vanes, thermometers, light intensity meters, and hygrometers, which collect environmental parameter data at the installation location in real time.
[0097] Obtain historical coconut falling image data. Industrial cameras are used for real-time monitoring and also store historical coconut falling image data.
[0098] Obtaining falling time data: Install a timer on the coconut falling collection device or use the internal clock of the control device, combined with the trigger mechanism of the pressure sensor or interception net. When the coconut touches the interception net or pressure sensor, the timer is triggered to record the current time as the coconut falling time, providing data support in the time dimension for subsequent coconut falling collection and analysis, which helps to understand the law of coconut falling and predict the maturity time of coconut.
[0099] Obtaining data on the number of coconuts dropped: Combining pressure sensors and image recognition technology. Whenever a coconut falls and is successfully caught, the pressure sensor detects the pressure change, and the image recognition technology confirms the presence of the coconut, and the counter automatically increases by one. The number of coconuts dropped is counted in real time, providing direct data for the evaluation of the efficiency of coconut collection.
[0100] Obtaining coconut status data: Use industrial cameras to collect images of fallen coconuts. Through the preset image recognition algorithm, the appearance (such as shape, skin condition), color (such as color changes related to maturity), weight and diameter of the coconut are analyzed to determine the maturity of the coconut. This provides an important basis for coconut classification, storage and sales, and also helps to understand the growth status and yield of coconut trees.
[0101] Obtain environmental parameter data: Install environmental sensors, including anemometers, wind vanes, thermometers, light meters, and hygrometers. The sensors collect environmental parameters such as wind speed, wind direction, temperature, light, and humidity at the installation location in real time. Providing environmental background data for the coconut falling collection process helps analyze the impact of the environment on coconut flowering, fruiting, fruiting, falling, and collection efficiency, and also provides environmental reference for coconut storage and transportation.
[0102] Obtain historical coconut falling image data: Use the real-time monitoring function of industrial cameras and set up a data storage mechanism. Industrial cameras are not only used to monitor the falling of coconuts in real time, but also responsible for storing historical coconut falling image data. Providing rich image data for subsequent coconut falling collection and analysis will help review and analyze the laws of coconut falling, abnormal conditions, and the working status of the device.
[0103] In summary, the data acquisition module in the coconut falling collection device of the present invention realizes comprehensive and accurate monitoring and data collection of the coconut falling collection process by integrating multiple sensors and image processing technologies.
[0104] Specifically, the coconut falling collection device described in the present invention, the data processing module includes:
[0105] Pre-processing the acquired falling time data, falling quantity data, coconut status data, environmental parameter data, historical coconut falling image data, device status data and geographic location information data;
[0106] The preprocessed data set is randomly divided into a test set and a validation set. The test set is used to train the neural network model, and the validation set is used to evaluate the performance of the model and adjust the model parameters.
[0107] Select a neural network model for training, use the test set to train the model, use the validation set to evaluate the trained model, and obtain a trained coconut falling collection prediction model.
[0108] Data preprocessing: Clean, normalize, and process missing values of the acquired drop time data, drop quantity data, coconut status data, environmental parameter data, historical coconut drop image data, device status data, and geographic location information data to ensure data quality, improve the efficiency and accuracy of subsequent model training, and reduce the impact of data noise on model performance.
[0109] Dataset partitioning: The preprocessed dataset is randomly divided into a test set and a validation set. The test set accounts for the majority of the dataset (e.g., 70%-80%) and is used to train the neural network model; the validation set accounts for a small portion of the dataset (e.g., 20%-30%) and is used to evaluate the performance of the model and adjust the model parameters. This partitioning method helps avoid overfitting of the model and improves the generalization ability of the model. It ensures that the model can fully learn the characteristics of the data during training, and at the same time, it can evaluate the performance of the model on the validation set so that necessary adjustments can be made.
[0110] Neural network model selection and training: Select a suitable neural network model for training, such as convolutional neural network (CNN) for image processing and recurrent neural network (RNN) for time series prediction. Use the test set to train the model, and adjust the model parameters through the back propagation algorithm so that the model can gradually fit the characteristics of the data. At the same time, use the validation set to evaluate the trained model, and calculate the model's accuracy, recall rate, F1 score and other indicators to evaluate the performance of the model.
[0111] Through continuous training and evaluation, a model is obtained that can accurately predict the collection of coconut falls, such as the time, quantity, maturity, etc. of coconut falls, as well as abnormal conditions predicted based on environmental parameters and device status. Based on the evaluation results on the validation set, the model is optimized, such as adjusting the network structure, increasing or decreasing the number of neurons, changing the activation function, etc. Through continuous trials and adjustments, the model parameters and structure with the best performance are found. Finally, the trained model is saved for prediction in practical applications. The prediction accuracy and stability of the model are improved to ensure that it can work reliably in practical applications.
[0112] In summary, the data processing module in the coconut falling collection device of the present invention finally obtains a model that can accurately predict the coconut falling collection situation through the steps of data preprocessing, data set division, neural network model selection and training, and model optimization and preservation, thereby providing strong support for the monitoring, prediction and control of the coconut falling collection process.
[0113] Specifically, the coconut falling collection device described in the present invention, the prediction module includes:
[0114] Receiving real-time image data from an industrial camera 11, the real-time image data is transmitted in the form of a video stream or an image frame, and the real-time image data also includes dynamic information of the environment around the coconut falling collection device;
[0115] Preprocess the received real-time image data, and load the pre-trained coconut image analysis model into the prediction module. The coconut image analysis model is built based on deep learning or traditional image processing algorithms to identify and count coconuts in the image.
[0116] The pre-processed image data is input into the coconut image analysis model, and the coconut image analysis model analyzes the image frame by frame or region by region to identify the coconuts in the image;
[0117] Using the counting function of the coconut image analysis model, the number of coconuts in each frame or each area is counted, and the identified coconut targets are marked, tracked, and counted;
[0118] The results of image recognition and counting are organized into coconut quantity monitoring results, which include the total number of coconuts, the number of newly added coconuts, and the coconut location information in the current frame or current time period.
[0119] Receiving real-time image data: receiving real-time image data from the industrial camera 11, which is transmitted in the form of video stream or image frame and contains dynamic information of the environment around the coconut falling collection device. Providing raw data for subsequent image processing to ensure that the real-time status of the coconut can be captured.
[0120] Image data preprocessing: Preprocess the received real-time image data, such as denoising, enhancement, cropping, etc., to improve image quality and reduce the complexity of subsequent processing. Provide clear and accurate input data for the coconut image analysis model to improve recognition accuracy.
[0121] Loading coconut image analysis model: The prediction module loads a pre-trained coconut image analysis model, which is built based on deep learning or traditional image processing algorithms and has the ability to identify coconuts in images. The model is used to identify coconuts in pre-processed image data, providing a basis for subsequent counting and tracking.
[0122] Coconut identification and counting: The pre-processed image data is input into the coconut image analysis model. The model analyzes the image frame by frame or region by region, identifies the coconuts in the image, and uses the counting function to count the number of coconuts in each frame or region. The model scans the image frame by frame or region by region, identifies the coconuts using feature extraction and classification technology, and marks, tracks and counts the identified coconut targets. The number of coconuts is obtained in real time to provide data support for the monitoring and control of the coconut fall collection process.
[0123] Organize the coconut quantity monitoring results: Organize the results of image recognition and counting into the coconut quantity monitoring results, including the total number of coconuts in the current frame or current time period, the number of newly added coconuts, and the location information of coconuts. Present the recognition results in a structured form to facilitate subsequent data analysis and processing, and to show the real-time status of the coconut falling collection device to the user.
[0124] In summary, the prediction module in the coconut falling collection device of the present invention realizes the real-time recognition, counting and tracking of coconuts by receiving real-time image data, performing preprocessing, loading a coconut image analysis model, performing coconut recognition and counting, and collating coconut quantity monitoring results, thereby providing important technical support for the monitoring and control of the coconut falling collection process.
[0125] Specifically, the coconut falling collection device described in the present invention, the detection module includes:
[0126] Before the detection module is started or the coconut falling collection device starts working, the pressure sensor 7 is initialized and set to a preset safety value, which is used as the maximum pressure or weight to be borne. If the preset safety value is exceeded, it is considered to be an unsafe state;
[0127] The detection module continuously monitors the output data of the pressure sensor 7. When the coconut falls onto the interception net, the pressure sensor senses the pressure changes and converts these changes into electrical signals for transmission;
[0128] Process the collected pressure data in real time, and compare the real-time monitored pressure data with the preset safety value. If the pressure data is less than or equal to the preset safety value, the detection module continues to monitor without taking further action. If the pressure data is greater than the preset safety value, proceed to the next step;
[0129] When it is detected that the pressure data exceeds the preset safety value, the detection module immediately generates coconut falling collection information, which includes the current time, pressure value, degree of exceeding the safety value, and alarm signal;
[0130] The coconut falling collection information is used to trigger notification to users, initiate emergency procedures, and adjust the working status of the coconut falling collection device.
[0131] Pressure sensor initialization setting: before the detection module is started or the coconut falling collection device starts working, the pressure sensor 7 is initialized, including setting a preset safety value.
[0132] Preset safety value: This is a critical threshold that defines the maximum pressure or weight that can be borne. Exceeding this value is considered an unsafe state, meaning that the coconut fall collection device is bearing too much weight or there are other potential safety issues. It provides a clear reference standard for subsequent pressure monitoring to ensure that the device operates within a safe range.
[0133] Continuously monitor pressure data: The detection module continuously monitors the output data of the pressure sensor 7. When the coconut falls onto the interception net, the pressure sensor senses the changes in pressure and converts these changes into electrical signals for transmission. Real-time pressure data during the coconut fall collection process is obtained to provide a basis for subsequent safety judgment.
[0134] Pressure data processing and comparison: The collected pressure data is processed in real time and compared with the preset safety value. If the pressure data is less than or equal to the preset safety value, the detection module continues to monitor without taking further action; if the pressure data is greater than the preset safety value, it indicates that there is a safety problem and needs to proceed to the next step for processing. Timely discovery and handling of potential safety hazards ensures the safety of the coconut falling collection process.
[0135] Generate coconut falling collection information: When the pressure data is detected to exceed the preset safety value, the detection module immediately generates coconut falling collection information.
[0136] Coconut Falls collects information including current time, pressure value, degree of exceeding safety value, and alarm signals.
[0137] Trigger emergency response: Coconut fall collection information is used to trigger a series of emergency response measures. Emergency response measures include but are not limited to notifying users, initiating emergency procedures, adjusting the working status of coconut fall collection devices, etc. When safety issues are detected, countermeasures can be taken quickly and effectively to minimize potential safety risks.
[0138] In summary, the detection module in the coconut falling collection device of the present invention ensures the safety of the coconut falling collection process through the steps of initialization setting, continuous monitoring of pressure data, real-time processing and comparison of pressure data, generation of coconut falling collection information and triggering of emergency response. When potential safety problems are detected, an alarm can be issued in a timely and accurate manner and corresponding measures can be taken to protect the safety of users and equipment.
[0139] Specifically, the coconut falling collection device described in the present invention, the data synchronization module includes:
[0140] The data synchronization module first receives the coconut quantity monitoring result from the prediction module and the coconut falling collection information from the detection module, and integrates the received coconut quantity monitoring result and coconut falling collection information to match the timestamps of the coconut quantity monitoring result and the coconut falling collection information;
[0141] Perform necessary preprocessing on the integrated data and substitute the preprocessed data into the coconut falling collection prediction model. The coconut falling collection prediction model is trained based on historical data and machine learning algorithms to predict the future trend of coconut falling collection.
[0142] The prediction results of coconut falling collection are generated by calculating the coconut falling collection prediction model, and the coconut falling collection prediction model includes the prediction of coconut quantity, collection efficiency and device status;
[0143] The generated coconut falling collection prediction results are sent to the background control terminal and the user's mobile terminal.
[0144] The data synchronization module first receives the coconut quantity monitoring results from the prediction module and the coconut falling collection information from the detection module. The received coconut quantity monitoring results and coconut falling collection information are integrated, and the key is to match the timestamps of the two. For each coconut quantity monitoring result, the coconut falling collection information at the corresponding time point can be found, and vice versa.
[0145] The integrated data is preprocessed as necessary. Preprocessing includes data cleaning (removing outliers, filling missing values, etc.), data normalization (converting data of different dimensions to the same dimension), data smoothing (reducing the impact of data fluctuations on prediction), etc. Improve data quality and make the data more suitable for input into the coconut fall collection prediction model.
[0146] Substitute into the coconut falling collection prediction model: Substitute the preprocessed data into the coconut falling collection prediction model.
[0147] Coconut Fall Collection Prediction Model: A model trained based on historical data and machine learning algorithms is used to predict the future trend of coconut fall collection. The model's ability to learn from historical data is used to predict the current situation of coconut fall collection.
[0148] Generate coconut falling collection prediction results: Generate coconut falling collection prediction results through calculation of coconut falling collection prediction model.
[0149] Prediction results: including coconut quantity prediction (the number of coconuts collected in the future), collection efficiency prediction (the trend of collection speed or efficiency), and device status prediction (the working status or fault warning of the coconut falling collection device). It provides decision support for the backend control terminal and user mobile terminal, helping users to better plan and adjust the coconut falling collection process.
[0150] Send prediction results: Send the generated coconut fall collection prediction results to the background control terminal and the user's mobile terminal.
[0151] The information is sent via wireless networks, wired networks or other communication methods, enabling users to obtain the collected forecast information in real time and make more timely and accurate decisions.
[0152] In summary, the data synchronization module in the coconut falling collection device described in the present invention realizes accurate prediction of the coconut falling collection process and real-time information transmission through the steps of data reception and integration, data preprocessing, substituting into the coconut falling collection prediction model, generating coconut falling collection prediction results and sending the prediction results.
[0153] The technical solution of the present invention aims at the problem that the existing coconut falling collection device cannot predict the load-bearing weight, and the device collapses and the coconuts are damaged after overloading, and proposes the following solutions:
[0154] Pressure sensors are installed at the interception net of the coconut falling collection device to monitor the pressure generated by coconuts falling onto the interception net in real time. The data collected by the pressure sensors can be used to understand the current weight of the coconut falling collection device in real time, providing a basis for prediction and judgment.
[0155] A safety value is preset in the control device as the threshold of the maximum pressure or weight that the coconut falling collection device can withstand. When the data collected by the pressure sensor exceeds this preset safety value, the control device will immediately generate coconut falling collection information and send out an alarm signal to prompt the user to pay attention.
[0156] When it is detected that the pressure data exceeds the standard, the control device will quickly trigger the emergency handling procedure, including notifying the user to take emergency measures and adjusting the working status of the coconut falling collection device (such as stopping collection or releasing some coconuts) to prevent the device from collapsing.
[0157] The data acquisition module in the control device will comprehensively collect multi-dimensional data including falling time, quantity, coconut status, environmental parameters, historical images, device status, and geographic location. The data is pre-processed and trained by the data processing module to form a coconut falling collection prediction model. This model can predict the future trend of coconut falling collection and provide support for decision-making.
[0158] The prediction module uses the real-time image data collected by the industrial camera to analyze and output the coconut quantity monitoring results. The data synchronization module integrates the coconut quantity monitoring results and the coconut falling collection information, substitutes them into the coconut falling collection prediction model for calculation, and generates the coconut falling collection prediction results. The prediction results are sent to the background control terminal and the user's mobile terminal in a timely manner, so that the user can understand the status and prediction of the coconut falling collection device at any time.
[0159] In summary, the technical solution of the present invention effectively solves the problem that the coconut falling collection device cannot bear the weight for prediction and is overloaded through technical means such as pressure sensor monitoring, preset safety value comparison, emergency handling procedures, data acquisition and analysis, prediction and synchronization.
Claims
1. A coconut falling collection device, characterized in that: include: A first fastening hoop (1), a second fastening hoop (2), a first layer support rod (3), a lateral support rod (4), a second layer support rod (5), a top connecting rod (6), a pressure sensor (7), an external equipment mounting platform (8), a coconut tree (9), an industrial camera (10), a first solar panel bracket (11), a second solar panel bracket (12), a third solar panel bracket (13), a solar panel (14), a rubber-coated wire mesh bag (15), support rod welding fixing screws (16), and a coconut tree trunk rubber protection (17); A first-layer support rod (3) is installed on the side wall of the first fastening hoop (1), a second-layer support rod 3 is installed on the side wall of the second fastening hoop (2), a transverse support rod (4) is installed at the connection between the first-layer support rod (3) and the second-layer support rod (5), an interception net is installed between the transverse support rod (4), the second-layer support rod (5) and the top connecting rod (6), and the interception net is used to receive coconuts; the second fastening hoop (2) is connected to the bottom of the transverse support rod (4), and the second fastening hoop (2) is used to support the transverse support rod (4); An external equipment installation platform (8) is installed on the side wall of the first fastening hoop (1), an industrial camera and a GPS locator (10) are installed on the top of the external equipment installation platform (8), the industrial camera (10) is used to collect image data at the transverse support rod (4), and a pressure sensor (7) and an environmental sensor are installed on the interception net located at the transverse support rod (4); A first solar panel bracket (11), a second solar panel bracket (12) and a third solar panel bracket (13) are installed on the outer wall of the second support rod (5); the first solar panel bracket (11) is connected to the second solar panel bracket (12) and the third solar panel bracket (13); and a solar panel (14) is installed on the third solar panel bracket (13); The first fastening hoop (1) and the second fastening hoop (2) are used to fasten the coconut tree. A control device, a power supply and an inverter are also installed below the first fastening hoop (1). One end of the inverter is connected to the solar panel (14), and the other end of the inverter is connected to the power supply. The power supply is connected to the GPS locator, the control device, the industrial camera (10) and the pressure sensor (7). The control device establishes a communication connection with the pressure sensor (7) and the industrial camera (10). The pressure sensor (7) and the industrial camera (10) transmit the collected data to the control device. After analyzing the data, the control device generates coconut falling collection information, and transmits the coconut falling collection information to a background control terminal and a user mobile terminal.
2. A coconut falling collection device as claimed in claim 1, characterized in that: The control device comprises: Data acquisition module, which acquires the falling time data, falling quantity data, coconut status data, environmental parameter data, historical coconut falling image data, device status data and geographic location information data; The data processing module divides the falling time data, the falling quantity and weight data, the coconut status data, the historical coconut falling image data, the environmental parameter data, the device status data and the geographical location information data into a test set and a validation set, and uses the test set and the validation set to train the neural network model to obtain a coconut falling collection prediction model; A prediction module receives real-time image data collected by an industrial camera (10), substitutes the real-time image data into a coconut image analysis model, and outputs a coconut quantity monitoring result; The detection module obtains data collected by the pressure sensor when the coconut falls onto the interception net and touches the pressure sensor (7), compares the data collected by the pressure sensor with a preset safety value of the coconut falling collection device, and generates coconut falling collection information if the data collected by the pressure sensor is greater than the preset safety value of the coconut falling collection device; The data synchronization module substitutes the coconut quantity monitoring results and the coconut falling collection information into the coconut falling collection prediction model to obtain the coconut falling collection prediction results, and sends the coconut falling collection prediction results to the background control terminal and the user's mobile terminal.
3. A coconut falling collection device as claimed in claim 1, characterized in that: The data acquisition module comprises: Obtaining the falling time data, by installing a timer on the coconut falling collection device or using the internal clock of the control device to record the time when each coconut falls, when the coconut touches the interception net or the pressure sensor, the timer automatically records the current time as the falling time; Obtain the data on the number of coconuts dropped. Using pressure sensors and image recognition technology, each time a coconut falls and is successfully caught, the counter automatically increases by one, thereby accumulating the number and weight of coconuts dropped. Obtain coconut status data, collect images of fallen coconuts through industrial cameras, and use preset image recognition algorithms to analyze the appearance, color, and size characteristics of coconuts to determine the maturity and fruit quality of coconuts; Obtain environmental parameter data. Environmental sensors include anemometers, wind vanes, thermometers, light accumulated temperature, and hygrometers to collect environmental parameter data at the installation location in real time. Obtain historical coconut falling image data. Industrial cameras are used for real-time monitoring and also store historical coconut falling image data.
4. A coconut falling collection device as claimed in claim 1, characterized in that: The data processing module comprises: Pre-processing the acquired falling time data, falling quantity data, coconut status data, environmental parameter data, historical coconut falling image data, device status data and geographic location information data; The preprocessed data set is randomly divided into a test set and a validation set. The test set is used to train the neural network model, and the validation set is used to evaluate the performance of the model and adjust the model parameters. Select a neural network model for training, use the test set to train the model, use the validation set to evaluate the trained model, and obtain a trained coconut falling collection prediction model.
5. A coconut falling collection device as claimed in claim 1, characterized in that: The prediction module comprises: Receiving real-time image data from an industrial camera (10), the real-time image data being transmitted in the form of a video stream or an image frame, the real-time image data also including dynamic information of the environment surrounding the coconut falling collection device; Preprocess the received real-time image data, and load the pre-trained coconut image analysis model into the prediction module. The coconut image analysis model is built based on deep learning or traditional image processing algorithms to identify and count coconuts in the image. The pre-processed image data is input into the coconut image analysis model, and the coconut image analysis model analyzes the image frame by frame or region by region to identify the coconuts in the image; Using the counting function of the coconut image analysis model, the number of coconuts in each frame or each area is counted, and the identified coconut targets are marked, tracked, and counted; The results of image recognition and counting are organized into coconut quantity monitoring results, which include the total number of coconuts, the number of newly added coconuts, and the coconut location information in the current frame or current time period.
6. A coconut falling collection device as claimed in claim 1, characterized in that: The detection module comprises: Before the detection module is started or the coconut falling collection device starts working, the pressure sensor (7) is initialized and set to a preset safety value, the preset safety value is used as the maximum pressure or weight to be borne, and exceeding the preset safety value is regarded as an unsafe state; The detection module continuously monitors the output data of the pressure sensor (7). When the coconut falls onto the interception net, the pressure sensor senses the pressure changes and converts these changes into electrical signals for transmission; Process the collected pressure data in real time, and compare the real-time monitored pressure data with the preset safety value. If the pressure data is less than or equal to the preset safety value, the detection module continues to monitor without taking further action. If the pressure data is greater than the preset safety value, proceed to the next step; When it is detected that the pressure data exceeds the preset safety value, the detection module immediately generates coconut falling collection information, which includes the current time, pressure value, degree of exceeding the safety value, and alarm signal; The coconut falling collection information is used to trigger notification to users, initiate emergency procedures, and adjust the working status of the coconut falling collection device.
7. A coconut falling collection device as claimed in claim 1, characterized in that: The data synchronization module comprises: The data synchronization module first receives the coconut quantity monitoring result from the prediction module and the coconut falling collection information from the detection module, and integrates the received coconut quantity monitoring result and coconut falling collection information to match the timestamps of the coconut quantity monitoring result and the coconut falling collection information; Perform necessary preprocessing on the integrated data and substitute the preprocessed data into the coconut falling collection prediction model. The coconut falling collection prediction model is trained based on historical data and machine learning algorithms to predict the future trend of coconut falling collection. The prediction results of coconut falling collection are generated by calculating the coconut falling collection prediction model, and the coconut falling collection prediction model includes the prediction of coconut quantity, collection efficiency and device status; The generated coconut falling collection prediction results are sent to the background control terminal and the user's mobile terminal.