Horticultural harvest prediction system

Abstract

To provide a harvest time prediction system for horticultural products capable of predicting harvest time of the horticultural products within a certain range (range from the earliest coloring scheduled date of flower buds of the horticultural products to the latest coloring scheduled date of the flower buds of the horticultural products).SOLUTION: A horticultural product harvest time prediction system 1 determines, as a flower bud diameter range at the time of coloring of horticultural products, a range from the flower bud diameter minimum value at the time of coloring when the flower bud diameter at the time of flower bud coloring was the smallest to the flower bud diameter maximum value at the time of coloring when the flower bud diameter at the time of flower bud coloring was the largest, based on past sample data of horticultural products cultivated in the past, estimates how many days it will take from now for the flower buds of currently cultivated horticultural products to color, based on current sample data of the currently cultivated horticultural products and the flower bud diameter range at the time of coloring, and predicts, as harvest time of the currently cultivated horticultural products, the time from the earliest scheduled coloring date that is estimated to be the earliest coloring date to the latest scheduled coloring date that is estimated to be the latest coloring date.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a system for predicting the harvest time of horticultural products such as flowers and vegetables. [Background technology]

[0002] Conventionally, a system for predicting the harvest time of horticultural products has been proposed (see, for example, Patent Document 1). In this conventional system, the harvest time of a tomato fruit is determined based on the ratio of the diameter to the height of the captured image of the fruit. For example, when the ratio of the diameter to the height of the captured image of the tomato fruit exceeds a predetermined threshold (for example, the value of diameter / height is 1.2), it is determined that the tomato is ready to be harvested. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-154510 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in conventional systems, the time when the size of the captured tomato fruit reaches a certain value (threshold) is determined to be the time to harvest the tomatoes (including all tomatoes, including those not captured in the image), but because the growth status of each fruit varies, fruits that are not suitable for harvest (those that are not yet ready for harvest or have passed their harvest time) are also determined to be ready for harvest. Conventional systems cannot predict the time to harvest tomatoes with a certain range (the range from the earliest to latest date for harvest).

[0005] Furthermore, tomatoes (especially tomatoes supplied to the market) are generally grown in greenhouses. In greenhouse cultivation, environmental control is easy and cultivation can be carried out in a uniform environment. Therefore, crops grown in greenhouses such as tomatoes are less affected by weather conditions (temperature, humidity, etc.) and environmental conditions (hours of sunlight, etc.), making it easy to predict the growth of the fruit. On the other hand, open-field cultivation is often affected by weather and environmental conditions. Therefore, it is difficult to predict the growth of horticultural products grown in the open, and it is also difficult to predict the harvest time. While conventional systems can determine the harvest time for fruits grown in greenhouses, it is extremely difficult to predict the harvest time for horticultural products grown in the open.

[0006] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a system that can predict the harvest time of horticultural products within a certain range (the range from the earliest scheduled date on which the flower buds of the horticultural products will color to the latest scheduled date on which they will color). [Means for solving the problem]

[0007] The harvest time prediction system for horticultural products of the present invention includes a means for storing, for horticultural products cultivated in the past, data on the flower bud diameters of a plurality of reference samples contained in the horticultural product and data on whether the flower buds of the reference samples are colored as past sample data of the horticultural product, and a means for determining, based on the past sample data, the width from the minimum flower bud diameter at coloring, which is the smallest diameter of the flower buds when colored, to the maximum flower bud diameter at coloring, which is the largest diameter of the flower buds when colored, as the flower bud diameter width at coloring of the horticultural product. means for acquiring, for the horticultural product currently being cultivated, data on the flower bud diameter of a prediction sample contained in the horticultural product as current sample data for the horticultural product; and means for estimating how many days from the present time it will take for the flower buds of the horticultural product currently being cultivated to color based on the current sample data and the flower bud diameter width at coloring, and predicting the period from the earliest expected coloring date, which is estimated to be the earliest when coloring will occur, to the latest expected coloring date, which is estimated to be the latest when coloring will occur, as the harvest time for the horticultural product currently being cultivated.

[0008] With this configuration, the flower bud diameter width at coloration (the width from the minimum to maximum flower bud diameter) of a horticultural product is determined based on past sample data obtained from horticultural products grown in the past (data on the flower bud diameter of multiple reference samples from horticultural products grown in the past and data on whether the flower buds have colored or not), and the harvest time of the horticultural product (the period from the earliest to latest scheduled coloration date) can be predicted based on current sample data obtained from the horticultural product currently being grown (flower bud diameter data of prediction samples from the horticultural product currently being grown) and the above-mentioned flower bud diameter width at coloration.In this way, by using the flower bud diameter width at coloration (the width of the flower bud diameter when the flower buds colored) of horticultural products grown in the past, the harvest time of the horticultural product currently being grown (the period from the earliest to latest scheduled coloration date) can be appropriately predicted.

[0009] In addition, in the horticultural product harvest time prediction system of the present invention, the storing means stores, as the past sample data, data regarding the date on which the flower bud diameter of the reference sample was measured, in correspondence with the flower bud diameter of the reference sample and data on whether the flower bud diameter has colored or not; the determining means calculates, based on the past sample data, a growth model function that indicates the change in flower bud diameter size relative to the number of days elapsed since a specified reference date, and calculates a range of days corresponding to the flower bud diameter range at coloring; the acquiring means acquires, as the current sample data, data regarding the date on which the flower bud diameter of the prediction sample was measured, along with data on the flower bud diameter of the prediction sample; and the predicting means may determine the earliest expected coloring date and the latest expected coloring date based on the current sample data, the growth model function, and the range of days.

[0010] With this configuration, a growth model function (a function showing changes in flower bud diameter over the number of days elapsed) for a horticultural product is calculated based on past sample data (including data on the dates the flower bud diameter was measured) obtained from a horticultural product grown in the past, a range of days corresponding to the flower bud diameter range at coloration is determined, and the harvest time (earliest and latest expected coloration dates) for the horticultural product can be predicted based on current sample data (including data on the dates the flower bud diameter of the prediction sample was measured) obtained from the horticultural product currently being cultivated, the growth model function, and the range of days. In this way, by using the growth model function (a representative function derived from multiple reference samples) derived from past sample data of horticultural products grown in the past and the range of days (corresponding to the flower bud diameter range at coloration), it is possible to appropriately predict the harvest time for a horticultural product currently being cultivated.

[0011] In addition, in the horticultural product harvest time prediction system of the present invention, the acquiring means acquires, as the current sample data, for each of the plurality of prediction samples, data on the flower bud diameter of the prediction sample, as well as data on the date the flower bud diameter of the prediction sample was measured, at least twice on different days, and the predicting means calculates, based on the current sample data, a growth model function for each of the plurality of prediction samples that indicates the change in flower bud diameter size relative to the number of days elapsed from a specified reference date, estimates the expected coloring date of the flower buds of each of the plurality of prediction samples based on the growth model function and the flower bud diameter width at coloring, and predicts the harvest time of the horticultural product currently being cultivated based on the expected coloring date of the flower buds of each of the plurality of prediction samples.

[0012] According to this configuration, a growth model function (a function showing the change in flower bud diameter over the number of days elapsed) for each of the multiple prediction samples is calculated based on current sample data obtained from the horticultural products currently being cultivated (which includes data on the date the flower bud diameter of each of the multiple prediction samples was measured, and is obtained at least twice on different days), and the expected coloring date (earliest expected coloring date and latest expected coloring date) for each of the multiple prediction samples is estimated, making it possible to appropriately predict the harvest time for the horticultural products currently being cultivated.

[0013] In addition, in the horticultural product harvest time prediction system of the present invention, the prediction means may calculate the number of horticultural products currently being cultivated that are expected to be harvested each day at the harvest time based on the expected coloring time of the flower buds of each of the multiple prediction samples and the total number of horticultural products currently being cultivated that are expected to be harvested at the harvest time.

[0014] With this configuration, it is possible to appropriately calculate the number of horticultural products currently being cultivated that are expected to be harvested each day at the harvest time (e.g., June 29th: 250, June 30th: 450, ..., July 7th: 200) from the expected coloring period of each flower bud of multiple prediction samples (e.g., Sample 1: June 29th to July 2nd, Sample 2: June 30th to July 5th, Sample 3: July 3rd to July 7th) and the total expected harvest number of horticultural products currently being cultivated at the harvest time (e.g., 3,000).

[0015] In addition, in the horticultural product harvest time prediction system of the present invention, the storage means stores multiple pieces of past sample data in correspondence with at least one of the production area, production year, crop type, and variety of the horticultural product, and when at least one of the production area, production year, crop type, and variety is input by a user, the determining means may determine the flower bud diameter width at coloring based on past sample data selected from the multiple pieces of past sample data in accordance with the user input.

[0016] According to this configuration, the flower bud diameter at coloring is determined based on past sample data selected in accordance with user input (at least one of the production area, production year, crop type, and variety of the horticultural product), and the harvest time of the horticultural product currently being cultivated is predicted. By selecting and using past sample data that has a cultivation environment (weather conditions, environmental conditions, etc.) similar to that of the horticultural product currently being cultivated based on at least one of the production area (field, region, prefecture, etc.), production year, crop type, and variety input by the user, it is possible to appropriately predict the harvest time of the horticultural product.

[0017] The method of the present invention is a method executed by a harvest time prediction system for horticultural products, which includes storing data on the flower bud diameters of a plurality of reference samples contained in a horticultural product cultivated in the past and data on whether the flower buds of the reference samples are colored as past sample data for the horticultural product, and calculating, based on the past sample data, a range from a minimum value of the flower bud diameter at coloring, which is the smallest value when the flower bud diameter when the flower buds colored, to a maximum value of the flower bud diameter at coloring, which is the largest value when the flower buds colored, as the flower bud diameter at coloring of the horticultural product. The method includes determining the flower bud diameter as a diameter width, obtaining, for the horticultural product currently being cultivated, data on the flower bud diameter of a prediction sample contained in the horticultural product as current sample data for the horticultural product, estimating how many days from the present time it will take for the flower buds of the horticultural product currently being cultivated to color based on the current sample data and the flower bud diameter width at coloring, and predicting the period from the earliest expected coloring date, which is estimated to be the earliest when coloring will occur, to the latest expected coloring date, which is estimated to be the latest when coloring will occur, as the harvest time for the horticultural product currently being cultivated.

[0018] Like the system described above, this method determines the flower bud diameter width at coloration (the width from the minimum to maximum flower bud diameter) of a horticultural product based on past sample data (flower bud diameter data of multiple reference samples from previously cultivated horticultural products and data on whether the flower buds have colored) obtained from that horticultural product. The harvest time of the horticultural product (the period from the earliest to latest expected coloration date) can be predicted based on current sample data (flower bud diameter data of prediction samples from currently cultivated horticultural products) obtained from the horticultural product being cultivated and the above-mentioned flower bud diameter width at coloration. In this way, by using the flower bud diameter width at coloration (the width of the flower bud diameter when the flower buds colored) of previously cultivated horticultural products, the harvest time of the currently cultivated horticultural product (the period from the earliest to latest expected coloration date) can be appropriately predicted.

[0019] The program of the present invention is a program executed in a harvest time prediction system for horticultural products, wherein data on the flower bud diameters of a plurality of reference samples contained in a horticultural product cultivated in the past and data on whether the flower buds of the reference samples have colored are stored in the harvest time prediction system as past sample data for the horticultural product, and the program causes the harvest time prediction system to calculate a value ranging from a minimum flower bud diameter at coloring, which was the smallest diameter when the flower buds had colored, to a maximum flower bud diameter at coloring, based on the past sample data. The system executes the following processes: determining the width up to the maximum bud diameter as the flower bud diameter width at coloring of the horticultural product; acquiring, for the horticultural product currently being cultivated, data on the flower bud diameter of a prediction sample contained in the horticultural product as current sample data of the horticultural product; and estimating how many days from the present it will be until the flower buds of the horticultural product currently being cultivated will color based on the current sample data and the flower bud diameter width at coloring, and predicting the period from the earliest expected coloring date, which is estimated to be the earliest when coloring will occur, to the latest expected coloring date, which is estimated to be the latest when coloring will occur, as the harvest time of the horticultural product currently being cultivated.

[0020] Like the system described above, this program determines the flower bud diameter width at coloration (the width from the minimum to maximum flower bud diameter) of a horticultural product based on past sample data obtained from previously cultivated horticultural products (data on the flower bud diameter of multiple reference samples from previously cultivated horticultural products and data on whether the flower buds have colored). Based on current sample data obtained from currently cultivated horticultural products (data on the flower bud diameter of prediction samples from currently cultivated horticultural products) and the above flower bud diameter width at coloration, the harvest time of the horticultural product (the period from the earliest to latest expected coloration date) can be predicted. In this way, by using the flower bud diameter width at coloration (the width of the flower bud diameter when the flower buds colored) of previously cultivated horticultural products, the harvest time of the currently cultivated horticultural product (the period from the earliest to latest expected coloration date) can be appropriately predicted. [Effects of the Invention]

[0021] According to the present invention, it is possible to predict the harvest time of a horticultural product within a certain range (the range from the earliest expected date of coloring of the flower buds of the horticultural product to the latest expected date of coloring). [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a diagram showing a configuration of a harvest time prediction system according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram showing an example of a screen for a user to input data from a user terminal. [Figure 3] FIG. 2 is a diagram showing an example of data stored in a harvest time prediction system (storage unit). [Figure 4] FIG. 10 is a diagram showing an example of calculating the minimum and maximum values ​​of flower bud diameter at coloration using past sample data selected based on user input. [Figure 5] FIG. 2 is a diagram illustrating a harvest time prediction (prediction based on past growth transitions) in the first embodiment. [Figure 6] FIG. 2 is a sequence diagram illustrating the flow of processing in the harvest time prediction system according to the first embodiment. [Figure 7] FIG. 10 is a diagram illustrating a harvest time prediction (prediction based on the current growth transition) in the second embodiment. [Figure 8] FIG. 10 is a diagram illustrating a method for predicting the number of trees to be harvested each day in the second embodiment. [Figure 9] FIG. 10 is a sequence diagram illustrating the flow of processing in the harvest time prediction system according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0023] A system for predicting the harvest time of horticultural products according to an embodiment of the present invention will be described below with reference to the drawings. In this embodiment, a system for predicting the harvest time of horticultural products such as flowers grown outdoors will be described as an example. The harvest time prediction in this embodiment can be realized by operating a computer using a program stored in the system's memory.

[0024] (First embodiment) The configuration of a harvest time prediction system according to a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing the configuration of the harvest time prediction system according to this embodiment. As shown in FIG. 1, the harvest time prediction system 1 is configured, for example, by a server device, and is connected to an external user terminal U via a network N such as the Internet. The user terminal U is a terminal device such as a mobile terminal such as a smartphone or a personal computer, and is used by a user (grower) in a field where flowers are grown (for example, field A, field B, etc.).

[0025] 1, this harvest time prediction system 1 includes a receiving unit 2 that receives various data transmitted from a user device, and a transmitting unit 3 that transmits various data to the user device. Harvest time prediction system 1 also includes a memory unit 4 that stores various data for predicting the harvest time, and a control unit 5 that executes various processes for predicting the harvest time. The control unit 5 includes, as functional blocks, a past sample data reading unit 6, a decision processing unit 7, a current sample data acquisition unit 8, and a prediction processing unit 9.

[0026] The memory unit 4 stores, as past sample data, data on the bud diameters of multiple reference samples of flowers cultivated in the past, input from the user terminal U, and data on whether the buds of the reference samples are colored. In this case, the memory unit 4 stores, as past sample data, data on the dates on which the bud diameters of the reference samples were measured, in association with data on the bud diameters of the reference samples and whether the bud diameters are colored. The memory unit 4 also stores the multiple past sample data in association with at least one of the flower production area, production year, crop type, and variety.

[0027] 2 is a diagram showing an example of a screen for user input of data from the user terminal U. As shown in FIG. 2, from the user terminal U, data such as the flower sample number, the date and measurement value (first and second) on which the flower bud diameter of the flower was measured, the date on which the flower bud colored and the flower bud diameter can be input. The data input from the user terminal U is transmitted to the harvest time prediction system 1 and stored in the memory unit 4.

[0028] Fig. 3 is a diagram showing an example of data stored in the memory unit 4 of the harvest time prediction system 1. In the example of Fig. 3, data on the flower production year, crop type, variety, illuminated / non-illuminated, field name, control number (sample number), date and measurement value (first and second) when the flower bud diameter was measured, and date when the flower bud colored and flower bud diameter are stored.

[0029] The past sample data reading unit 6 has a function of reading out past sample data stored in the storage unit 4. When the user inputs at least one of the production area, production year, crop type, and variety, the past sample data reading unit 6 has a function of reading out past sample data selected in accordance with the user input from among multiple past sample data. The past sample data read out from the storage unit 4 is used by the determination processing unit 7.

[0030] The determination processing unit 7 has a function of determining, based on past sample data read from the storage unit 4, the flower bud diameter width at coloration, which is the range from the minimum flower bud diameter at coloration, where the flower bud diameter was smallest when the flower bud colored, to the maximum flower bud diameter at coloration, where the flower bud diameter was largest when the flower bud colored. In this case, the determination processing unit 7 calculates, based on the past sample data, a growth model function (in the example of Figure 5, the calibration curve y = ax + b) that indicates the change in flower bud diameter size with respect to the number of days elapsed since a predetermined reference date, and calculates the range of days corresponding to the flower bud diameter width at coloration. Furthermore, when the user inputs at least one of the production area, production year, crop type, and variety, the determination processing unit 7 can determine the flower bud diameter width at coloration based on past sample data selected in accordance with the user input from among multiple past sample data.

[0031] Figure 4 is a diagram showing an example of calculating the minimum and maximum flower bud diameter at coloring using past sample data selected based on user input. In the example of Figure 4, the minimum flower bud diameter at coloring is calculated to be "9.125 mm" and the maximum flower bud diameter at coloring is calculated to be "9.725 mm" using past sample data selected based on user input of "Production area: Field A, Production year 2019, Cropping type: Shipped in August, Variety: Variety A, Illumination / non-illumination: Non-illumination."

[0032] The current sample data acquisition unit 8 has a function of acquiring, for a flower currently being cultivated, data on the bud diameter of a prediction sample contained in that flower as the current sample data of the flower. In this case, the current sample data acquisition unit 8 can acquire, as the current sample data, data on the bud diameter of the prediction sample as well as data on the date on which the bud diameter of the prediction sample was measured.

[0033] The prediction processing unit 9 has the function of estimating how many days from now it will be until the flower buds of the flower currently being cultivated will color, based on the current sample data and the flower bud diameter width at coloring, and predicting the period from the earliest expected coloring date, which is estimated to be the earliest, to the latest expected coloring date, which is estimated to be the latest, as the harvest time for the flower currently being cultivated. In this case, the prediction processing unit 9 can determine the earliest expected coloring date and the latest expected coloring date based on the current sample data, the growth model function, and the range of days.

[0034] Fig. 5 is a diagram illustrating harvest time prediction (prediction based on past growth trends) in this embodiment. As shown in Fig. 5, in harvest time prediction in this embodiment, a growth model function (in Fig. 5, calibration curve y = ax + b) that represents the degree of growth of flower bud diameter y with respect to the number of days x is calculated based on flower bud diameter data (past sample data) measured on at least two different dates for multiple reference samples (samples 1, 2, 3, etc. in Fig. 5), and a range of days (range of days until coloring) corresponding to the flower bud diameter range at coloring (range of flower bud diameter at coloring) is calculated. Then, using the growth model function and the range of days, the earliest expected coloring date (in Figure 5, the date d1 days after the measurement date at the time of prediction) and the latest expected coloring date (in Figure 5, the date d2 days after the measurement date at the time of prediction) are calculated from the current date and the bud diameter (in Figure 5, the measurement date at the time of prediction and the bud diameter at the time of prediction), and the harvest time of the flowers (in Figure 5, coloring will occur d1 to d2 days after the measurement date) is predicted.

[0035] The operation of the harvest time prediction system 1 configured as above will be described with reference to the sequence diagram of FIG.

[0036] In the harvest time prediction system 1 of the first embodiment, data such as the flower sample numbers of multiple reference samples (e.g., samples 1, 2, and 3), the dates and measurements (first and second times) on the flower bud diameters, the dates on which the buds turned colored, and the bud diameters are input to the user terminal U of a user who grows flowers as needed (see FIG. 2). The data input from the user terminal U is sent to the harvest time prediction system 1 and stored in the memory unit 4 of the harvest time prediction system 1 as past sample data (see FIG. 3).

[0037] When predicting the harvest time, as shown in Figure 4, data such as "production area, crop type, variety, illuminated / non-illuminated" used for predicting the harvest time, and data on the bud diameter and date of the prediction sample contained in the flowers currently being cultivated (current sample data) are input from the user terminal U (S10), and the data input from the user terminal U (user input) is sent to the harvest time prediction system 1 (S11).

[0038] In the harvest time prediction system 1, past sample data with similar cultivation environments (weather conditions, environmental conditions, etc.) is read out based on the data on "production area, crop type, variety, and illuminated / non-illuminated" input from the user terminal U (S12), a growth model function (in Figure 5, the calibration curve y = ax + b) is calculated, and the range of days (range of days until coloring) corresponding to the flower bud diameter range at coloring (range of flower bud diameter at coloring) is calculated (S13).

[0039] Then, the harvest time prediction system 1 acquires the current sample data (in Figure 5, the measurement date at the time of prediction and the flower bud diameter at the time of prediction) input from the user terminal U (S14), calculates the earliest expected coloring date (in Figure 5, the date d1 days after the measurement date at the time of prediction) and the latest expected coloring date (in Figure 5, the date d2 days after the measurement date at the time of prediction), and predicts the harvest time of the flowers (in Figure 5, coloring will occur d1 to d2 days after the measurement date) (S15).

[0040] The predicted flower harvest time (prediction result) is transmitted from the harvest time prediction system 1 to the user terminal U (S16) and displayed on the screen of the user terminal U (S17).

[0041] According to the first embodiment of the harvest time prediction system 1, the flower bud diameter width at coloration (the width from the minimum to maximum colored flower bud diameter) is determined based on past sample data (data on the flower bud diameter of multiple reference samples from flowers grown in the past and data on whether the flower buds are colored) acquired from flowers grown in the past, and the harvest time of the flower (the period from the earliest to latest expected coloration date) can be predicted based on current sample data (data on the flower bud diameter of prediction samples from flowers currently grown) acquired from flowers currently grown and the above-mentioned flower bud diameter width at coloration. In this way, by using the flower bud diameter width at coloration (the width of the flower bud diameter when the flower buds colored) of flowers grown in the past, the harvest time of the flower currently grown (the period from the earliest to latest expected coloration date) can be appropriately predicted.

[0042] In this embodiment, a flower growth model function (a function showing the change in flower bud diameter over the number of days elapsed) is calculated based on past sample data (including data on the dates when flower bud diameter was measured) obtained from flowers grown in the past, a range of days corresponding to the flower bud diameter range at coloration is determined, and the harvest time (earliest and latest expected coloration dates) of the flower can be predicted based on current sample data (including data on the dates when flower bud diameter of the prediction sample was measured) obtained from flowers currently being grown. In this way, by using the growth model function (a representative function derived from multiple reference samples) derived from past sample data of flowers grown in the past and the range of days (corresponding to the flower bud diameter range at coloration), the harvest time of the flower currently being grown can be appropriately predicted.

[0043] Furthermore, in this embodiment, the flower bud diameter at coloring is determined based on past sample data selected in accordance with user input (at least one of the flower production area, production year, crop type, and variety), and the harvest time of the flower currently being cultivated is predicted. Although it is particularly difficult to cultivate flowers in an open-field environment while controlling the environment, the harvest time of the flower can be appropriately predicted by selecting and using past sample data that has a cultivation environment (weather conditions, environmental conditions, etc.) similar to that of the flower currently being cultivated, based on at least one of the production area (field, region, prefecture, etc.), production year, crop type, and variety input by the user.

[0044] (Second embodiment) Next, a harvest time prediction system 1 according to a second embodiment of the present invention will be described. Here, the differences between the harvest time prediction system 1 according to the second embodiment and the first embodiment will be mainly described. Unless otherwise specified, the configuration and operation of this embodiment are the same as those of the first embodiment.

[0045] In this embodiment, the current sample data acquisition unit 8 acquires, as current sample data, data on the bud diameter of each of the multiple prediction samples, as well as data on the date on which the bud diameter of the prediction sample was measured, at least twice on different days.

[0046] Then, based on the current sample data, the prediction processing unit 9 calculates a growth model function for each of the multiple prediction samples that indicates the change in flower bud diameter size relative to the number of days elapsed since a specified reference date, estimates the expected coloring time of each flower bud of the multiple prediction samples based on the growth model function and the flower bud diameter width at coloring, and predicts the harvest time of the flowers currently being cultivated based on the expected coloring time of each flower bud of the multiple prediction samples.

[0047] In addition, the prediction processing unit 9 calculates the number of flowers currently being cultivated that are expected to be harvested each day at the harvest time based on the expected coloring time of each flower bud of multiple prediction samples and the total number of flowers currently being cultivated that are expected to be harvested at the harvest time.

[0048] Fig. 7 is a diagram illustrating harvest time prediction (prediction based on current growth trends) in this embodiment. As shown in Fig. 7, in harvest time prediction in this embodiment, data on flower bud diameters (current sample data) measured on at least two different dates are acquired for each of a plurality of prediction sample flowers (samples 1, 2, 3, etc. in Fig. 7) currently being cultivated, and a growth model function (calibration curve) is calculated for each of the plurality of prediction sample flowers (samples 1, 2, 3, etc. in Fig. 7).

[0049] On the other hand, as in the first embodiment, based on the flower bud diameter data (past sample data) measured on at least two different dates for multiple reference samples (samples 1, 2, 3, etc. in Figure 5), a growth model function (in Figure 5, the calibration curve y = ax + b) representing the degree of growth of the flower bud diameter y relative to the number of days x is calculated, and the number of days width (width of the number of days until coloring) corresponding to the flower bud diameter width at coloring (width of the flower bud diameter at coloring) is calculated.

[0050] Then, using the growth model function (calibration curve) and day range for each of the multiple prediction sample flowers (samples 1, 2, 3, etc. in Figure 7), the earliest and latest expected coloring dates (sample 1: June 29th to July 2nd in Figure 7, sample 2: June 30th to July 5th, sample 3: July 3rd to July 7, etc.) are determined for each of the multiple prediction sample flowers (samples 1, 2, 3, etc. in Figure 7), and the flower harvest time (June 29th to July 7th in Figure 7) is predicted.

[0051] In this embodiment, it is possible to predict the number of flowers to be harvested each day. FIG. 8 is a diagram illustrating a method for predicting the number of flowers to be harvested each day. In this embodiment, as shown in FIG. 8, a daily harvest rate is calculated for each of a plurality of prediction sample flowers (samples 1, 2, 3, etc. in FIG. 8). For example, since the harvest period for sample 1 is four days from June 29th to July 2nd, the harvest rate for sample 1 (1.0 in total) is divided into four equal parts to calculate "June 29th: 0.25, June 30th: 0.25, July 1st: 0.25, July 2nd: 0.25."

[0052] Similarly, for samples 2, 3, etc., the daily harvest rates are calculated and added together. The total harvest rate for samples 1, 2, 3, etc. (total of 3.0) is calculated as follows: June 29: 0.25, June 30: 0.45, ..., July 7: 0.2. Then, when this total harvest rate is normalized (total of 1.0), it becomes: June 29: 0.083, June 30: 0.15, ..., July 7: 0.066.

[0053] If a user is cultivating flowers with the expectation of harvesting a total of 3,000 stems, the expected daily harvest numbers can be calculated as "June 29th: 250 stems, June 30th: 450 stems, ..., July 7th: 200 stems" by multiplying the normalized harvest rates above (June 29th: 0.083, June 30th: 0.15, ..., July 7th: 0.066) by the number of stems to be harvested.

[0054] The operation of the harvest time prediction system 1 configured as above will be described with reference to the sequence diagram of FIG.

[0055] In the harvest time prediction system 1 of the second embodiment, as in the first embodiment, a user who grows flowers inputs data such as the flower sample numbers of multiple reference samples (e.g., samples 1, 2, and 3), the dates and measurements (first and second times) on the flower bud diameters, the dates on which the buds turned colored, and the bud diameters into the user terminal U as appropriate (see FIG. 2). The data input from the user terminal U is sent to the harvest time prediction system 1 and stored in the memory unit 4 of the harvest time prediction system 1 as past sample data (see FIG. 3).

[0056] When predicting the harvest time, data such as "production area, crop type, variety, illuminated / non-illuminated" used for predicting the harvest time, as well as data on the bud diameter and date of multiple prediction samples contained in the flowers currently being cultivated (current sample data) are input from the user terminal U (S20), and the data input from the user terminal U (user input) is sent to the harvest time prediction system 1 (S21).

[0057] The harvest time prediction system 1 obtains data on the flower bud diameter (current sample data) measured on at least two different dates for each of a plurality of prediction sample flowers currently being cultivated (samples 1, 2, 3, etc. in Figure 7) (S22), and calculates a growth model function (calibration curve) for each of the plurality of prediction sample flowers (samples 1, 2, 3, etc. in Figure 7) (S23).

[0058] Furthermore, based on the data on "production area, crop type, variety, and illuminated / non-illuminated" input from the user terminal U, past sample data with similar cultivation environments (weather conditions, environmental conditions, etc.) is read out (S24), and, as in the first embodiment, the range of days (range of days until coloring) corresponding to the flower bud diameter range at coloring (range of flower bud diameter at coloring) is calculated (S25).

[0059] Then, in the harvest time prediction system 1, the growth model function (calibration curve) and the number of days range for each of the plurality of prediction sample flowers (samples 1, 2, 3, etc. in Figure 7) are used to determine the earliest and latest expected coloring dates for each of the plurality of prediction sample flowers (samples 1, 2, 3, etc. in Figure 7) (sample 1: June 29th to July 2nd, sample 2: June 30th to July 5th, sample 3: July 3rd to July 7, etc. in Figure 7), and the harvest time for the flowers (June 29th to July 7th in Figure 7) is predicted (S26).

[0060] Furthermore, in this embodiment, the number of flowers currently being cultivated that are expected to be harvested each day at the harvest time (in Figure 8, June 29th: 250, June 30th: 450, ..., July 7th: 200) is calculated based on the expected coloring period of the flower buds of each of the multiple prediction samples (in Figure 8, Sample 1: June 29th to July 2nd, Sample 2: June 30th to July 5th, Sample 3: July 3rd to July 7th) and the total expected harvest number of flowers currently being cultivated at the harvest time (3,000 in Figure 8) (S27).

[0061] The predicted flower harvest time and the number of flowers to be harvested each day (prediction results) are transmitted from the harvest time prediction system 1 to the user terminal U (S28) and displayed on the screen of the user terminal U (S29).

[0062] The harvest time prediction system 1 of the second embodiment also provides the same effects as those of the first embodiment.

[0063] Furthermore, in this embodiment, a growth model function (a function showing the change in flower bud diameter over the number of days elapsed) for each of the multiple prediction samples is calculated based on current sample data obtained from flowers currently being cultivated (which includes data on the date the bud diameter of each of the multiple prediction samples was measured, and is obtained at least twice on different days), and the expected coloring date (earliest expected coloring date and latest expected coloring date) for each of the multiple prediction samples is estimated, making it possible to appropriately predict the harvest time of flowers currently being cultivated.

[0064] Furthermore, in this embodiment, it is possible to appropriately calculate the number of flowers currently being cultivated that are expected to be harvested each day at the harvest time (June 29th: 250, June 30th: 450, ..., July 7th: 200) from the expected coloring period of each flower bud of multiple prediction samples (e.g., Sample 1: June 29th to July 2nd, Sample 2: June 30th to July 5th, Sample 3: July 3rd to July 7th) and the total expected harvest number of flowers currently being cultivated at the harvest time (e.g., 3,000).

[0065] Although the embodiments of the present invention have been described above by way of example, the scope of the present invention is not limited to these, and can be modified and changed according to the purpose within the scope of the claims. [Industrial Applicability]

[0066] As described above, the harvest time prediction system of the present invention has the effect of being able to predict the harvest time of horticultural products within a certain range (the range from the scheduled date when the flower buds of horticultural products will color earliest to the scheduled date when they will color latest), and is useful as a system for predicting the harvest time of horticultural products such as flowers. [Explanation of symbols]

[0067] 1. Harvest time prediction system 2. Receiving section 3. Transmitter 4 Storage section 5. Control section 6 Past sample data readout section 7 Decision Processing Unit 8 Current sample data acquisition section 9 Prediction processing section N Network U User Terminal

Claims

1. A means for storing data on the flower bud diameters of a plurality of reference samples contained in a horticultural product cultivated in the past and data on whether the flower buds of the reference samples are colored as past sample data of the horticultural product; A means for determining the width of the flower bud diameter at coloring of the horticultural product from the minimum value of the flower bud diameter at coloring, which is the smallest diameter of the flower bud when coloring, to the maximum value of the flower bud diameter at coloring, which is the largest diameter of the flower bud when coloring, based on the past sample data; A means for acquiring data on the flower bud diameter of a prediction sample contained in the horticultural product currently being cultivated as current sample data of the horticultural product; A means for estimating how many days from now the flower buds of the currently cultivated horticultural product will color based on the current sample data and the diameter width of the flower buds at coloring, and predicting the period from the earliest expected coloring date, which is estimated to be the earliest when the flower buds will color, to the latest expected coloring date, which is estimated to be the latest when the flower buds will color, as the harvest time of the currently cultivated horticultural product; Equipped with The storage means includes: As the past sample data, data on the date when the flower bud diameter of the reference sample was measured is stored in association with the flower bud diameter of the reference sample and data on whether the flower bud diameter is colored or not; The determining means comprises: Based on the past sample data, a growth model function showing the change in the size of the flower bud diameter with respect to the number of days elapsed since a predetermined reference date is calculated, and a range of days corresponding to the flower bud diameter width at coloring is calculated; The acquiring means includes: As the current sample data, data on the bud diameter of the prediction sample is obtained together with data on the date on which the bud diameter of the prediction sample was measured; The means for predicting includes: A system for predicting the harvest time of horticultural products, characterized in that the earliest expected coloring date and the latest expected coloring date are determined based on the current sample data, the growth model function, and the range of days.

2. A means for storing data on the flower bud diameters of a plurality of reference samples contained in a horticultural product cultivated in the past and data on whether the flower buds of the reference samples are colored as past sample data of the horticultural product; A means for determining the width of the flower bud diameter at coloring of the horticultural product from the minimum value of the flower bud diameter at coloring, which is the smallest diameter of the flower bud when coloring, to the maximum value of the flower bud diameter at coloring, which is the largest diameter of the flower bud when coloring, based on the past sample data; A means for acquiring data on the flower bud diameter of a prediction sample contained in the horticultural product currently being cultivated as current sample data of the horticultural product; A means for estimating how many days from now the flower buds of the currently cultivated horticultural product will color based on the current sample data and the diameter width of the flower buds at coloring, and predicting the period from the earliest expected coloring date, which is estimated to be the earliest when the flower buds will color, to the latest expected coloring date, which is estimated to be the latest when the flower buds will color, as the harvest time of the currently cultivated horticultural product; Equipped with The acquiring means includes: As the current sample data, for each of the plurality of prediction samples, data on the bud diameter of the prediction sample and data on the date on which the bud diameter of the prediction sample was measured are obtained at least twice on different days; The means for predicting includes: Calculating a growth model function that indicates a change in the size of the flower bud diameter with respect to the number of days elapsed since a predetermined reference date for each of the plurality of prediction samples based on the current sample data; estimating the expected time of coloring of each of the plurality of prediction samples based on the growth model function and the flower bud diameter width at coloring; A harvest time prediction system for horticultural products that predicts the harvest time of the horticultural products currently being cultivated based on the expected coloring time of the flower buds of each of the plurality of prediction samples.

3. A method implemented in a horticultural harvest time prediction system, comprising: For horticultural products cultivated in the past, data on the flower bud diameters of a plurality of reference samples contained in the horticultural product and data on whether the flower buds of the reference samples are colored are stored as past sample data of the horticultural product; Based on the past sample data, the width from the minimum value of the flower bud diameter at coloring, which is the smallest diameter of the flower bud when coloring, to the maximum value of the flower bud diameter at coloring, which is the largest diameter of the flower bud when coloring, is determined as the flower bud diameter width at coloring of the horticultural product; For the horticultural product currently being cultivated, data on the flower bud diameter of a prediction sample contained in the horticultural product is acquired as current sample data of the horticultural product; Based on the current sample data and the flower bud diameter width at coloring, estimate how many days from now the flower buds of the currently cultivated horticultural product will color, and predict the harvest time of the currently cultivated horticultural product from the earliest expected coloring date, which is estimated to be the earliest when coloring occurs, to the latest expected coloring date, which is estimated to be the latest when coloring occurs; Including, The harvest time prediction system includes: As the past sample data, data on the date when the flower bud diameter of the reference sample was measured is stored in association with the flower bud diameter of the reference sample and data on whether the flower bud diameter is colored or not; The determining step comprises: Calculating a growth model function that indicates a change in the size of the flower bud diameter with respect to the number of days elapsed since a predetermined reference date based on the past sample data, and calculating a range of days corresponding to the flower bud diameter width at coloration, The obtaining includes: As the current sample data, data on the bud diameter of the prediction sample is obtained together with data on the date on which the bud diameter of the prediction sample is measured; The predicting step comprises: determining the earliest planned coloring date and the latest planned coloring date based on the current sample data, the growth model function, and the range of days.

4. A method implemented in a horticultural harvest time prediction system, comprising: For horticultural products cultivated in the past, data on the flower bud diameters of a plurality of reference samples contained in the horticultural product and data on whether the flower buds of the reference samples are colored are stored as past sample data of the horticultural product; Based on the past sample data, the width from the minimum value of the flower bud diameter at coloring, which is the smallest diameter of the flower bud when coloring, to the maximum value of the flower bud diameter at coloring, which is the largest diameter of the flower bud when coloring, is determined as the flower bud diameter width at coloring of the horticultural product; For the horticultural product currently being cultivated, data on the flower bud diameter of a prediction sample contained in the horticultural product is acquired as current sample data of the horticultural product; Based on the current sample data and the flower bud diameter width at coloring, estimate how many days from now the flower buds of the currently cultivated horticultural product will color, and predict the harvest time of the currently cultivated horticultural product from the earliest expected coloring date, which is estimated to be the earliest when coloring occurs, to the latest expected coloring date, which is estimated to be the latest when coloring occurs; Including, The obtaining includes: As the current sample data, for each of the plurality of prediction samples, data on the bud diameter of the prediction sample and data on the date on which the bud diameter of the prediction sample was measured are obtained at least twice or more times on different days; The predicting step comprises: Calculating a growth model function that indicates a change in the size of the flower bud diameter with respect to the number of days elapsed since a predetermined reference date for each of the plurality of prediction samples based on the current sample data; estimating the expected time of coloring of each of the plurality of prediction samples based on the growth model function and the flower bud diameter width at coloring; The method includes predicting the harvest time of the currently cultivated horticultural product based on the expected time of coloring of the flower buds of each of the plurality of prediction samples.

5. A program executed in a horticultural harvest time prediction system, The harvest time prediction system includes: For a horticultural product cultivated in the past, data on the flower bud diameters of a plurality of reference samples contained in the horticultural product and data on whether the flower buds of the reference samples are colored are stored as past sample data for the horticultural product; The program is configured to the harvest time prediction system, A process of determining the width of the flower bud diameter at coloring of the horticultural product from the minimum flower bud diameter at coloring, which was the smallest diameter of the flower bud when coloring, to the maximum flower bud diameter at coloring, which was the largest diameter of the flower bud when coloring, based on the past sample data; A process of acquiring data on the flower bud diameter of a prediction sample contained in the horticultural product currently being cultivated as current sample data of the horticultural product; A process of estimating how many days from now it will take for the flower buds of the currently cultivated horticultural product to color based on the current sample data and the flower bud diameter width at coloring, and predicting the harvest time of the currently cultivated horticultural product from the earliest expected coloring date, which is estimated to be the earliest when coloring occurs, to the latest expected coloring date, which is estimated to be the latest when coloring occurs; Execute The harvest time prediction system includes: As the past sample data, data on the date when the flower bud diameter of the reference sample was measured is stored in association with the flower bud diameter of the reference sample and data on whether the flower bud diameter is colored or not; The determining process includes: Calculating a growth model function showing a change in the size of the flower bud diameter with respect to the number of days elapsed since a predetermined reference date based on the past sample data, and calculating a range of days corresponding to the flower bud diameter width at coloration, The acquiring process includes: The method includes a process of acquiring, as the current sample data, data on the bud diameter of the prediction sample, as well as data on the date on which the bud diameter of the prediction sample was measured; The prediction process includes: A program including a process for determining the earliest planned coloring date and the latest planned coloring date based on the current sample data, the growth model function, and the range of days.

6. A program executed in a horticultural harvest time prediction system, The harvest time prediction system includes: For a horticultural product cultivated in the past, data on the flower bud diameters of a plurality of reference samples contained in the horticultural product and data on whether the flower buds of the reference samples are colored are stored as past sample data for the horticultural product; The program is configured to the harvest time prediction system, A process of determining the width of the flower bud diameter at coloring of the horticultural product from the minimum flower bud diameter at coloring, which was the smallest diameter of the flower bud when coloring, to the maximum flower bud diameter at coloring, which was the largest diameter of the flower bud when coloring, based on the past sample data; A process of acquiring data on the flower bud diameter of a prediction sample contained in the horticultural product currently being cultivated as current sample data of the horticultural product; A process of estimating how many days from now it will take for the flower buds of the currently cultivated horticultural product to color based on the current sample data and the flower bud diameter width at coloring, and predicting the harvest time of the currently cultivated horticultural product from the earliest expected coloring date, which is estimated to be the earliest when coloring occurs, to the latest expected coloring date, which is estimated to be the latest when coloring occurs; Execute The acquiring process includes: The method includes a process of acquiring, as the current sample data, data on the bud diameter of each of the plurality of prediction samples, as well as data on the date on which the bud diameter of each of the prediction samples was measured, at least twice on different days; The prediction process includes: Calculating a growth model function that indicates a change in the size of the flower bud diameter with respect to the number of days elapsed since a predetermined reference date for each of the plurality of prediction samples based on the current sample data; estimating the expected time of coloring of each of the plurality of prediction samples based on the growth model function and the flower bud diameter width at coloring; A program including a process for predicting the harvest time of the horticultural product currently being cultivated based on the expected time of coloring of the flower buds of each of the plurality of prediction samples.

Images