Tea leaf yield prediction device, tea leaf yield prediction program, and tea leaf yield prediction method

Abstract

To easily predict a yield of tea leaves that reflects the cultivation for each tea field.SOLUTION: A tea leaf yield prediction device for predicting a yield of tea leaves obtained by plucking and harvesting tea leaves from tea trees in trimmed tea fields includes: a storage section 20 for storing a prediction model (formula (1)) to be used for prediction; an acquisition section 10 for acquiring sample measurement information related to the tea trees per unit measurement space when the tea field is trimmed; a calculation section 30 for calculating a predicted value of a yield, on the basis of the sample measurement information and the prediction model; and an output section 40 for outputting the predicted value. The prediction model is determined according to a calibration curve indicating a relationship between: a cross-sectional area of a branch of a tea tree when a tea field is trimmed; and an actual value of a yield of tea leaves when the tea field was plucked. The sample measurement information contains branch information related to specific branches that hold a branch cross-sectional area of a predetermined area or more, among a plurality of branches in the unit measurement space.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a tea leaf yield prediction device, a tea leaf yield prediction program, and a tea leaf yield prediction method.

Background Art

[0002] In the cultivation of tea plants, when the tea plants awaken from winter dormancy, the budding of new shoots begins in spring. "Budding" refers to the state in which the overwintering buds sprout due to the rise in temperature, that is, when the length of the new shoots becomes twice the length of the bract leaves. The time when the buds in this budding state reach 70% of the whole is called the budding period.

[0003] Around late April to early May after a lapse of a predetermined period (about one month) from budding, the tea leaves are harvested. The tea leaves harvested at this time are so-called first flush tea. About 50 days after the harvest of the first flush tea, so-called second flush tea is harvested. Since the temperature rises after the harvest of the second flush tea, about 30 to 40 days after the harvest of the second flush tea, so-called third flush tea is harvested.

[0004] After the completion of the tea leaf harvest, as the temperature drops in autumn, the growth of the tea plants slows down. Around October, the pruning of the tea plants in the entire tea garden is carried out for the next year's tea leaf harvest. Among the prunings, the pruning carried out in autumn for the next year's tea leaf harvest is "autumn pruning", and "autumn pruning" is an operation of cutting in a part of the autumn shoots that have grown after the final tea harvest. Autumn pruning is carried out for the purpose of adjusting the harvesting surface (cutting surface) of the tea plants so that old leaves do not mix when harvesting the first flush tea of the next year, and for the purpose of adjusting the uniformity of the buds and the branches that sprout.

[0005] Tea farmers control the number of buds that can be harvested in the next year by autumn pruning, and adjust (prepare) the yield and quality of the tea leaves (first flush tea) of the next year.

[0006] Here, a prediction model has been proposed to predict the yield of tea leaves (first flush tea) harvested in the next spring from the state (tree vigor) of the tea plants at the time of autumn pruning (see, for example, Non-Patent Document 1).

Prior Art Documents

Patent Document

[0007]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0008] However, the method of adjusting the yield of tea leaves (first flush tea) depends on the intuition and experience of skilled tea farmers.

[0009] Conventional prediction models are determined based on general plant growth models, and the methods based on the intuition and experience of different tea farmers in each region (so-called the skills of tea farmers) are not considered (reflected) at all. Therefore, while conventional prediction models have many coefficients for calculating the predicted yield and the calculation is complex, they cannot predict the yield that reflects the method for each tea garden.

[0010] An object of the present invention is to provide a tea leaf yield prediction device, a tea leaf yield prediction program, and a tea leaf yield prediction method that can easily predict the yield of tea leaves reflecting the method for each tea garden.

Means for Solving the Problems

[0011] The tea leaf yield prediction device according to the present invention is a tea leaf yield prediction device that predicts the yield of tea leaves harvested by picking tea leaves from tea trees in a tea garden to be pruned, and includes a storage unit that stores a prediction model used for prediction, an acquisition unit that acquires sample measurement information regarding the tea trees per unit measurement space at the time when the tea garden is pruned, a calculation unit that calculates a predicted value of the yield based on the sample measurement information and the prediction model, and an output unit that outputs the predicted value. The prediction model is determined based on a calibration curve showing the relationship between the cross-sectional area of the branches of the tea trees at the time when the tea garden is pruned and the measured value of the yield of the tea leaves at the time of picking in the tea garden. The sample measurement information includes branch information regarding specific branches having a branch cross-sectional area of a predetermined area or more among a plurality of branches in the unit measurement space.

Effect of the Invention

[0012] According to the present invention, it is possible to easily predict the yield of tea leaves reflecting the preparation for each tea garden.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0014] Hereinafter, embodiments of the tea leaf yield prediction device, the tea leaf yield prediction program, and the tea leaf yield prediction method according to the present invention will be described with reference to the drawings.

[0015] Here, in the embodiments described below, a case where a prediction model derived based on a tea garden in which the area where the tea garden for predicting the yield of tea leaves exists, the budding degree of tea leaves (new shoots), the light shielding rate of the material covering the tea tree, and the covering period of covering the tea leaves are specified will be described as an example. In the present embodiment, the area of the tea garden for predicting the yield of tea leaves applied to derive the prediction model is "Yabuko", the budding degree of tea leaves (new shoots) is "50%", the light shielding rate of the material covering the tea tree is "80%", and the covering period of covering the tea leaves is "14 days". That is, the tea garden in the area of "Yabuko" is an example of the specific tea garden in the present invention. Also, the budding degree of "50%" is an example of the specific budding degree in the present invention. Furthermore, the light shielding rate of "80%" and the covering period of "14 days" are examples of the specific covering state in the present invention. Note that the tea farmers in the specific tea garden "Yabuko" pick from about 20 mm above the pruning surface of autumn pruning in the following spring. The prediction model in the present embodiment reflects the workmanship of these Yabuko tea farmers.

[0016] A "tea garden" is a field in which a plurality of tea trees (plants) are ridged or naturally ridged at a predetermined interval (inter-plant spacing).

[0017] The "opening rate" is the ratio of the number of fully opened leaves to the total number of new buds within a certain area (for example, within one tea garden). "Fully opened" refers to the state where all the leaves that should open on a single tea leaf (new bud) have fully opened and the leaf expansion has temporarily stopped.

[0018] ● Tea leaf yield prediction device FIG. 1 is a functional block diagram showing an embodiment of a tea leaf yield prediction device according to the present invention.

[0019] The tea leaf yield prediction device (hereinafter referred to as "this device") 1 according to the present invention predicts the yield of tea leaves harvested by picking tea leaves in the following spring from tea trees in a tea garden (specific tea garden) to be pruned (autumn pruning). This device 1 is composed of, for example, a personal computer, a portable computer terminal such as a mobile phone (for example, a smartphone), a tablet PC (Personal Computer), or a PDA (Personal Digital Assistant). This device 1 includes an acquisition unit 10, a storage unit 20, a calculation unit 30, and an output unit 40.

[0020] In this device 1, a tea leaf yield prediction program (hereinafter referred to as "this program") according to the present invention operates, and this program cooperates with the hardware resources of this device 1 to implement a tea leaf yield prediction method (hereinafter referred to as "this method") according to the present invention described later. That is, this program causes a computer (not shown) to function as this device 1.

[0021] The acquisition unit 10 acquires sample measurement information regarding tea trees per unit measurement space at the time when the specific tea garden is pruned. The sample measurement information is input by the user of this device 1 using the input means K. The input means K is, for example, a keyboard connected to this device 1 or a touch panel provided in this device 1. The sample measurement information acquired by the acquisition unit 10 is stored in the storage unit 20.

[0022] The "unit measurement space" is a predetermined space from the surface of the tree crown in the tea garden (for example, the space within a frame of 200 mm in length × 200 mm in width × 30 mm in height). In other words, the acquisition unit 10 acquires information obtained by measuring a plurality of branches existing inside a frame of 200 mm in length × 200 mm in width × 30 mm, which is arranged on the surface of the tree crown at the time when the specific tea garden is pruned.

[0023] Note that in the present embodiment, the number of frames for defining the unit measurement space for measuring a specific branch is arbitrary. That is, for example, one or more frames are arranged per row constituting the ridge of the tea garden.

[0024] The "sample measurement information" is information (branch information) regarding branches having a branch cross-sectional area of a predetermined area (for example, 1.43 mm 2 ) or more among the plurality of branches in the unit measurement space. The "branch having a branch cross-sectional area of a predetermined area (1.43 mm 2 ) or more" is an example of a specific branch in the present invention. In other words, the sample measurement information is branch information regarding a specific branch in the unit measurement space.

[0025] The "branch information" includes information regarding the number of specific branches in the unit space. The "information regarding the number of specific branches" includes information regarding cut branches and non-cut branches. A "cut branch" is a branch that has been pruned (cut) by pruning. A "non-cut branch" is a branch that has not been pruned (cut) by pruning, that is, the apical bud.

[0026] Here, the specific branch in the present invention may be a branch having a branch cross-sectional area of a predetermined area or more and a branch having a branch cross-sectional area of a predetermined area or less. That is, if the branch cross-sectional area at the time of autumn pruning is not a predetermined area or more, the buds emerging from thin branches are short and do not reach the picking surface for harvesting, and thus are not the targets for picking in the following spring. On the other hand, if the branch cross-sectional area at the time of autumn pruning exceeds the predetermined area, it will be pruned by pruning immediately before picking in the following spring, or the internodes will become wider, and thus it will not be included in the yield of picking in the following spring.

[0027] In the present embodiment, the "number of specific branches" in a specific tea garden is plural (for example, 11 or more). Regarding "cut branches pruned by pruning" and "non-cut branches not pruned by pruning", they will be described later.

[0028] The storage unit 20 stores information necessary for the present apparatus 1 to execute information processing described later. The storage unit 20 is composed of, for example, a recording device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory. The storage unit 20 stores sample measurement information, a prediction model, a regional correction value, an opening degree correction value, a covering correction value, etc. The prediction model, the regional correction value, the opening degree correction value, the covering correction value, and the operation of the storage unit 20 will be described later.

[0029] The calculation unit 30 reads out the sample measurement information and the prediction model from the storage unit 20, and calculates the predicted yield (predicted value of the yield) of the specific tea garden.

[0030] The output unit 40 outputs the predicted yield of the specific tea garden calculated by the calculation unit 30 to the display D. The operation of the output unit 40 will be described later.

[0031] The display D displays the output result of the output unit 40. The display D is, for example, a display unit (not shown) connected to the present apparatus 1.

[0032] Note that the display may be a display unit provided in an external portable computer terminal different from the present apparatus, such as a mobile phone (for example, a smartphone), a tablet PC, or a PDA.

[0033] Also, the output mode of the predicted yield by the output unit may not be limited to display on the display or an external display unit. That is, for example, the output unit may transmit the predicted yield to a printer or the like connected to the present apparatus and cause the printer to print it.

[0034] ● Prediction model (1) Here, the prediction model stored in the storage unit 20 will be described. The prediction model can be expressed by the following formula (1).

[0035] Formula (1): Y 50% = 72.5×(CA au,i 1.38 ) ave + 275

[0036] As described above, formula (1) is the formula for predicting the yield when the budding degree of tea leaves (new shoots) in a specific tea garden is "50%".

[0037] In formula (1), "Y 50% " indicates that it is a formula for predicting the yield when the budding degree is "50%".

[0038] Also, in formula (1), "(CA au,i 1.38 ) ave " is the average value of the value obtained by raising the cross-sectional area of each specific branch in the unit measurement space to the 1.38th power.

[0039] Formula (1) is a formula derived by the present inventor based on the calibration curve obtained from the measured values (actual yields) of past yield data.

[0040] Figure 2 is the calibration curve used for deriving the prediction model (formula (1)) stored in the storage unit 20.

[0041] This figure shows that the horizontal (X) axis is the average of the value obtained by raising the cross-sectional area of the specific branch to the 1.38th power, and the vertical (Y) axis is the actual yield. From this figure, it is specified that the slope of formula (1) is approximately 72.5 and the Y-intercept is approximately 275.

[0042] ● Tea leaf yield prediction method (1) Next, this method will be described. This method is executed by the present apparatus 1. The following description refers to both Figure 1 in addition to the drawings described later.

[0043] First, the acquisition unit 10 acquires sample measurement information input by the user using the input means K. The sample measurement information acquired by the acquisition unit 10 is stored in the storage unit 20.

[0044] FIG. 3 is a schematic diagram showing an example of an input screen of information input to the present apparatus 1. This figure shows a (fixed) state in which "region", "opening degree", "light shielding rate", and "coating period" are pre-input on the input screen. Also, this figure is configured to enable input of "number of cut branches", "number of non-cut branches", and "branch cross-sectional area", and the user has input "X" for the "number of cut branches", "Y" for the "number of non-cut branches", and "Z" mm 2 for the "branch cross-sectional area".

[0045] Note that, for convenience of explanation, although illustration is omitted, the "branch cross-sectional area" input on the input screen is the total cross-sectional area of all specific branches in the unit measurement space. The display "Z" of the "branch cross-sectional area" shown in FIG. 3 indicates that the average value of the cross-sectional area of specific branches in the unit measurement space is "Z". The average value "Z" is calculated by the present apparatus 1.

[0046] Next, the calculation unit 30 reads out the sample measurement information and the prediction model from the storage unit 20 and calculates the predicted yield. The predicted yield calculated by the calculation unit 30 is stored in the storage unit 20.

[0047] Next, the output unit 40 reads out the calculation result of the calculation unit 30, that is, the predicted yield, from the storage unit 20 and outputs it to the display D.

[0048] FIG. 4 is a schematic diagram showing an example of an output screen on which the predicted yield of a specific tea garden calculated by the calculation unit 30 is displayed. This figure shows that when the "region" is Yatsushiro, the "opening degree" is 50%, the "light shielding rate" is 80%, the "coating period" is 14 days, the "number of cut branches" is "X", the "number of non-cut branches" is "Y", and the "branch cross-sectional area" is "Z" mm 2 a yield of 500 kg / 10a is predicted.

[0049] ● Prediction model (2) The prediction model of the foregoing formula (1) is determined under certain preconditions. That is, the prediction model of formula (1) predicts the yield under the same conditions of "region", "opening degree", "light shielding rate", and "coating period". In other words, the prediction model of formula (1) is a reference prediction model (hereinafter referred to as the "reference prediction model").

[0050] The present apparatus 1 can also predict the yield under conditions different from the preconditions used when determining the reference prediction model, not limited to this. When predicting the yield under conditions different from the preconditions of the reference prediction model, for example, three correction values are used. The three correction values are the region correction value CC, the opening degree correction value CRB, and the coating correction value DC. As described above, the region correction value CC, the opening degree correction value CRB, and the coating correction value DC are stored in the storage unit 20.

[0051] Here, the prediction model using the three correction values (hereinafter referred to as the "corrected prediction model") can be represented by the following formula (2). That is, the prediction model in the present invention includes the reference prediction model (formula (1)) and the corrected prediction model (formula (2)).

[0052] Formula (2): Y RBS,covered =CRB×DC×CC×(72.5×(CA au,i 1.38 ) ave +275)

[0053] In formula (2), "RBS" in "Y RBS,covered " indicates the opening degree, and "covered" indicates that it is a covered tea garden.

[0054] The "region correction value CC" is a value multiplied by the reference prediction model when predicting the yield in a tea garden in a region different from the specific tea garden based on the reference prediction model. The "tea garden in a region different from the specific tea garden" is an example of the predicted tea garden in the present invention. That is, the region correction value CC is a value used to calculate the predicted yield of the predicted tea garden in a region different from the specific tea garden.

[0055] The regional correction value CC is calculated based on the specific number of new shoots and the predicted number of new shoots. The regional correction value CC is "CC = a' / 43.9", and "a' = m1N au,1 + m2N au,2 ".

[0056] "a'" is the number of new shoots at the time of spring harvesting in the predicted tea garden (hereinafter referred to as "predicted number of new shoots"). That is, the "predicted number of new shoots" is the number of new shoots with specific branches as the mother branches per unit measurement space at the time of harvesting in the predicted tea garden.

[0057] "43.9" is the number of new shoots at the time of spring harvesting in the specific tea garden (hereinafter referred to as "specific number of new shoots"). That is, the "specific number of new shoots" is the number of new shoots with specific branches as the mother branches per unit measurement space at the time of harvesting in the specific tea garden. The value of the specific number of new shoots is a value obtained from the measured values (actual yields) in the past specific tea gardens.

[0058] Figure 5 is a graph used for deriving the regional correction value CC. This figure shows that the vertical axis is the ratio of the number of cut branches (autumn) to the number of new shoots (spring), and the horizontal axis is the number of cut branches (autumn) after autumn pruning. Based on the slope of the graph in this figure, the specific number of new shoots is determined to be approximately 43.9.

[0059] "m1" used for deriving "a'" is the number of cut branches of new shoots at the time of spring harvesting in the predicted tea garden (hereinafter referred to as "predicted number of cut branches"). That is, the "predicted number of cut branches" is the number of new shoots with cut branches as the mother branches at the time of harvesting in the predicted tea garden. "m2" is the number of non-cut branches of new shoots at the time of spring harvesting in the predicted tea garden (hereinafter referred to as "predicted number of non-cut branches"). That is, the "predicted number of non-cut branches" is the number of new shoots with non-cut branches as the mother branches at the time of harvesting in the predicted tea garden.

[0060] Also, "N au,1 " used for deriving "a'" is the number of cut branches at the time of pruning (autumn pruning) in the predicted tea garden. "N au,2 " is the number of non-cut branches at the time of pruning (autumn pruning) in the predicted tea garden.

[0061] The "opening degree correction value CRB" is a value that is multiplied by the reference prediction model when predicting the yield at an opening degree different from the specific opening degree in a tea garden (specific tea garden, predicted tea garden) based on the reference prediction model. The "opening degree different from the specific opening degree" is an example of the predicted opening degree in the present invention. That is, the opening degree correction value CRB is a value used to calculate the predicted value of the yield at a predicted opening degree different from the specific opening degree.

[0062] The opening degree correction value CRB is determined based on the opening degree of tea leaves at the time of spring picking in the tea garden (specific tea garden, predicted tea garden) and the actual measured value of the yield in the tea garden (specific tea garden, predicted tea garden).

[0063] The opening degree correction value CRB is, for example, "CRB = (0.005 × opening degree + 0.1) / 0.35". That is, for example, when the opening degree is "95%", the opening degree correction value CRB is "1.64".

[0064] The "coating correction value DC" is a value that is multiplied by the reference prediction model when predicting the yield in a tea garden (specific tea garden, predicted tea garden) in a coating state different from the specific coating state based on the reference prediction model. The "coating state different from the specific coating state" is an example of the predicted coating state in the present invention. That is, the coating correction value DC is a value used to calculate the predicted value of the yield in a predicted coating state different from the specific coating state.

[0065] The "coating correction value DC" is determined based on the coating state of tea trees in the tea garden (specific tea garden, predicted tea garden) and the actual measured value of the yield in the tea garden (specific tea garden, predicted tea garden).

[0066] Figure 6 is a graph showing the influence of covered cultivation on the yield. (a) is a graph showing the relationship between the predicted yield and the actual yield when the light shading rate is constant, and (b) is a graph showing the relationship between the predicted yield and the actual yield when the opening degree is constant.

[0067] Figure (a) shows the relationship between the predicted yield and the actual yield at the opening degree of "100%" and the opening degree of "50%" when the light shielding rate is "0%". That is, Figure (a) is a graph when the tea garden is cultivated by open-field cultivation.

[0068] Figure (b) shows the relationship between the predicted yield and the actual yield at the light shielding rate of "0%" and the light shielding rate of "80%" when the opening degree is "95 - 100%". That is, Figure (b) is a graph when the tea garden is cultivated by covered cultivation.

[0069] Figure (a) shows that the yield can be predicted only by correcting the slope even if the opening degree is different. That is, according to the present invention, the yield at the opening degree desired by tea farmers can be predicted. Figure (b) shows that the yield can be predicted only by correcting the slope even if the light shielding rate is different. That is, according to the present invention, the yield in covered cultivation performed to improve the quality of tea can be predicted. Thus, according to the present invention, the yield can be predicted in consideration of the opening degree and the light shielding rate related to quality and profit.

[0070] ● Tea leaf yield prediction method (2) Next, this method using the correction prediction model will be described. This method is executed by the present apparatus 1. The following description refers to FIG. 1 in addition to the drawings described later.

[0071] Also, the following description will be given by taking the case where the region, that is, the predicted tea garden is "Region A" as an example.

[0072] First, the acquisition unit 10 acquires the sample measurement information input by the user using the input means K, the predicted tea garden, the branch information of the predicted tea garden, the predicted opening degree, and the predicted covering state. The sample measurement information and the like acquired by the acquisition unit 10 are stored in the storage unit 20.

[0073] FIG. 7 is a schematic diagram showing another example of the input screen of the information input to the present apparatus 1. The figure is configured such that inputs of "region", "opening degree", "light shielding rate", "coating period", "number of cut branches", "number of non-cut branches", and "branch cross-sectional area" can be made on the input screen. The user has input "region A" for "region", "50" % for "opening degree", "80" % for "light shielding rate", "14" days for "coating period", "X" branches for "number of cut branches", "Y" branches for "number of non-cut branches", and "Z" mm for "branch cross-sectional area". 2 It shows that the above inputs have been made.

[0074] Also, in the figure, "region A" is selected from among Yatsushiro, region A, region B, and region C (regions A to C are all regions outside Fukuoka Prefecture) for "region", and the same values as those shown on the input screen in Figure 3 are input for "opening degree", "light shielding rate", "coating period", "number of cut branches", "number of non-cut branches", and "branch cross-sectional area". The present apparatus 1 calculates the predicted yield using a correction prediction model composed of the correction value of "region correction value CC" corresponding to "region A" in the correction prediction model, the correction value of "opening degree correction value CRB" of "1" (i.e., the coefficient is "1"), and the correction value of "coating correction value DC" of "1" (i.e., the coefficient is "1"). That is, Equation (2) is expressed as the following Equation (2').

[0075] Equation (2'): Y 50% = CC × (72.5 × (CA au,i 1.38 )) ave + 275)

[0076] That is, Equation (2') is the equation obtained by multiplying Equation (1) by the region correction value CC.

[0077] Next, the calculation unit 30 reads out the sample measurement information, the branch information of the predicted tea garden, the prediction model (correction prediction model), and the region correction value CC from the storage unit 20, and calculates the predicted yield of the predicted tea garden. The predicted yield of the predicted tea garden calculated by the calculation unit 30 is stored in the storage unit 20.

[0078] Next, the output unit 40 reads out the calculation result of the calculation unit 30, that is, the predicted yield of the predicted tea garden, from the storage unit 20, and outputs it to the display D.

[0079] FIG. 8 is a schematic diagram showing an example of an output screen on which the predicted yield of the predicted tea garden calculated by the calculation unit 30 is displayed. In this figure, "region" is region A, "opening degree" is 50%, "light shading rate" is 80%, "coating period" is 14 days, "number of cut branches" is "X", "number of non-cut branches" is "Y", and "branch cross-sectional area" is "Z" mm 2 It shows that a yield of 400 kg / 10a is predicted when.

[0080] FIG. 9 is a graph showing the relationship between the predicted yield and the actual yield in tea gardens in each region calculated based on the prediction models (reference prediction model, correction prediction model) stored in the storage unit 20. This figure shows that the correction prediction model in the present invention can accurately predict the yield even when the region of the tea garden changes.

[0081] Table 1 shows the relationship between the predicted yield before and after correction of the regional correction value CC and the measured value. By predicting the yield using the regional correction value CC, a yield close to the measured value was predicted.

[0082]

Table 1

[0083] ●Summary● According to the embodiment described above, the present apparatus 1 calculates a predicted yield based on the prediction model of formula (1) and the sample measurement information. The prediction model of formula (1) is determined based on a calibration curve showing the relationship between the cross-sectional area of the branches of the tea tree at the time when the tea garden is pruned (the time of autumn pruning) and the actual value of the yield of tea leaves at the time of spring harvesting in the tea garden. The sample measurement information includes branch information regarding specific branches having a branch cross-sectional area of a predetermined area (1.43 mm 2 ) or more among a plurality of branches in a unit measurement space (for example, a space within a frame of 200 mm in length × 200 mm in width × 30 mm in height). By using such a prediction model and sample measurement information, it is possible to easily predict the yield of tea leaves that can be harvested in the following spring at the time of autumn pruning.

[0084] As described above, the prediction model of formula (1) in this embodiment is a model for predicting the yield when picking about 20 mm above the pruning surface of the autumn pruning of tea farmers in the specific tea garden "Yabuki", that is, in the following spring. Therefore, when predicting the yield of a tea garden that is pruned about 10 mm above the pruning surface of the autumn pruning and picked in the following spring, for example, which is a pruning method different from the one reflected in the prediction model of formula (1), it is necessary to correct the yield calculated by the model of formula (1).

[0085] According to the actual measurement by the present inventors, the correction value CH of the increase in yield with respect to the prediction model of formula (1) due to the difference in picking height, taking the case of picking 2 cm above the autumn pruning surface as a reference, and setting the increase or decrease in height as H, for example, is "CH = -0.086 × increase or decrease H (cm) with respect to the reference + 1.00". That is, when picking 1 cm above the autumn pruning surface, it is 1 cm lower than the reference picking surface, that is, the increase or decrease in height is -1, so "CH = -0.086 × (-1) + 1.00 = 1.086", and it can be predicted that the yield will increase by about 8.6% compared to the yield prediction calculated by the prediction model of formula (1).

[0086] Also, according to the embodiment described above, the present apparatus 1 can predict the yield of a predicted tea garden in a region different from the specific tea garden by multiplying the regional correction value CC by formula (1) (by using the prediction model of formula (2)). At this time, the calculation unit of the present apparatus calculates the predicted yield (predicted value of the yield) of the predicted tea garden based on the branch information of the predicted tea garden, the prediction model of formula (1), and the regional correction value.

[0087] Furthermore, according to the embodiment described above, the present apparatus 1 can predict the yield at an arbitrary budding degree (predicted budding degree) different from the specific budding degree by multiplying the budding degree correction value CRB by formula (1) (by using the prediction model of formula (2)). At this time, the calculation unit of the present apparatus calculates the predicted yield (predicted value of the yield) at the predicted budding degree based on the prediction model of formula (1), the predicted budding degree, and the budding degree correction value.

[0088] Furthermore, according to the embodiments described above, the present apparatus 1 can predict the yield in an arbitrary coating state (predicted coating state) different from the specific coating state by multiplying the coating correction value DC by the formula (1) (by using the prediction model of the formula (2)). At this time, the calculation unit of the present apparatus calculates the predicted yield (predicted value of the yield) in the predicted coating state based on the prediction model of the formula (1), the predicted coating state, and the coating correction value.

[0089] Hereinafter, the features of the tea leaf yield prediction apparatus, the tea leaf yield prediction program, and the tea leaf yield prediction method according to the present invention described so far will be collectively described. (Feature 1) A tea leaf yield prediction apparatus for predicting the yield of tea leaves harvested by picking tea leaves from tea trees in a tea garden to be pruned, a storage unit (storage unit 20) that stores a prediction model (prediction model of formula (1)) used for the prediction, an acquisition unit (acquisition unit 10) that acquires sample measurement information regarding the tea trees per unit measurement space at the time when the tea garden is pruned, a calculation unit (calculation unit 30) that calculates a predicted value of the yield based on the sample measurement information and the prediction model, an output unit (output unit 40) that outputs the predicted value, comprising: the prediction model is the cross-sectional area of the branches of the tea trees at the time when the tea garden is pruned, the measured value of the yield of the tea leaves at the time of picking in the tea garden, determined based on a calibration curve showing the relationship between the sample measurement information is branch information regarding specific branches having a branch cross-sectional area of a predetermined area or more among a plurality of the branches in the unit measurement space, including A tea leaf yield prediction apparatus (the present apparatus 1) characterized by the above. (Feature 2) The branch information is information regarding specific branches of a predetermined number or more among the plurality of the branches in the unit measurement space, including The tea leaf yield prediction device according to Feature 1. (Feature 3) The branch information is information regarding the number of the specific branches within the unit measurement space, including The tea leaf yield prediction device according to Feature 1. (Feature 4) The specific branch is the cut branches pruned by the pruning, and the non-cut branches not pruned by the pruning, including The branch information is the number of the cut branches, and the number of the non-cut branches, including The tea leaf yield prediction device according to Feature 3. (Feature 5) The yield of the tea leaves is predicted for a specific tea garden, the prediction model is determined based on the measured values of the yield in the specific tea garden and stored in the storage unit, The storage unit is a regional correction value (regional correction value CC) used for calculating the predicted value of the yield of a predicted tea garden in a region different from the specific tea garden, storing The regional correction value is the number of specific new shoots with the specific branch as the mother branch per unit measurement space at the time of picking in the specific tea garden, and the predicted number of new shoots with the specific branch as the mother branch per unit measurement space at the time of picking in the predicted tea garden, a value calculated based on The acquisition unit is acquiring the branch information of the predicted tea garden, and The calculation unit is calculating the predicted value of the yield of the predicted tea garden based on the branch information of the predicted tea garden, the prediction model, the regional correction value, and The tea leaf yield prediction device according to Feature 1. (Feature 6) The specific branch includes the cut branches pruned by the pruning, and the non-cut branches not pruned by the pruning, and the predicted number of new shoots is the number of the cut branches at the time of pruning of the predicted tea garden, the number of the non-cut branches at the time of pruning of the predicted tea garden, the predicted number of cut branches of the new shoots with the cut branches as the mother branches at the time of harvesting of the predicted tea garden, the predicted number of non-cut branches of the new shoots with the non-cut branches as the mother branches at the time of harvesting of the predicted tea garden, and is calculated based on the tea leaf yield prediction device according to Feature 5. (Feature 7) The prediction model is determined based on the measured value of the yield when the degree of opening of the tea garden at the time of harvesting is a specific degree of opening, The storage unit stores the degree of opening correction value (degree of opening correction value CRB) used for calculating the predicted value of the yield at a predicted degree of opening different from the specific degree of opening, and the degree of opening correction value is determined based on the degree of opening of the tea leaves at the time of harvesting of the tea garden, the measured value of the yield of the tea garden, and The acquisition unit acquires the predicted degree of opening, and The calculation unit calculates the predicted value of the yield at the predicted degree of opening based on the prediction model, the predicted degree of opening, and the degree of opening correction value. The tea leaf yield prediction device according to Feature 1. (Feature 8) The prediction model is determined based on the measured value of the yield in a specific covering state of the tea garden, The storage unit A covering correction value (covering correction value DC) used for calculating a predicted value of the yield in a predicted covering state different from the specific covering state is stored The covering correction value is determined based on the covering state of the tea plants in the tea garden and the measured value of the yield of the tea garden and is The acquisition unit acquires the predicted covering state and The calculation unit calculates a predicted value of the yield in the predicted covering state based on the prediction model, the predicted covering state, the covering correction value, and The tea leaf yield prediction device according to Feature 1 (Feature 9) A tea leaf yield prediction program characterized by causing a computer to function as the tea leaf yield prediction device according to Feature 1 (Feature 10) A tea leaf yield prediction method executed by a tea leaf yield prediction device that predicts the yield of tea leaves harvested by picking tea leaves from tea plants in a tea garden to be pruned, wherein the tea leaf yield prediction device includes a storage unit that stores a prediction model used for the prediction and the tea leaf yield prediction method includes an acquisition step of acquiring sample measurement information regarding the tea plants per unit measurement space at the time when the tea garden is pruned, a calculation step of calculating a predicted value of the yield based on the sample measurement information and the prediction model, an output step of outputting the predicted value, and the prediction model is determined based on a calibration curve showing the relationship between the cross-sectional area of the branches of the tea plants at the time when the tea garden is pruned and the measured value of the yield of the tea leaves at the time of picking of the tea garden and​ The sample measurement information is branch information regarding specific branches having a branch cross-sectional area of a predetermined area or more among the plurality of the branches in the unit measurement space, including A method for predicting tea leaf yield, characterized by the above.

Explanation of symbols

[0090] 1 Tea leaf yield prediction device (this device) 10 Acquisition unit 20 Storage unit 30 Calculation unit 40 Output unit CC Regional correction value CRB Opening degree correction value DC Coating correction value K Input means D Display

Claims

1. A tea leaf yield prediction device for predicting the yield of tea leaves harvested by picking tea leaves in the following spring from tea trees in a specific tea garden to be pruned, comprising: a storage unit that stores a prediction model used for the prediction and a regional correction value used for calculating a predicted value of the yield of a predicted tea garden in a region different from the specific tea garden; an acquisition unit that acquires sample measurement information regarding the tea trees per unit measurement space at the time when the specific tea garden is pruned; a calculation unit that calculates a predicted value of the yield based on the regional correction value calculated based on the sample measurement information and the prediction model; an output unit that outputs the predicted value; characterized in that the prediction model is the average value of the values obtained by raising to the 1.38th power the cross-sectional area of each specific branch having a cross-sectional area of a predetermined area or more among a plurality of branches within the unit measurement space of the tea trees at the time when the specific tea garden is pruned, the actually measured value of the yield of the tea leaves at the time of picking of the specific tea garden, and is determined based on a calibration curve showing the relationship therebetween; the sample measurement information includes branch information regarding the specific branch within the unit measurement space, and the branch information is the specific number of new shoots with the specific branch as the mother branch per unit measurement space at the time of picking of the specific tea garden, and the predicted number of new shoots with the specific branch as the mother branch per unit measurement space at the time of picking of the predicted tea garden, and the regional correction value is a value calculated based on the specific number of new shoots and the predicted number of new shoots; the acquisition unit acquires the branch information of the predicted tea garden; and the calculation unit calculates by multiplying the actually measured value of the yield of the tea leaves corresponding to the average value of the values obtained by raising to the 1.38th power the cross-sectional area of each specific branch within the unit measurement space in the calibration curve by the regional correction value. A tea leaf yield prediction device characterized by the above.

2. The specific branch includes a cut branch pruned by the pruning, and a non-cut branch not pruned by the pruning; and the predicted number of new shoots is the number of cut branches at the time of pruning of the predicted tea garden, the number of non-cut branches at the time of pruning of the predicted tea garden, the predicted number of cut-branch new shoots with the cut branch as the mother branch at the time of picking of the predicted tea garden, and the predicted number of non-cut-branch new shoots with the non-cut branch as the mother branch at the time of picking of the predicted tea garden, and is calculated based on the above. The tea leaf yield prediction device according to Claim 1.

3. The prediction model is determined based on the measured yield value at the specific budding degree at the time of picking in the specific tea garden. The storage unit stores a budding degree correction value used for calculating a predicted yield value at a predicted budding degree different from the specific budding degree. The budding degree correction value is determined based on the budding degree of the tea leaves at the time of picking in the specific tea garden and the measured yield value of the specific tea garden. The acquisition unit acquires the predicted budding degree. The calculation unit multiplies the measured yield value of the tea leaves corresponding to the average value of the values obtained by raising the cross-sectional area of each specific branch in the unit measurement space in the calibration curve to the power of 1.38 by the regional correction value and the budding degree correction value to calculate a predicted yield value at the predicted budding degree. The tea leaf yield prediction device according to claim 1.

4. The prediction model is determined based on the measured yield value in the specific covering state of the specific tea garden. The storage unit stores a covering correction value used for calculating a predicted yield value in a predicted covering state different from the specific covering state. The covering correction value is determined based on the covering state of the tea trees in the specific tea garden and the measured yield value of the specific tea garden. The acquisition unit acquires the predicted covering state. The calculation unit multiplies the measured yield value of the tea leaves corresponding to the average value of the values obtained by raising the cross-sectional area of each specific branch in the unit measurement space in the calibration curve to the power of 1.38 by the regional correction value and the covering correction value to calculate a predicted yield value in the predicted covering state. The tea leaf yield prediction device according to claim 1.

5. A tea leaf yield prediction program characterized by causing a computer to function as the tea leaf yield prediction device according to any one of claims 1 to 4.

6. A tea leaf yield prediction method executed by a tea leaf yield prediction device that predicts the yield of tea leaves harvested by picking tea leaves from tea trees in a specific tea garden to be pruned in the following spring, wherein the tea leaf yield prediction device comprises a storage unit that stores a prediction model used for the prediction and a regional correction value used for calculating a predicted yield value of a predicted tea garden in a region different from the specific tea garden. The tea leaf yield prediction method includes an acquisition step of acquiring sample measurement information regarding the tea trees per unit measurement space at the time when the specific tea garden is pruned, and a calculation step of calculating a predicted yield value based on the regional correction value calculated based on the sample measurement information and the prediction model. ​ An output step for outputting the predicted value; comprising: the prediction model is the average value of the values obtained by raising to the power of 1.38 the cross-sectional area of each specific branch having a cross-sectional area of a predetermined area or more among a plurality of branches within the unit measurement space of the tea plants at the time of pruning of the specific tea garden; the measured value of the yield of the tea leaves at the time of harvesting of the specific tea garden; determined based on a calibration curve showing the relationship therebetween; the sample measurement information is branch information regarding the specific branch within the unit measurement space; including; the branch information being the specific number of new shoots with the specific branch within the unit measurement space as the mother branch at the time of harvesting of the specific tea garden; the predicted number of new shoots with the specific branch within the unit measurement space as the mother branch at the time of harvesting of the predicted tea garden; the regional correction value is a value calculated based on the specific number of new shoots and the predicted number of new shoots; the acquisition step is acquiring the branch information of the predicted tea garden; and the calculation step is multiplying the measured value of the yield of the tea leaves corresponding to the average value of the values obtained by raising to the power of 1.38 the cross-sectional area of each specific branch within the unit measurement space in the calibration curve by the regional correction value for calculation; A method for predicting tea leaf yield, characterized by the above.

Images