Tomato yield prediction method and system based on normal distribution function

Through the tomato yield prediction method based on normal distribution function, combined with tomato detection and tracking model, the ripening time data is fitted, and the problem of failure to fully consider the tomato's own characteristics in the existing technology is solved, and the accuracy and application of yield prediction are improved. , adapt to the growth conditions of different regions and climatic conditions.

CN118940879APending Publication Date: 2024-11-12EAST CHINA AGRI-TECH CENTER OF CHINESE ACADEMY OF AGRICULTURAL SCIENCES
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410886827.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-03
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

When used, the existing tomato yield prediction methods are missing details of remote sensing images and fail to fully consider the characteristics of tomato fruits, resulting in low accuracy of the yield prediction model, and an intelligent prediction method is urgently needed to improve accuracy.

Method used

Tomato yield prediction method based on normal distribution function is adopted, and the detection model and tracking model are constructed by obtaining tomato fruit images and videos, and the tomato ripening time data is fitted with normal distribution function to calculate the predicted yield and adapt to different regions and climatic conditions. tomato growth.

Benefits of technology

Improve the accuracy and flexibility of tomato yield prediction, can better capture the ripening time of tomato fruit, provide accurate results for yield prediction, and adapt to the growth conditions of different regions and climatic conditions, and has strong practicality and applicability. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940879A_ABST
    Figure CN118940879A_ABST
Patent Text Reader

Abstract

The invention discloses a tomato yield prediction method and system based on a normal distribution function, and the method comprises the steps: obtaining a tomato fruit image and video after a tomato fruit enters a green ripe stage; constructing a tomato fruit detection model according to the obtained tomato fruit image; constructing a tomato fruit tracking model according to the obtained tomato fruit video; and inputting a detection result of the tomato fruit detection model into the tomato fruit tracking model to obtain a tomato fruit tracking counting result, obtaining probability distribution of tomato maturation time through an established normal distribution function, and calculating the predicted yield of tomatoes in the greenhouse. According to the tomato yield prediction method and system based on the normal distribution function, the normal distribution function is used for tomato yield prediction, the maturation time of tomato fruits can be better captured, an accurate result is provided for yield prediction, the method and system can adapt to tomato growth conditions of different regions and weather conditions, and the method and system are suitable for large-scale popularization and application. And the method has relatively high practicability and applicability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of agricultural information technology processing technology, and in particular to a tomato yield prediction method and system based on a normal distribution function. Background Art

[0002] The tomato yield prediction method is a method for predicting and calculating tomato yield. The yield prediction can not only predict the overall tomato yield in the greenhouse, so that farmers can effectively plan tomato storage and sales plans, but also optimize resource allocation and utilization, improve tomato quality, reduce production costs and increase profits. It has important value and significance for the development of agriculture. With the continuous development of science and technology, people have higher and higher requirements for the manufacturing process of tomato yield prediction methods.

[0003] The existing tomato yield prediction methods have certain drawbacks when used. Most of the existing yield prediction methods establish yield prediction models by obtaining the growth status information data of crops, and then obtain the yield prediction results. For example, CN110414738B (a method and system for predicting crop yield) constructs a data set containing remote sensing images and meteorological data, and then performs correlation analysis, principal component analysis and importance analysis on the data set. After obtaining the corresponding data set, four yield prediction models are trained and the optimal model is selected as the crop yield prediction model. This method uses remote sensing images and some meteorological data, and fully considers the impact of external factors of crops on crop yield. However, due to the lack of details in remote sensing images, the inherent characteristics of crops are often ignored, which makes the accuracy of the yield prediction model low; tomato fruits have four key stages in the ripening process, namely green ripening period-color change period-hard ripening period-full ripening period. After tomatoes enter the green ripening period, yield prediction can be started. Therefore, there is an urgent need for an intelligent prediction method and system, which aims to fully consider the inherent characteristics of tomato fruits to improve the accuracy of tomato yield prediction. To this end, we propose a tomato yield prediction method and system based on normal distribution function. Summary of the invention

[0004] Technical problem to be solved: In view of the shortcomings of the prior art, the present invention provides a tomato yield prediction method and system based on the normal distribution function. The use of the normal distribution function for tomato yield prediction can better capture the ripening time of tomato fruits, provide accurate results for yield prediction, and can adapt to the growth of tomatoes in different regions and climatic conditions. It has strong practicality and applicability, and can effectively solve the problems in the background technology.

[0005] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is: a tomato yield prediction method based on a normal distribution function, comprising the following steps: S101: when the tomato fruit enters the green ripe stage, obtaining tomato fruit images and videos; S102: constructing a tomato fruit detection model according to the obtained tomato fruit image; S103: constructing a tomato fruit tracking model according to the obtained tomato fruit video; S104: inputting the detection result of the tomato fruit detection model into the tomato fruit tracking model to obtain the tomato fruit tracking counting result; S105: fitting the tomato fruit maturity time data at each key stage before reaching full maturity using a normal distribution function to obtain a probability distribution of the tomato maturity time; S106: Calculate the predicted yield of tomatoes in the greenhouse by combining the established normal distribution function with the counting results of the tomato tracking model.

[0006] As a preferred technical solution of the present application, the calculation formula for predicting the output in step S106 is: In the formula, n represents the forecast week number, Y n represents the forecast output for the next week, S i Indicates the three stages of cherry tomatoes before they reach full maturity. S1, S2, and S3 represent the green maturity stage, the hard maturity stage, and the color change stage, respectively. si represents the number of fruits detected by the fruit tracking model, P si Indicates S i The probability of the predicted days corresponding to the normal distribution function of the stage, X represents the predicted time.

[0007] As a preferred technical solution of the present application, the step S102 specifically includes the following operation steps: A1: Collect tomato fruit images in the greenhouse and use data augmentation methods to expand the images to 10 times the number; A2: LabelImg annotation tool is used to annotate tomato fruit images. During the annotation process, tomatoes are divided into four categories, namely Green, Turning, Orange and Red, corresponding to the four key stages of tomato fruit ripening, namely green ripening, color change, hard ripening and full ripening, to establish a tomato fruit dataset; and the tomato fruit dataset is divided into a tomato fruit training set and a tomato fruit validation set; A3: The tomato fruit detection model is any one of the target detection models YOLO, Faster RCNN and SSD; the prediction method used is linear regression; the loss function used is composed of three parts, the classification loss is calculated by Focal Loss, the confidence loss is calculated by BCE Loss, and the coordinate loss is calculated by GIoU Loss; S4: The tomato fruit training set is used to train the fruit detection model, the tomato fruit verification set is used to verify the accuracy of the tomato fruit detection model, and the tomato fruit detection model with the best performance is obtained.

[0008] As a preferred technical solution of the present application, the step S103 specifically includes the following operation steps: B1: Collect tomato fruit videos in the greenhouse, use the Darklabel multi-target tracking and annotation tool to annotate the tomato fruit videos frame by frame, customize the annotation format, and make the annotated categories consistent with the tomato fruit dataset to establish a tomato fruit tracking dataset; and divide the tomato fruit tracking dataset into a tomato fruit tracking training set and a tomato fruit tracking verification set; B2: The tomato fruit tracking model is any one of the multi-target tracking algorithms Deep SORT, MOTDT and CenterTrack; the trajectory prediction method used is the Kalman filter; the trajectory matching algorithm used is the Hungarian algorithm; B3: The tomato fruit tracking training set is used to train the tomato fruit tracking model, and the tomato fruit tracking verification set is used to verify the tomato fruit tracking model, and the tomato fruit tracking model with the best performance is obtained.

[0009] As a preferred technical solution of the present application, the step S105 specifically includes the following operation steps: C1: Use ultra-high-definition cameras to continuously monitor the greenhouse, write a python script to control the camera to take photos at fixed monitoring points every day, record the tomato ripening time after checking the photos, obtain the tomato ripening time data, and store the data in the tomato ripening time database; C2: Obtain the time when tomato fruits reach maturity at each stage from the tomato maturity time database, and use the normal distribution function to fit the time when tomato fruits reach maturity at each stage, and obtain the mathematical expectation and standard deviation of the normal distribution curve after fitting. The horizontal axis of the fitting curve is time, and the vertical axis is the probability value, and the probability distribution is calculated.

[0010] As a preferred technical solution of the present application, the probability distribution formula is: In the formula, μ represents the mathematical expectation, σ represents the standard deviation, x represents the maturity time in days, and f(x) represents the probability of maturity in x days.

[0011] A tomato yield prediction system based on normal distribution function comprises a data acquisition module, a tomato fruit data set establishment module, a tomato fruit tracking data set establishment module, a tomato detection model establishment module, a tomato tracking model establishment module, a fruit ripening time fitting module, a yield prediction module and a result display module.

[0012] As a preferred technical solution of the present application, the data acquisition module acquires tomato fruit images and videos, the tomato fruit data set establishment module performs data enhancement on the tomato fruit images, labels them according to the four key stages of tomato fruit ripening, and establishes a tomato fruit data set, the tomato fruit tracking data set establishment module labels the tomato fruit videos, and establishes a tomato fruit tracking data set, the tomato detection model establishment module establishes a tomato detection model, the tomato tracking model establishment module establishes a tomato fruit tracking model, and the result display module displays the tomato fruit detection results, the tomato fruit tracking counting results, and the measurement prediction results.

[0013] Beneficial effects: Compared with the prior art, the present invention provides a tomato yield prediction method and system based on the normal distribution function, which has the following beneficial effects: This tomato yield prediction method and system based on the normal distribution function fully considers the four key stages of tomato fruit ripening, and fits the normal distribution curve of ripening time by counting the time it takes to reach full maturity in the first three stages, and completes the tomato yield prediction in combination with the fruit tracking and counting results.

[0014] High accuracy: The normal distribution function is a reliable prediction model widely used in statistics and is suitable for describing many natural phenomena, including many agricultural data. Using the normal distribution function for tomato yield prediction can better capture the ripening time of tomato fruits and provide accurate results for yield prediction.

[0015] Strong flexibility: The normal distribution function can be adjusted and optimized according to different maturity time databases and requirements. The method of the present invention can adapt to the growth of tomatoes in different regions and climatic conditions, and has strong practicality and applicability. The entire tomato yield prediction method has a simple structure, is easy to operate, and has better results than traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a schematic diagram of the overall structure of a tomato yield prediction method based on a normal distribution function according to the present invention.

[0017] Figure 2 The figure is a schematic diagram of fruits at four key stages of tomato fruit ripening in a tomato yield prediction method based on a normal distribution function of the present invention.

[0018] Figure 3The present invention is a schematic diagram of the curve fitting results of the time from the green ripe stage to the fully ripe stage in a tomato yield prediction method based on the normal distribution function.

[0019] Figure 4 The present invention is a schematic diagram of the curve fitting results of the time from the color change period to the full maturity period in a tomato yield prediction method based on the normal distribution function.

[0020] Figure 5 The present invention is a schematic diagram of the curve fitting results of the time from the hard-ripe stage to the fully ripe stage in a tomato yield prediction method based on the normal distribution function.

[0021] Figure 6 The schematic diagram of a tomato yield prediction system based on a normal distribution function of the present invention is shown in FIG. DETAILED DESCRIPTION

[0022] The technical scheme of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments, but it will be understood by those skilled in the art that the following described embodiments are part of embodiments of the present invention, rather than all embodiments, and are only used to illustrate the present invention, and should not be considered as limiting the scope of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of the present invention. If specific conditions are not specified in the embodiments, they are carried out according to normal conditions or conditions recommended by the manufacturer. If the manufacturer is not specified in the reagents or instruments used, they are all conventional products that can be purchased commercially.

[0023] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0024] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] like Figure 1-6 As shown, a tomato yield prediction method based on a normal distribution function includes the following steps: S101: when the tomato fruit enters the green ripe stage, obtaining tomato fruit images and videos; S102: constructing a tomato fruit detection model according to the obtained tomato fruit image; S103: constructing a tomato fruit tracking model according to the obtained tomato fruit video; S104: inputting the detection result of the tomato fruit detection model into the tomato fruit tracking model to obtain the tomato fruit tracking counting result; S105: fitting the tomato fruit maturity time data at each key stage before reaching full maturity using a normal distribution function to obtain a probability distribution of the tomato maturity time; S106: Calculate the predicted yield of tomatoes in the greenhouse by combining the established normal distribution function with the counting results of the tomato tracking model.

[0026] Using the normal distribution function to predict tomato yield can better capture the maturity time of tomato fruits, provide accurate results for yield prediction, and can adapt to the growth of tomatoes in different regions and climatic conditions, with strong practicality and applicability.

[0027] Furthermore, the calculation formula for predicting output in step S106 is: In the formula, n represents the forecast week number, Y n represents the forecast output for the next week, S i Indicates the three stages of cherry tomatoes before they reach full maturity. S1, S2, and S3 represent the green maturity stage, the hard maturity stage, and the color change stage, respectively. si represents the number of fruits detected by the fruit tracking model, P si Indicates S i The probability of the predicted days corresponding to the normal distribution function of the stage, X represents the predicted time.

[0028] Furthermore, step S102 specifically includes the following steps: A1: Collect tomato fruit images in the greenhouse and use data augmentation methods to expand the images to 10 times the number; A2: LabelImg annotation tool is used to annotate tomato fruit images. During the annotation process, tomatoes are divided into four categories: Green, Turning, Orange, and Red, corresponding to the four key stages of tomato fruit ripening, namely green ripening, color change, hard ripening, and full ripening. A tomato fruit dataset is established; and the tomato fruit dataset is divided into a tomato fruit training set and a tomato fruit validation set. A3: The tomato fruit detection model is any one of the target detection models YOLO, Faster RCNN and SSD; the prediction method used is linear regression; the loss function used is composed of three parts: the classification loss is calculated by Focal Loss, the confidence loss is calculated by BCE Loss, and the coordinate loss is calculated by GIoU Loss; S4: The fruit detection model is trained using the tomato fruit training set, and the accuracy of the tomato fruit detection model is verified using the tomato fruit validation set, and the tomato fruit detection model with the best performance is obtained.

[0029] Furthermore, step S103 specifically includes the following steps: B1: Collect tomato fruit videos in the greenhouse, use the Darklabel multi-target tracking and annotation tool to annotate the tomato fruit videos frame by frame, customize the annotation format, and make the annotated categories consistent with the tomato fruit dataset to establish a tomato fruit tracking dataset; and divide the tomato fruit tracking dataset into a tomato fruit tracking training set and a tomato fruit tracking verification set; B2: The tomato fruit tracking model is any one of the multi-target tracking algorithms Deep SORT, MOTDT and CenterTrack; the trajectory prediction method used is the Kalman filter; the trajectory matching algorithm used is the Hungarian algorithm; B3: The tomato fruit tracking training set is used to train the tomato fruit tracking model, and the tomato fruit tracking verification set is used to verify the tomato fruit tracking model, and the tomato fruit tracking model with the best performance is obtained.

[0030] Furthermore, step S105 specifically includes the following steps: C1: Use ultra-high-definition cameras to continuously monitor the greenhouse, write a python script to control the camera to take photos at fixed monitoring points every day, record the tomato ripening time after checking the photos, obtain the tomato ripening time data, and store the data in the tomato ripening time database; C2: Obtain the time when tomato fruits reach maturity at each stage from the tomato maturity time database, and use the normal distribution function to fit the time when tomato fruits reach maturity at each stage. Obtain the mathematical expectation and standard deviation of the normal distribution curve after fitting. The horizontal axis of the fitting curve is time, and the vertical axis is the probability value. Calculate the probability distribution.

[0031] Furthermore, the probability distribution formula is: In the formula, μ represents the mathematical expectation, σ represents the standard deviation, x represents the maturity time in days, and f(x) represents the probability of maturity in x days.

[0032] Figure 2 These are pictures of tomatoes at four key stages of ripening. In the pictures, a, b, c, and d represent tomatoes in the green ripe stage, color change stage, hard ripe stage, and fully ripe stage, respectively.

[0033] Figure 3 , 4 5 are the results of curve fitting for the time from green ripening stage, color change stage and hard ripening stage to full ripening stage according to the measured data of the present invention, and the mathematical expectation μ and variance σ of the curve are 2 They are: 14.7 and 3.8, 9.8 and 3.8, 2.8 and 2.2 respectively.

[0034] A tomato yield prediction system based on normal distribution function comprises a data acquisition module, a tomato fruit data set establishment module, a tomato fruit tracking data set establishment module, a tomato detection model establishment module, a tomato tracking model establishment module, a fruit ripening time fitting module, a yield prediction module and a result display module.

[0035] A data acquisition module for acquiring tomato fruit images and videos; The tomato fruit dataset establishment module is used to perform data enhancement on tomato fruit images, annotate them according to the four key stages of tomato fruit ripening, establish a tomato fruit dataset, and divide the tomato fruit dataset into a tomato fruit training set and a tomato fruit validation set; A tomato fruit tracking dataset establishment module is used to annotate tomato fruit videos, establish a tomato fruit tracking dataset, and divide the tomato fruit tracking dataset into a tomato fruit tracking training set and a tomato fruit tracking verification set; A tomato detection model building module is used to build a tomato detection model, train the tomato detection model using a tomato fruit training set, verify the tomato detection model using a tomato fruit verification set, and obtain a tomato fruit detection model with the best performance; A tomato tracking model building module is used to build a tomato fruit tracking model, train the tomato fruit tracking model using a tomato fruit tracking training set, verify the tomato fruit tracking model using a tomato fruit tracking verification set, and obtain a tomato fruit tracking model with the best performance; The fruit ripening time fitting module is used to fit the data of the ripening time of tomato fruits in the first three key stages that have not reached full maturity, and obtain fitting curves respectively. The horizontal axis of the curve is time in days, and the vertical axis is the probability value; The yield prediction module inputs the results of the tomato fruit detection model into the tomato fruit tracking model to obtain the tomato fruit tracking and counting results. Then, the module combines the tomato fruit tracking and counting results with the fitting curve to calculate the yield prediction results for a period of time in the future. The result display module is used to display the tomato fruit detection results, tomato fruit tracking and counting results and measurement prediction results.

[0036] It should be noted that, in this article, relational terms such as first and second (number one, number two), etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0037] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A tomato yield prediction method based on normal distribution function, Features: The steps include: S101: when the tomato fruit enters the green ripe stage, obtaining tomato fruit images and videos; S102: constructing a tomato fruit detection model according to the obtained tomato fruit image; S103: constructing a tomato fruit tracking model according to the obtained tomato fruit video; S104: inputting the detection result of the tomato fruit detection model into the tomato fruit tracking model to obtain the tomato fruit tracking counting result; S105: fitting the tomato fruit maturity time data at each key stage before reaching full maturity using a normal distribution function to obtain a probability distribution of the tomato maturity time; S106: Calculate the predicted yield of tomatoes in the greenhouse by combining the established normal distribution function with the counting results of the tomato tracking model.

2. A tomato yield prediction method based on normal distribution function according to claim 1, Features: The calculation formula for predicting the output in step S106 is: In the formula, n represents the forecast week number, Y n represents the forecast output for the next week, S i Indicates the three stages before cherry tomatoes reach full maturity. 1 , S 2 , S 3 Represent green ripening stage, hard ripening stage and color change stage, N si represents the number of fruits detected by the fruit tracking model, P si Indicates S i The probability of the predicted days corresponding to the normal distribution function of the stage, X represents the predicted time.

3. A tomato yield prediction method based on normal distribution function according to claim 1, Features: The step S102 specifically includes the following steps: A1: Collect tomato fruit images in the greenhouse and use data augmentation methods to expand the images to 10 times the number; A2: LabelImg annotation tool is used to annotate tomato fruit images. During the annotation process, tomatoes are divided into four categories, namely Green, Turning, Orange and Red, corresponding to the four key stages of tomato fruit ripening, namely green ripening, color change, hard ripening and full ripening, to establish a tomato fruit dataset; and the tomato fruit dataset is divided into a tomato fruit training set and a tomato fruit validation set; A3: The tomato fruit detection model is any one of the target detection models YOLO, Faster RCNN and SSD; the prediction method used is linear regression; the loss function used is composed of three parts, the classification loss is calculated by Focal Loss, the confidence loss is calculated by BCE Loss, and the coordinate loss is calculated by GIoU Loss; S4: The tomato fruit training set is used to train the fruit detection model, the tomato fruit verification set is used to verify the accuracy of the tomato fruit detection model, and the tomato fruit detection model with the best performance is obtained.

4. A tomato yield prediction method based on normal distribution function according to claim 1, Features: The step S103 specifically includes the following steps: B1: Collect tomato fruit videos in the greenhouse, use the Darklabel multi-target tracking and annotation tool to annotate the tomato fruit videos frame by frame, customize the annotation format, and make the annotated categories consistent with the tomato fruit dataset to establish a tomato fruit tracking dataset; and divide the tomato fruit tracking dataset into a tomato fruit tracking training set and a tomato fruit tracking verification set; B2: The tomato fruit tracking model is any one of the multi-target tracking algorithms Deep SORT, MOTDT and CenterTrack; the trajectory prediction method used is the Kalman filter; the trajectory matching algorithm used is the Hungarian algorithm; B3: The tomato fruit tracking training set is used to train the tomato fruit tracking model, and the tomato fruit tracking verification set is used to verify the tomato fruit tracking model, and the tomato fruit tracking model with the best performance is obtained.

5. The tomato yield prediction method based on normal distribution function according to claim 1, Features: The step S105 specifically includes the following steps: C1: Use ultra-high-definition cameras to continuously monitor the greenhouse, write a python script to control the camera to take photos at fixed monitoring points every day, record the tomato ripening time after checking the photos, obtain the tomato ripening time data, and store the data in the tomato ripening time database; C2: Obtain the time when tomato fruits reach maturity at each stage from the tomato maturity time database, and use the normal distribution function to fit the time when tomato fruits reach maturity at each stage, and obtain the mathematical expectation and standard deviation of the normal distribution curve after fitting. The horizontal axis of the fitting curve is time, and the vertical axis is the probability value, and the probability distribution is calculated.

6. A tomato yield prediction method based on normal distribution function according to claim 5, Features: The probability distribution formula is: In the formula, μ represents the mathematical expectation, σ represents the standard deviation, x represents the maturity time in days, and f(x) represents the probability of maturity in x days.

7. A tomato yield prediction system based on normal distribution function, Features: It includes a data acquisition module, a tomato fruit data set building module, a tomato fruit tracking data set building module, a tomato detection model building module, a tomato tracking model building module, a fruit ripening time fitting module, a yield prediction module and a result display module.

8. A tomato yield prediction system based on normal distribution function according to claim 7, Features: The data acquisition module acquires tomato fruit images and videos, the tomato fruit data set establishment module performs data enhancement on tomato fruit images, labels them according to four key stages of tomato fruit ripening, and establishes a tomato fruit data set, the tomato fruit tracking data set establishment module labels tomato fruit videos, and establishes a tomato fruit tracking data set, the tomato detection model establishment module establishes a tomato detection model, the tomato tracking model establishment module establishes a tomato fruit tracking model, and the result display module displays tomato fruit detection results, tomato fruit tracking counting results, and measurement prediction results.

Citation Information

Cited By

  • Tomato picking method and system based on binocular vision and machine readable storage medium

    CN120982297A