Quality control methods and programs for fruits and vegetables, as well as agricultural support methods and programs for agricultural support.

Abstract

This system accurately estimates the hardness of fruits and vegetables based on the number of days since harvest. [Solution] The method for quality control of fruits and vegetables involves generating an estimation model for each variety based on information about the variety of the cultivated fruits and vegetables, information about the cultivation environment during the cultivation period of the fruits and vegetables, and information about the changes in hardness of the fruits and vegetables after harvest, based on information about the cultivation environment during the cultivation period of the fruits and vegetables, and for each variety, an estimation model is generated that estimates the hardness of the fruits and vegetables according to the number of days after harvest from the information about the cultivation environment during the cultivation period of the fruits and vegetables. The estimation model is then performed by a computer to obtain information about the variety of the first fruits and vegetables to be estimated and information about the cultivation environment during the cultivation period of the first fruits and vegetables, and input the information about the cultivation environment during the cultivation period of the first fruits and vegetables into the estimation model corresponding to the variety of the first fruits and vegetables, thereby estimating the hardness of the first fruits and vegetables according to the number of days after harvest.

Description

Technical Field

[0001] The present invention relates to a method for quality control of fruits and vegetables, a program for quality control of fruits and vegetables, an agricultural support method, and an agricultural support program.

Background Art

[0002] Fruits and vegetables such as fruits and vegetables continue their metabolic activities even after harvesting, and it is known that the quality such as fruit components and hardness changes during the storage period after harvesting. Fruits and vegetables shift from the cultivation environment until then to the post-harvest environment by harvesting, and enter an environment where water, nutrients, and light are not supplied. Since they continue their life activities using limited accumulated nutrients, freshness deterioration and quality degradation that were not seen before harvesting occur (for example, see Non-Patent Document 1, etc.).

[0003] Conventionally, various studies have been conducted regarding quality degradation, etc. (for example, see Non-Patent Documents 1-6, etc.). In addition, a technique for estimating the quality of fruits and vegetables from the appearance information and cultivation history of fruits and vegetables is known (for example, see Patent Documents 1-4, etc.).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Non-Patent Documents

[0005]

Non-Patent Document 1

[0006] However, there is no known technique for estimating the quality of harvested fruit during storage, nor is there any known technique for proposing cultivation conditions to achieve the desired quality during storage.

[0007] Therefore, the present invention aims to provide a quality control method and program for fruits and vegetables that can accurately estimate the hardness of fruits and vegetables after harvest. Furthermore, the present invention aims to provide an agricultural support method and program that can suggest appropriate cultivation conditions for fruits and vegetables. [Means for solving the problem]

[0008] The present invention relates to a method for managing the quality of fruits and vegetables after harvest, which involves a computer performing the following steps: generating an estimation model for each variety based on information about the variety of the cultivated fruits and vegetables, information about the cultivation environment during the cultivation period of the fruits and vegetables, and information about the changes in the hardness of the fruits and vegetables after harvest; acquiring information about the variety of the first fruits and vegetables to be estimated and information about the cultivation environment during the cultivation period of the first fruits and vegetables; and inputting the information about the cultivation environment during the cultivation period of the first fruits and vegetables into the estimation model corresponding to the variety of the first fruits and vegetables to be estimated, thereby estimating the hardness of the first fruits and vegetables according to the number of days after harvest.

[0009] The present invention relates to an agricultural support method for presenting cultivation conditions for fruits and vegetables, and is an agricultural support method in which a computer performs the following processing: generating an estimation model for each variety to estimate second cultivation environment information to match the day on which the hardness after harvest falls outside a predetermined range with a desired date, based on information about the variety of the cultivated fruits and vegetables, information about a first cultivation environment and information about a second cultivation environment during the cultivation period of the fruits and vegetables, and information about the progression of the hardness of the fruits and vegetables after harvest; acquiring information about the variety of the first fruits and vegetables, information about the desired date and information about the first cultivation environment; inputting the information about the desired date and the information about the first cultivation environment into the estimation model corresponding to the variety of the first fruits and vegetables, thereby estimating and outputting second cultivation environment information to match the day on which the hardness after harvest falls outside a predetermined range with a desired date. [Effects of the Invention]

[0010] The present invention provides a method and program for quality control of fruits and vegetables, which have the effect of accurately estimating the hardness of fruits and vegetables after harvest. Furthermore, the present invention provides an agricultural support method and program, which have the effect of suggesting appropriate cultivation conditions for fruits and vegetables. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 is a schematic diagram showing the configuration of a quality control system for fruits and vegetables according to the first embodiment. [Figure 2] Figure 2 shows an example of a server hardware configuration. [Figure 3] Figure 3 is a functional block diagram of the server. [Figure 4] Figure 4 shows an example of cultivation and storage history information. [Figure 5] Figure 5 shows an example of cultivation and storage history information (after pretreatment). [Figure 6]Fig. 6(a) is a graph showing the relationship between the cumulative temperature from fruit set to harvest and the endocarp hardness of the cultivar "CF Momotaro York", and Fig. 6(b) is a graph showing the relationship between the cumulative temperature from fruit set to harvest and the endocarp hardness of the cultivar "Tom丸 Muncho". [Figure 7] Fig. 7(a) is a graph showing the relationship between the average cumulative solar radiation from fruit set to harvest and the endocarp hardness of the cultivar "CF Momotaro York", and Fig. 7(b) is a graph showing the relationship between the average cumulative solar radiation from fruit set to harvest and the endocarp hardness of the cultivar "Tom丸 Muncho". [Figure 8] Fig. 8(a) is a graph showing the relationship between the average EC from fruit set to harvest and the endocarp hardness of the cultivar "CF Momotaro York", and Fig. 8(b) is a graph showing the relationship between the average EC from fruit set to harvest and the endocarp hardness of the cultivar "Tom丸 Muncho". [Figure 9] Fig. 9(a) is a graph showing the relationship between the cumulative temperature from fruit set to harvest and the exocarp hardness of the cultivar "CF Momotaro York", and Fig. 9(b) is a graph showing the relationship between the cumulative temperature from fruit set to harvest and the exocarp hardness of the cultivar "Tom丸 Muncho". [Figure 10] Fig. 10(a) is a graph showing the relationship between the average cumulative solar radiation from fruit set to harvest and the exocarp hardness of the cultivar "CF Momotaro York", and Fig. 10(b) is a graph showing the relationship between the average cumulative solar radiation from fruit set to harvest and the exocarp hardness of the cultivar "Tom丸 Muncho". [Figure 11] Fig. 11(a) is a graph showing the relationship between the average EC from fruit set to harvest and the exocarp hardness of the cultivar "CF Momotaro York", and Fig. 11(b) is a graph showing the relationship between the average EC from fruit set to harvest and the exocarp hardness of the cultivar "Tom丸 Muncho". [Figure 12] Fig. 12 is a diagram showing an example of input information. [Figure 13] Fig. 13 is a diagram showing an example of the input information (after preprocessing). [Figure 14] Fig. 14 is a flowchart showing the process flow of the hardness estimation model generation process by the server. [Figure 15] Fig. 15 is a flowchart showing the process flow of the estimation process by the server. [Figure 16]Figure 16(a) is a graph showing the relationship between the estimated and measured values ​​of endocarp hardness using a hardness estimation model, and Figure 16(b) is a graph showing the relationship between the estimated and measured values ​​of exocarp hardness using a hardness estimation model. [Figure 17] Figure 17(a) is a graph showing the relationship between the average cumulative solar radiation from fruit setting to green ripening and the hardness of the inner peel for the variety "CF Momotaro York," and Figure 17(b) is a graph showing the relationship between the average cumulative solar radiation from fruit setting to green ripening and the hardness of the inner peel for the variety "Tomimaru Mucho." [Figure 18] Figure 18(a) is a graph showing the relationship between average soil moisture content from fruit setting to harvest and endocarp hardness for the variety "CF Momotaro York," and Figure 18(b) is a graph showing the relationship between average soil moisture content from fruit setting to harvest and endocarp hardness for the variety "Tomimaru Mucho." [Figure 19] Figure 19(a) is a graph showing the relationship between average soil moisture during the fruit-setting to green-ripening period and endocarp hardness for the variety "CF Momotaro York," and Figure 19(b) is a graph showing the relationship between average soil moisture during the fruit-setting to green-ripening period and endocarp hardness for the variety "Tomimaru Mucho." [Figure 20] Figure 20(a) is a graph showing the relationship between the average EC during fruit setting to green ripening and the hardness of the endocarp for the variety "CF Momotaro York," and Figure 20(b) is a graph showing the relationship between the average EC during fruit setting to green ripening and the hardness of the endocarp for the variety "Tomimaru Mucho." [Figure 21] Figure 21(a) is a graph showing the relationship between average cumulative solar radiation from fruit setting to green ripening and the hardness of the outer peel for the variety "CF Momotaro York," and Figure 21(b) is a graph showing the relationship between average cumulative solar radiation from fruit setting to green ripening and the hardness of the outer peel for the variety "Tomimaru Mucho." [Figure 22] Figure 22(a) is a graph showing the relationship between average soil moisture content from fruit setting to harvest and peel hardness for the variety "CF Momotaro York," and Figure 22(b) is a graph showing the relationship between average soil moisture content from fruit setting to harvest and peel hardness for the variety "Tomimaru Mucho." [Figure 23] Figure 23(a) is a graph showing the relationship between average soil moisture during the fruit-setting to green-ripening period and the hardness of the outer peel for the variety "CF Momotaro York," and Figure 23(b) is a graph showing the relationship between average soil moisture during the fruit-setting to green-ripening period and the hardness of the outer peel for the variety "Tomimaru Mucho." [Figure 24] Figure 24(a) is a graph showing the relationship between the average EC during fruit setting to green ripening and the hardness of the outer peel for the variety "CF Momotaro York," and Figure 24(b) is a graph showing the relationship between the average EC during fruit setting to green ripening and the hardness of the outer peel for the variety "Tomimaru Mucho." [Figure 25] Figure 25 is a diagram illustrating a modified example 1 of the first embodiment. [Figure 26] Figure 26 is a schematic diagram showing the configuration of the agricultural support system according to the second embodiment. [Figure 27] Figure 27 is a functional block diagram of the server shown in Figure 26. [Figure 28] Figure 28 is a flowchart showing the processing flow of the environmental control information estimation model generation process by the server according to the second embodiment. [Figure 29] Figure 29 is a flowchart showing the processing flow of the estimation process by the server according to the second embodiment. [Figure 30] Figure 30 shows an example of input information according to the second embodiment. [Figure 31] Figure 31 is a diagram illustrating a modified example 1 of the second embodiment. [Figure 32] Figure 32 is a graph showing the relationship between the average flowering-to-harvest temperature and the apparent modulus of elasticity for the strawberry variety "Koiminori". [Figure 33] Figure 33 is a graph showing the relationship between the cumulative solar radiation from flowering to harvest and the apparent modulus of elasticity for the strawberry variety "Koiminori". [Figure 34] Figure 34 is a graph showing the relationship between the estimated apparent modulus using a hardness estimation model and the measured value. [Figure 35] Figure 35(a) is a graph showing the relationship between the cumulative temperature from flowering to harvest and the apparent modulus of elasticity for the strawberry variety "Koiminori," and Figure 35(b) is a graph showing the relationship between the average humidity from flowering to harvest and the apparent modulus of elasticity for the strawberry variety "Koiminori." [Figure 36]Figure 36(a) is a graph showing the relationship between the average solar radiation from flowering to harvest and the apparent modulus of elasticity for the strawberry variety "Koiminori," and Figure 36(b) is a graph showing the relationship between the average daytime CO2 concentration from flowering to harvest and the apparent modulus of elasticity for the strawberry variety "Koiminori." [Modes for carrying out the invention]

[0012] 《First Embodiment》 The quality control system for fruits and vegetables according to the first embodiment will be described in detail below.

[0013] Figure 1 schematically shows the configuration of the quality control system 100 for fresh produce according to the first embodiment. In this embodiment, the fresh produce is, for example, tomatoes.

[0014] The quality management system 100 comprises a server 10, an administrator terminal 60, and a user terminal 70. The server 10, administrator terminal 60, and user terminal 70 are connected to a network 80 such as the Internet, a mobile phone network, a WAN (Wide Area Network), or a LAN (Local Area Network), enabling information exchange between the devices.

[0015] Here, since fresh produce continues to respire even after harvest, its quality changes even after harvesting. Generally, the optimal ripeness of fresh produce is considered the peak of its quality, and thereafter, the quality gradually declines. Furthermore, in the case of tomatoes, the firmness of the fruit is an important factor in determining whether or not it can be sold. If the change in fruit firmness according to the number of days after harvest can be predicted, the storage period can be estimated, and the sales method can be devised according to the storage period. The quality control system 100 of this first embodiment is a system that estimates the change in quality of fresh produce (tomatoes) after harvest (firmness according to the number of days after harvest) and outputs the quality retention period (number of days that can be stored) after harvest based on the estimated change in firmness.

[0016] (Server 10) Server 10 generates a model (hardness estimation model) that estimates the changes in hardness of fruits and vegetables after harvest, based on past cultivation information and storage history information of fruits and vegetables (hereinafter referred to as "cultivation and storage history information") entered from administrator terminal 60 and user terminal 70. The hardness estimation model is an equation obtained by multivariate analysis (multivariate analysis model) or a trained model obtained by machine learning. Server 10 also uses the generated hardness estimation model to estimate the changes in hardness of fruits and vegetables after harvest and the number of days they can be stored, and outputs the estimation results to user terminal 70.

[0017] Figure 2 shows the hardware configuration of server 10. As shown in Figure 2, server 10 includes a CPU (Central Processing Unit) 190, ROM (Read Only Memory) 192, RAM (Random Access Memory) 194, storage (HDD (Hard Disk Drive) or SSD (Solid State Drive), etc.) 196, a network interface 197, a display unit 193, an input unit 195, and a portable storage medium drive 199, etc. The display unit 193 includes a liquid crystal display or an organic EL display, and the input unit 195 includes a keyboard, mouse, touch panel, etc. Each of these components of server 10 is connected to a bus 198. In server 10, the CPU 190 executes programs stored in the ROM 192 or storage 196, or programs read from the portable storage medium 191 by the portable storage medium drive 199, thereby realizing the functions of each component shown in Figure 3. The functions of each part in Figure 3 may be implemented by integrated circuits such as ASICs (Application Specific Integrated Circuits) or FPGAs (Field Programmable Gate Arrays).

[0018] In server 10, the CPU 190 executes a program, thereby realizing the functions of the cultivation / storage history acquisition unit 40, preprocessing unit 42, estimation model generation unit 44, input information acquisition unit 46, estimation model acquisition unit 48, hardness transition estimation unit 50, quality retention period estimation unit 52, and output unit 54, as shown in Figure 3. Figure 3 also illustrates the history information DB 30, post-preprocessing information DB 32, and estimation model DB 34 stored in storage 196, etc.

[0019] The cultivation and storage history acquisition unit 40 acquires cultivation and storage history information entered from the administrator terminal 60 and the user terminal 70. The cultivation and storage history information includes information on the variety of the cultivated produce, information on the cultivation environment during the cultivation period of the produce, and information on the changes in the hardness of the produce after harvest. In the case of produce with both an exocarp and an endocarp, such as tomatoes, the hardness change information includes information on the changes in the hardness of the exocarp and information on the changes in the hardness of the endocarp.

[0020] Figure 4 shows an example of cultivation and storage history information. As shown in Figure 4, the cultivation and storage history information includes information on the variety of the actual cultivated produce (tomato) (CF Momotaro York, Tomimaru Mucho, ...), the date of Tomato Tone treatment, and the harvest date. In this embodiment, the date of Tomato Tone treatment is considered to be the date of fruit setting.

[0021] Furthermore, the cultivation and storage history information includes environmental information for a specified period, including the cultivation period. This environmental information includes the daily average temperature (°C) and daily cumulative solar radiation (MJ / m²). 2 This includes EC (mS / cm, hereafter referred to as EC within the rockwool slab). A rockwool slab is a mat used as a growing medium in hydroponics.

[0022] Furthermore, the cultivation and storage history information includes data on the storage history of the produce, specifically the values ​​for endocarp hardness and exocarp hardness at each of the following post-harvest days (1 day, 7 days, 14 days, and 21 days). Note that the post-harvest days of 1, 7, 14, and 21 days are just examples, and other days may also be used. Endocarp hardness and exocarp hardness can be measured using a physical property measuring device (texture analyzer, Stable Micro Systems, Inc.). In the measurement, a penetration test is performed from the endocarp side, and the load at which the endocarp breaks and the load at which the exocarp breaks are used as the respective hardness values. A reference for this measurement is "Journal of the Japan Society for Food Engineering, 66(11).408-419, 2019".

[0023] The cultivation and storage history acquisition unit 40 stores the cultivation and storage history information shown in Figure 4 in the history information DB 30 each time it is transmitted from the administrator terminal 60 or user terminal 70. The history information DB 30 stores a large amount of cultivation and storage history information, as shown in Figure 4.

[0024] The preprocessing unit 42 generates information suitable for generating a hardness estimation model from the cultivation and storage history information stored in the history information DB 30 and stores it in the post-preprocessing information DB 32. Figure 5 shows the information after preprocessing of the cultivation and storage history information in Figure 4 (cultivation and storage history information (post-preprocessing)). The cultivation and storage history information (post-preprocessing) includes the cumulative temperature from fruit setting to harvest, the average cumulative solar radiation from fruit setting to harvest, and the average EC from fruit setting to harvest. The cumulative temperature from fruit setting to harvest is the sum of the daily average temperatures from the fruit setting day to the day before harvest, as shown in Figure 4. The average cumulative solar radiation from fruit setting to harvest and the average EC from fruit setting to harvest are the average values ​​of the daily cumulative solar radiation and EC within the slab from the fruit setting day to the day before harvest, as shown in Figure 4.

[0025] The estimation model generation unit 44 retrieves the large amount of cultivation and storage history information (after preprocessing) generated by the preprocessing unit 42 from the preprocessing information DB 32 for each variety, and generates a hardness estimation model to estimate the hardness progression of each variety.

[0026] Here, we will explain the relationship between environmental information during the cultivation period (temperature, solar radiation, EC) and the hardness of tomatoes (endocarp hardness, exocarp hardness).

[0027] Figure 6(a) shows the relationship between the cumulative temperature from fruit setting to harvest and the hardness of the endocarp (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York". The cultivation sites were two high-ceiling greenhouses at the Tsukuba Station and the Anno Vegetable Research Station of the National Agriculture and Food Research Organization (NARO), and hydroponic cultivation using rockwool was performed. In addition, cultivation was performed in a "high sugar content plot" in which NaCl was added to the nutrient solution and a "control plot" in which NaCl was not added to the nutrient solution. The storage conditions were a temperature of 4°C. Figure 6(b) shows the relationship between the cumulative temperature from fruit setting to harvest and the hardness of the endocarp (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho".

[0028] In Figures 6(a) and 6(b), a positive correlation was found between the cumulative temperature from fruit setting to harvest and the hardness of the inner peel in almost all graphs, with a negative correlation only observed on day 21 in the high-sugar content section of Tomimaru Mucho.

[0029] Figure 7(a) shows the relationship between the average cumulative solar radiation from fruit setting to harvest and the hardness of the inner peel (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 7(b) shows the relationship between the average cumulative temperature from fruit setting to harvest and the hardness of the inner peel (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." In both Figure 7(a) and Figure 7(b), it was found that there was a positive correlation between the average cumulative solar radiation from fruit setting to harvest and the hardness of the inner peel in almost all graphs, with only the 21st day in the high-sugar content section of Tomimaru Mucho showing a negative correlation.

[0030] Figure 8(a) shows the relationship between the average EC from fruit setting to harvest and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 8(b) shows the relationship between the average EC from fruit setting to harvest and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 8(a) and 8(b), it can be seen that a negative correlation was observed for the variety "CF Momotaro York" except for day 21, and a positive correlation was observed for the variety "Tomimaru Mucho," indicating that there are differences between the varieties.

[0031] Figure 9(a) shows the relationship between the cumulative temperature from fruit setting to harvest and the hardness of the outer peel (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 9(b) shows the relationship between the cumulative temperature from fruit setting to harvest and the hardness of the outer peel (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho."

[0032] Figure 9(a) shows a positive correlation in the high-sugar content group of CF Momotaro York, but no clear correlation was observed in the control group. On the other hand, as shown in Figure 9(b), a negative correlation was observed in both experimental groups for Tomimaru Mucho. It is thought that the upward trend at 14 days post-harvest was influenced by the presence of soft fruits despite low cumulative solar radiation up to harvest, and hard fruits despite high cumulative solar radiation up to harvest.

[0033] Figure 10(a) shows the relationship between the average cumulative solar radiation from fruit setting to harvest and the hardness of the outer peel (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 10(b) shows the relationship between the average cumulative temperature from fruit setting to harvest and the hardness of the outer peel (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figure 10(a), a positive correlation was observed in the high-sugar content plot of CF Momotaro York, but no clear correlation was observed in the control plot. On the other hand, from Figure 10(b), a negative correlation was observed in both experimental plots for Tomimaru Mucho. It should be noted that the upward trend on the 14th day after harvest is thought to be due to the presence of soft fruits despite low cumulative solar radiation up to harvest, and hard fruits despite high cumulative solar radiation up to harvest.

[0034] Figure 11(a) shows the relationship between the average EC from fruit setting to harvest and the hardness of the outer peel (7, 14, and 21 days after harvest) when cultivating the variety "CF Momotaro York," and Figure 11(b) shows the relationship between the average EC from fruit setting to harvest and the hardness of the outer peel (7, 14, and 21 days after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 11(a) and 11(b), it was found that a positive correlation was observed for both CF Momotaro York and Tomimaru Mucho.

[0035] Figures 6(a) to 11(b) show that for each variety, there is a correlation between environmental information during the cultivation period (cumulative temperature from fruit setting to harvest, average cumulative solar radiation from fruit setting to harvest, average EC from fruit setting to harvest) and the hardness of the tomato (endocarp hardness, exocarp hardness). Therefore, in this embodiment, based on the correlation described above, a hardness estimation model (first estimation model) is generated for each variety using cultivation and storage history information (after pretreatment) to estimate the change in endocarp hardness of tomatoes after harvest from environmental information during the cultivation period, and a hardness estimation model (second estimation model) is generated to estimate the change in exocarp hardness of tomatoes after harvest from environmental information during the cultivation period.

[0036] The estimation model generation unit 44 uses multivariate analysis to determine an equation showing the relationship between environmental information during the cultivation period (cumulative temperature from fruit setting to harvest, average cumulative solar radiation from fruit setting to harvest, average EC from fruit setting to harvest) and endocarp hardness. In this case, for example, the following equation (multivariate analysis model) is obtained. Note that the value "0" in equations (1) and (2) is the coefficient for the variety "CF Momotaro York", and the value "1" in equations (3) and (4) is the coefficient for the variety "Tomimaru Mucho".

[0037] (Estimation model for the hardness of the endocarp of CF Momotaro York) Endocarp hardness: Hi(N) = -0.041287 × (Storage days n) - 0.050377 × (Fruit setting - Average EC at harvest) + 0.002744 × (Fruit setting - Cumulative temperature at harvest) + 0.124773 × 0 + 0.122725 × 0 × (Fruit setting - Average EC at harvest) - 0.505659 …(1)

[0038] In other words, equation (1) can be expressed as equation (1'). Endocarp hardness: Hi(N) = -0.041287 × (Storage days n) - 0.050377 × (Fruit setting - Average EC at harvest) + 0.002744 × (Fruit setting - Cumulative temperature at harvest) - 0.505659 …(1')

[0039] (Estimation model for the hardness of the outer peel of CF Momotaro York) Exocarp hardness Ho(N) = -0.023669 × (Storage days n) + 2.874204 × (Fruit setting - Average cumulative solar radiation at harvest) + 0.056334 × (Fruit setting - Average EC at harvest) - 0.004276 × (Fruit setting - Cumulative temperature at harvest) + 13.353262 × 0 + 0.080009 × 0 × (Fruit setting - Average EC at harvest) + 0.006137 × 0 × (Fruit setting - Cumulative temperature at harvest) - 3.868557 × 0 × (Fruit setting - Average cumulative solar radiation at harvest) - 7.525949 …(2)

[0040] In other words, equation (2) can be expressed as equation (2'). Exocarp hardness Ho(N) = -0.023669 × (Storage days n) + 2.874204 × (Average cumulative solar radiation at harvest - fruit set) + 0.056334 × (Average EC at harvest - fruit set) - 0.004276 × (Cumulative temperature at harvest - fruit set) - 7.525949 …(2')

[0041] (Estimation model for the hardness of the endocarp of Tomimaru Mucho) Endocarp hardness: Hi(N) = -0.041287 × (Storage days n) - 0.050377 × (Fruit setting - Average EC at harvest) + 0.002744 × (Fruit setting - Cumulative temperature at harvest) + 0.124773 × 1 + 0.122725 × 1 × (Fruit setting - Average EC at harvest) - 0.505659 …(3)

[0042] In other words, equation (3) can be expressed as equation (3'). Endocarp hardness Hi(N) = -0.041287 × (Storage days n) + 0.072348 × (Fruit setting - Average EC at harvest) + 0.002744 × (Fruit setting - Cumulative temperature at harvest) - 0.380886 …(3')

[0043] (Estimation model for the hardness of the outer peel of Tomimaru Mucho) Exocarp hardness Ho(N) = -0.023669 × (Storage days n) + 2.874204 × (Average cumulative solar radiation at harvest - fruit set) + 0.056334 × (Average EC at harvest - fruit set) - 0.004276 × (Cumulative temperature at harvest - fruit set) + 13.353262 × 1 + 0.080009 × 1 × (Average EC at harvest - fruit set) + 0.006137 × 1 × (Cumulative temperature at harvest - fruit set) - 3.868557 × 1 × (Average cumulative solar radiation at harvest - fruit set) - 7.525949 …(4)

[0044] In other words, equation (4) can be expressed as equation (4'). Exocarp hardness Ho(N) = -0.023669 × (Storage days n) - 0.994353 × (Average cumulative solar radiation from fruit set to harvest) + 0.136343 × (Average EC from fruit set to harvest) + 0.001861 × (Cumulative temperature from fruit set to harvest) - 5.827313 …(4')

[0045] The estimation model generation unit 44 stores the hardness estimation model generated as described above in the estimation model DB 34, linked to the variety.

[0046] Returning to Figure 3, the input information acquisition unit 46 receives information from the user terminal 70 (information necessary for estimating the changes in the hardness of the produce (the first produce to be estimated), and here, it is assumed that the information shown in Figure 12 is input to the user terminal 70. Specifically, as shown in Figure 12, the input information includes variety information (CF Momotaro York, Tomimaru Mucho, ...), tomato tone treatment date (fruit setting date), and harvest date. The input information also includes environmental information for a predetermined period including the cultivation period. The environmental information includes the daily average temperature (°C) and daily cumulative solar radiation (MJ / m²) for each day. 2 This includes EC (mS / cm) within the slab.

[0047] Furthermore, the input information acquisition unit 46 preprocesses the information in Figure 12 to generate preprocessed input information (hereinafter referred to as preprocessed input information) as shown in Figure 13, and transmits the generated preprocessed input information to the hardness transition estimation unit 50. The preprocessing performed by the input information acquisition unit 46 is the same as the preprocessing performed by the preprocessing unit 42 described above, and is a process that calculates the fruit-setting-harvest cumulative temperature, fruit-setting-harvest average cumulative solar radiation, and fruit-setting-harvest average EC shown in Figure 13 based on the input information in Figure 12.

[0048] The estimation model acquisition unit 48 acquires hardness estimation models from the estimation model DB 34 that correspond to (are associated with) the variety information acquired by the input information acquisition unit 46. In the case of tomatoes, where both endocarp and exocarp exist, the estimation model acquisition unit 48 acquires a hardness estimation model for estimating the hardness of the endocarp (equation (1') or (3') above) and a hardness estimation model for estimating the hardness of the exocarp (equation (2') or (4') above), corresponding to the acquired variety.

[0049] The hardness transition estimation unit 50 estimates the hardness transition after harvest by inputting (substituting) the input information (after preprocessing) received from the input information acquisition unit 46 into the hardness estimation model.

[0050] In the case of the "CF Momotaro York" variety, the hardness transition estimation unit 50 substitutes the values ​​from Figure 13 into equations (1') and (2') above to obtain equations that show the transitions of endocarp hardness Hi and exocarp hardness Ho.

[0051] From equation (1') above, Endocarp hardness: High = -0.041287 × n - 0.050377 × 11.17061899 + 0.002744 × 1114.188194 - 0.505659 Therefore, the relationship between Hi and n is, Endocarp hardness Hi = -0.041287 × n + 1.98893113148 …(1”) This is the result.

[0052] Also, from equation (2') above, Exocarp hardness Ho =-0.023669×n+2.874204×5.433702304+0.056334×11.17061899-0.004276×1114.188194-7.525949 Therefore, the relationship between Ho and n is, Exocarp hardness Ho = -0.023669 × n + 3.9566368296 …(2") This is the result.

[0053] The hardness transition estimation unit 50 transmits equations (equations (1) and (2) above) showing the transitions in endocarp hardness Hi and exocarp hardness Ho to the quality retention period estimation unit 52. The process by which the hardness transition estimation unit 50 calculates equations (1) and (2) above can be described as a process of estimating hardness according to the number of days after harvest.

[0054] Furthermore, the hardness transition estimation unit 50 can estimate the hardness (N) on each day after harvest by substituting the number of days after harvest (for example, n=1, 7, 21, 28) for n in the above equations (1") and (2).

[0055] For example, on the 7th day after harvest (n=7), Endocarp hardness Hi = -0.041287 × 7 + 1.98893113148 = 1.69992213148 (N) Exocarp hardness Ho = -0.023669 × 7 + 3.9566368296 = 3.7909538269 (N) The hardness transition estimation unit 50 may also output the estimated hardness for each day to the user terminal 70 via the output unit 54.

[0056] The quality retention period estimation unit 52 estimates the quality retention period of tomatoes after harvest based on an equation showing the hardness change received from the hardness change estimation unit 50. For example, the quality retention period estimation unit 52 determines the quality retention period to be the earlier of the following two days: the day on which the endocarp hardness Hi falls below a predetermined value (e.g., 1.5N), or the day on which the exocarp hardness Ho falls below a predetermined value (e.g., 2.0N).

[0057] The number of days when the endocarp hardness Hi is less than 1.5N can be determined from the above formula (1”). -0.041287 × n + 1.98893113148 = 1.5 …(5) ⇔n=(1.5-1.98893113148) / (-0.041287) ⇔n=11.8422537719

[0058] In other words, the day when the endocarp hardness (Hi) falls below 1.5N can be estimated to be 11 days after harvest.

[0059] On the other hand, the number of days when the exocarp hardness Ho is less than 2.0 N can be determined from the above formula (2”). -0.023669 × n + 3.9566368296 = 2.0 …(6) ⇔n=(2.0-3.9566368296) / (-0.023669) ⇔n=82.6666453843

[0060] In other words, the day when the endocarp hardness (Hi) falls below 1.5N can be estimated to be 82 days after harvest.

[0061] Therefore, according to the example above, the shelf life of tomatoes after harvest is 11 days, so the shelf life estimation unit 52 transmits the estimated shelf life (=11 days) to the output unit 54.

[0062] The output unit 54 outputs the estimated result of the quality retention period estimation unit 52 to the user terminal 70.

[0063] (Administrator terminal 60) The administrator terminal 60 is a device such as a PC (Personal Computer) or smartphone used by the administrator. The administrator inputs cultivation and storage history information obtained when growing fruits and vegetables (tomatoes), as well as cultivation and storage history information collected from agricultural workers, etc., into the administrator terminal 60. The administrator terminal 60 transmits the entered cultivation and storage history information to the server 10.

[0064] (User terminal 70) The user terminal 70 is a device such as a PC or smartphone used by users such as agricultural workers. When a user cultivates produce (tomatoes) themselves, they input cultivation and storage history information into the user terminal 70, and the user terminal 70 transmits the input cultivation and storage history information to the server 10. In addition, to check the shelf life of harvested produce (tomatoes), the user inputs information about the cultivation environment during the cultivation period into 70. In this case, the user terminal 70 transmits the input cultivation and storage history information to the server 10. Furthermore, when the server 10 outputs the estimated shelf life of the produce (tomatoes), the user terminal 70 displays the estimated result on the display unit of the user terminal 70. If a user wants to know the shelf life of tomatoes after harvest, assuming they were cultivated under predetermined environmental conditions, they should input information about those predetermined environmental conditions into the user terminal 70.

[0065] (Regarding processing) Next, we will explain the processing of server 10 in detail. The following explanation will be divided into the hardness estimation model generation process and the estimation process.

[0066] (Hardness estimation model generation process) Figure 14 shows a flowchart of the hardness estimation model generation process. It should be noted that the process in Figure 14 assumes that the history information DB30 contains a large amount of cultivation and storage history information for each variety.

[0067] When the process shown in Figure 14 begins, in step S10, the preprocessing unit 42 first waits until it is time to generate or update the hardness estimation model. The timing for generating or updating the hardness estimation model may be at predetermined intervals, each time a predetermined number of new cultivation / storage history entries are added to the history information DB 30, or at any other time.

[0068] When it is time to generate or update the hardness estimation model, the process moves to step S12, where the preprocessing unit 42 performs preprocessing of the cultivation and storage history information. Specifically, it generates the cultivation and storage history information (after preprocessing) shown in Figure 5 from the cultivation and storage history information shown in Figure 4. The preprocessing unit 42 stores the generated cultivation and storage history information (after preprocessing) in the preprocessed information DB 32.

[0069] Next, in step S14, the estimation model generation unit 44 generates a hardness estimation model by multivariate analysis. Specifically, the estimation model generation unit 44 reads the cultivation and storage history information (after preprocessing) of one variety from the preprocessed information DB 32, and uses the read cultivation and storage history information (after preprocessing) to generate a hardness estimation model for estimating the hardness of the endocarp and a hardness estimation model for estimating the hardness of the exocarp. The estimation model generation unit 44 also generates hardness estimation models for estimating the hardness of the endocarp and exocarp for other varieties in the same manner. As an example, let's assume that equations (1') and (2') above are generated for the variety "CF Momotaro York", and equations (3') and (4') above are generated for the variety "Tomimaru Mucho".

[0070] Next, in step S16, the estimation model generation unit 44 stores the generated hardness estimation model in the estimation model DB 34.

[0071] With the above steps, the entire process of generating the hardness estimation model shown in Figure 14 is completed.

[0072] (Estimated process) Next, the estimation process using the hardness estimation model will be explained following the flowchart in Figure 15.

[0073] In the process shown in Figure 15, first, in step S30, the input information acquisition unit 46 waits until input information is transmitted from the user terminal 70. Once input information is transmitted from the user terminal 70, the input information acquisition unit 46 proceeds to step S32.

[0074] When the process moves to step S32, the input information acquisition unit 46 acquires the input information (see Figure 12). Next, in step S34, the input information acquisition unit 46 performs preprocessing of the input information. In this preprocessing, the input information acquisition unit 46 generates input information (after preprocessing) as shown in Figure 13 from the input information in Figure 12, and transmits the generated information to the estimation model acquisition unit 48 and the hardness transition estimation unit 50.

[0075] Next, in step S36, the estimation model acquisition unit 48 acquires (reads out) a hardness estimation model corresponding to the variety from the estimation model DB 34. For example, as shown in Figure 13, if the variety information included in the input information (after preprocessing) is "CF Momotaro York", the estimation model acquisition unit 48 reads out the above equations (1') and (2') as the hardness estimation model.

[0076] Next, in step S38, the hardness transition estimation unit 50 uses the acquired hardness estimation model to obtain equations that show the hardness transition. Specifically, the hardness transition estimation unit 50 obtains equations (1") and (2) by substituting the values ​​shown in Figure 13 into equations (1') and (2'). The hardness transition estimation unit 50 transmits equations (1") and (2) to the quality retention period estimation unit 52.

[0077] Next, in step S40, the quality retention period estimation unit 52 estimates the quality retention period using the above equations (1) and (2). In this case, from the above equations (5) and (6), the quality retention period can be estimated to be 11 days. The quality retention period estimation unit 52 transmits the estimation result to the output unit 54.

[0078] Next, in step S42, the output unit 54 outputs the estimation result of the quality retention period estimation unit 52 to the user terminal 70. The display unit of the user terminal 70 displays the quality retention period of the harvested produce (tomatoes), so the user can display the quality retention period on the tomatoes when selling them, or set a price that takes the quality retention period into account when selling tomatoes.

[0079] As described in detail above, according to this first embodiment, the estimation model generation unit 44 generates a hardness estimation model (equations (1') to (4') above) for each variety, which estimates the hardness of the tomato according to the number of days after harvest, based on the cultivation history information (which includes information on the variety of tomato actually cultivated, information on the cultivation environment during the cultivation period of the tomato, and information on the change in hardness of the tomato after harvest) (Figure 4). The input information acquisition unit 46 acquires information on the variety of tomato to be estimated and information on the cultivation environment during the cultivation period of the tomato to be estimated. The estimation model acquisition unit 48 then acquires a hardness estimation model corresponding to the variety of tomato to be estimated, and the hardness change estimation unit 50 estimates the hardness according to the number of days after harvest of the tomato to be estimated by inputting the information on the cultivation environment during the cultivation period of the tomato to be estimated into the hardness estimation model acquired by the estimation model acquisition unit 48 (it obtains an equation that shows the change in hardness according to the number of days). This allows users to check the changes in hardness of tomatoes based on the number of days after harvest simply by inputting information about the growing environment during the cultivation period of the tomato being estimated.

[0080] In this first embodiment, when the hardness of the endocarp was estimated using equations (1') and (3') above, it was found that there was a relationship between the estimated value and the measured value of the endocarp as shown in Figure 16(a). The approximation line for each plot in Figure 16(a) is y=x, and its coefficient of determination R 2 The value was 0.638, indicating that an appropriate value was obtained as an estimate of the hardness of the endocarp. Furthermore, when the hardness of the exocarp was estimated using the above equations (2') and (4'), it was found that there was a relationship between it and the measured value of the exocarp as shown in Figure 16(b). The approximation line for each plot in Figure 16(b) is y=x, and its coefficient of determination R 2 The value was 0.521, which was found to be an appropriate value for estimating the hardness of the exocarp.

[0081] Furthermore, in this first embodiment, the quality retention period estimation unit 52 identifies the number of days after harvest when the hardness of the tomato falls outside a predetermined range as the quality retention period, and the output unit 54 outputs the quality retention period information to the user terminal 70. As a result, the user can check how many days the quality of the tomato will be maintained after harvest simply by inputting information about the cultivation environment during the cultivation period of the tomato to be estimated. Therefore, the user can display the quality retention period on the tomatoes when selling them, or set prices that take the quality retention period into consideration when selling tomatoes. In addition, by knowing the quality retention period of tomatoes, consumers can reduce food waste (reduce food loss) and purchase tomatoes that are close to their quality retention period at a fair price.

[0082] Furthermore, in this first embodiment, when there is an endocarp and an exocarp, such as in tomatoes, the estimation model generation unit 44 generates a hardness estimation model for estimating the hardness of the endocarp and a hardness estimation model for estimating the hardness of the exocarp. The quality retention period estimation unit 52 then identifies the earlier of the number of days after harvest when the hardness of the endocarp falls below a predetermined value (outside a predetermined first range) and the number of days after harvest when the hardness of the exocarp falls below a predetermined value (outside a predetermined second range), and sets this as the quality retention period. This makes it possible to appropriately estimate the quality retention period based on the condition of both the endocarp and the exocarp.

[0083] In the first embodiment described above, the case where the environmental information used to generate the hardness estimation model and the environmental information input to the hardness estimation model are the cumulative temperature from fruit setting to harvest, the average cumulative solar radiation from fruit setting to harvest, and the average EC from fruit setting to harvest was explained (Figures 5 and 13). However, it is not limited to this, and other environmental information can also be used as the environmental information used to generate the hardness estimation model and the environmental information input to the hardness estimation model. For example, other environmental information that can be used includes "average cumulative solar radiation from fruit setting to green ripening," "soil moisture from fruit setting to green ripening," "average soil moisture from fruit setting to green ripening," and "average EC from fruit setting to green ripening." "Average cumulative solar radiation from fruit setting to green ripening" means the average cumulative solar radiation between the fruit setting day and the day when the green ripening stage is reached. The day when the green ripening stage is reached can be, for example, the day that is the number of days prior to the harvest day when a predetermined cumulative temperature (e.g., 420°C) is reached. Furthermore, "average soil moisture from fruit setting to harvest" refers to the average soil moisture from the day of fruit setting to the day before harvest, and "average soil moisture from fruit setting to green ripening" refers to the average soil moisture between the day of fruit setting and the day the green ripening stage is reached. Additionally, "average EC from fruit setting to green ripening" refers to the average EC between the day of fruit setting and the day the green ripening stage is reached.

[0084] Figure 17(a) shows the relationship between the average cumulative solar radiation from fruit setting to green ripening and the hardness of the inner peel (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 17(b) shows the relationship between the cumulative temperature from fruit setting to green ripening and the hardness of the inner peel (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 17(a) and 17(b), it was found that there is a positive correlation between the average cumulative solar radiation from fruit setting to green ripening and the hardness of the inner peel in almost all graphs, and a negative correlation only on day 21 in the high-sugar content section of Tomimaru Mucho.

[0085] Figure 18(a) shows the relationship between average soil moisture content from fruit setting to harvest and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 18(b) shows the relationship between average soil moisture content from fruit setting to harvest and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 18(a) and 18(b), it was found that there is a negative correlation between average soil moisture content from fruit setting to harvest and endocarp hardness in almost all graphs, and a positive correlation is observed only on day 21 in the high-sugar content section of Tomimaru Mucho.

[0086] Figure 19(a) shows the relationship between average soil moisture during the fruit-setting to green ripening period and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 19(b) shows the relationship between average soil moisture during the fruit-setting to green ripening period and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 19(a) and 19(b), it was found that there is a negative correlation between average soil moisture during the fruit-setting to green ripening period and endocarp hardness in almost all graphs, and a positive correlation is observed only on day 21 in the high-sugar content section of Tomimaru Mucho.

[0087] Figure 20(a) shows the relationship between the average EC during fruit setting to green ripening and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 20(b) shows the relationship between the average EC during fruit setting to green ripening and endocarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 20(a) and 20(b), it can be seen that CF Momotaro York showed a negative correlation except for day 21, while Tomimaru Mucho showed a positive correlation, indicating differences between the varieties.

[0088] Figure 21(a) shows the relationship between the average cumulative solar radiation from fruit setting to green ripening and the hardness of the outer peel (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 21(b) shows the relationship between the average cumulative temperature from fruit setting to green ripening and the hardness of the outer peel (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figure 21(a), a positive correlation was observed in the high-sugar content group of CF Momotaro York, but no clear relationship was observed in the control group. On the other hand, as shown in Figure 21(b), a negative correlation was observed in both experimental groups for Tomimaru Mucho. It should be noted that the upward trend on the 14th day after harvest is thought to be due to the presence of soft fruits despite low cumulative solar radiation up to harvest, and hard fruits despite high cumulative solar radiation up to harvest.

[0089] Figure 22(a) shows the relationship between average soil moisture content from fruit setting to harvest and pericarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 22(b) shows the relationship between average soil moisture content from fruit setting to harvest and pericarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 22(a) and 22(b), a negative correlation was observed in the high-sugar content section of CF Momotaro York, but no clear relationship was observed in the control section or for Tomimaru Mucho.

[0090] Figure 23(a) shows the relationship between average soil moisture during the fruit-setting to green ripening period and pericarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "CF Momotaro York," and Figure 23(b) shows the relationship between average soil moisture during the fruit-setting to green ripening period and pericarp hardness (7th, 14th, and 21st day after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 23(a) and 23(b), a negative correlation was observed in the high-sugar content section of CF Momotaro York, but no clear relationship was observed in the control section or for Tomimaru Mucho.

[0091] Figure 24(a) shows the relationship between the average EC during fruit setting to green ripening and the hardness of the outer peel (7, 14, and 21 days after harvest) when cultivating the variety "CF Momotaro York," and Figure 24(b) shows the relationship between the average EC during fruit setting to green ripening and the hardness of the outer peel (7, 14, and 21 days after harvest) when cultivating the variety "Tomimaru Mucho." From Figures 24(a) and 24(b), it was found that a positive correlation was observed for both CF Momotaro York and Tomimaru Mucho.

[0092] As described above, the correlation between each environmental piece of information and the endocarp hardness or exocarp hardness may change depending on the variety, cultivation conditions, etc., as shown in Figures 17(a) to 24(b). Therefore, each environmental piece of information should be incorporated into the hardness estimation model based on these factors. In addition, for environmental pieces of information other than those described in Figures 17(a) to 24(b), the correlation may be confirmed in the same manner as in Figures 17(a) to 24(b), and this correlation may be incorporated into the hardness estimation model. Examples of environmental pieces of information other than those described in Figures 17(a) to 24(b) include environmental data for the entire cultivation period (average cumulative solar radiation, average soil moisture, average EC, etc.) and environmental data for the same period in the previous year or in typical years (average cumulative solar radiation, average soil moisture, average EC, etc.).

[0093] (Variation 1) In the first embodiment described above, the hardness estimation model was described as an equation obtained by multivariate analysis, but it is not limited to this. For example, as shown in Figure 25, the estimation model generation unit 44 may generate a trained model that outputs the endocarp hardness and exocarp hardness on day n based on environmental information during the cultivation period (accumulated temperature from fruit setting to harvest, average accumulated solar radiation from fruit setting to harvest, average EC from fruit setting to harvest) by machine learning the cultivation and storage history information (after preprocessing) in the hardness estimation model generation process. In this case, the hardness transition estimation unit 50 should output the endocarp hardness and exocarp hardness on day n by inputting the input information (after preprocessing) to the trained model. The quality retention period estimation unit 52 should then estimate the quality retention period based on the days on which the endocarp hardness and exocarp hardness on day n output from the hardness transition estimation unit 50 exceed predetermined values. The same effects as in the first embodiment can be obtained in this way as well.

[0094] (Modification 2) Furthermore, the hardness estimation model of the first embodiment described above may be a model that estimates and outputs the hardness of fruits and vegetables according to the number of days after harvest, based on information about the cultivation environment and the storage conditions after harvest. The storage conditions after harvest include conditions such as whether or not a treatment to adsorb ethylene was carried out in the storage environment, whether or not a treatment to inhibit ethylene synthesis was carried out, and whether or not a treatment to suppress the softening of fruits and vegetables during long-term storage was carried out. Treatment to adsorb ethylene includes placing a freshness-preserving catalyst (palladium-treated zeolite or potassium permanganate-treated zeolite) in the storage environment. Treatment to inhibit ethylene synthesis includes treatment with 1-MCP (1-methylcyclopropene), CaCl2, SA (salicylic acid), and UV-C (deep ultraviolet) irradiation. Treatment to suppress the softening of fruits and vegetables (tomatoes) during long-term storage includes treatment at temperatures of 36°C or 40°C after harvest. Although these treatments are publicly known, by incorporating the effects of these treatments into the hardness estimation model, it becomes possible to estimate the hardness of fruits and vegetables according to the number of days after harvest with greater accuracy.

[0095] In the first embodiment and its modifications described above, the example described the case where the fruit is a tomato. However, the example is not limited to this, and may include other fruits and vegetables (for example, fruit-like vegetables (strawberries, watermelons, melons, etc.), fruit vegetables (tomatoes, cucumbers, eggplants, bell peppers, pumpkins, etc.), stone fruits (apricots, plums, cherries, Japanese apricots, nectarines, prunes, peaches, etc.), kernel fruits (pears, European pears, loquats, apples, etc.), and tropical fruits (avocados, kiwifruit, durian, papaya, mango, etc.)). In cases where both the endocarp and exocarp are absent, such as with tomatoes, only an exocarp hardness estimation model may be generated, and the quality retention period may be estimated using the estimated results of the change in exocarp hardness.

[0096] Furthermore, if the produce has a hard outer peel, such as an avocado, it may be harvested and sold before it is fully ripe. In such cases, the degree of ripeness may be determined by the hardness of the outer peel. Therefore, the server 10 and the user terminal 70 may use a hardness estimation model to estimate the hardness of the outer peel and estimate how many days after harvest the produce will be ready to eat once the estimated hardness exceeds a predetermined value.

[0097] (Example using strawberries as the fruit) Here, we will describe an example where the fruit is strawberries.

[0098] In this embodiment, the "apparent modulus of elasticity (MPa)" was used to represent the hardness of the strawberry fruit. Details about the apparent modulus of elasticity can be found, for example, in "Kaoru Kamiyama et al., 'A Simple Mechanical Index for Indicating the Quality of Strawberry Fruit During Storage,' Food Research Institute Report, No. 77, 1-11 (2013)." It can be determined using a texture analyzer (SMS Corporation) by pressing a plunger against the strawberry fruit and measuring the force and displacement of the plunger until just before the fruit breaks, along with the fruit diameter (thickness in the direction of crushing) and the cross-sectional area of ​​the plunger.

[0099] Here, we will explain the relationship between environmental information during the cultivation period (temperature, solar radiation) and the hardness of the strawberry fruit (apparent modulus of elasticity).

[0100] Figure 32 shows the relationship between the average flowering-to-harvest temperature and the apparent modulus of elasticity (immediately after harvest: 1 day of storage) when cultivating the strawberry variety "Koiminori". In Figure 32, a negative correlation was found between the average flowering-to-harvest temperature and the apparent modulus of elasticity.

[0101] Figure 33 shows the relationship between the cumulative solar radiation from flowering to harvest and the apparent modulus of elasticity (immediately after harvest) when cultivating the strawberry variety "Koiminori". In Figure 33, a negative correlation was found between the cumulative solar radiation from flowering to harvest and the apparent modulus of elasticity. Although not shown in the figure, similar correlations were found between the average temperature from flowering to harvest and the cumulative solar radiation from flowering to harvest for the apparent modulus of elasticity after 7, 14, and 21 days of storage.

[0102] Figures 32 and 33 show that, as described above, there is a correlation between environmental information during the cultivation period (average temperature from flowering to harvest, and cumulative solar radiation from flowering to harvest) and the hardness of the fruit (apparent modulus of elasticity) in strawberries. Therefore, in this example, based on the correlation described above, we decided to generate a hardness estimation model for each variety to estimate the changes in the hardness (apparent modulus of elasticity) of strawberries after harvest from the environmental information during the cultivation period, using cultivation and storage history information (average temperature, cumulative solar radiation, storage temperature, and number of storage days).

[0103] More specifically, we used multivariate analysis to derive an equation showing the relationship between environmental information during the cultivation period (average temperature from flowering to harvest and cumulative solar radiation from flowering to harvest), storage history information (storage temperature, number of storage days), and fruit hardness (apparent modulus of elasticity). In this case, for example, the following equation (multivariate analysis model) was obtained.

[0104] (Estimation model for the apparent elastic modulus of the "Koiminori" variety) Apparent modulus of elasticity (MPa) = -0.030324 × (Storage days n) - 0.176368 × (Storage temperature t) - 0.026237 × (Storage days n) × (Storage temperature t) - 0.551896 × (Average temperature from flowering to harvest) - 0.025821 × (Cumulative solar radiation from flowering to harvest) + 26.933166 …(7)

[0105] When the apparent modulus was predicted using equation (7) above, it was found that there is a relationship between the apparent modulus and the measured value as shown in Figure 34. The approximate straight line for each plot in Figure 34 is y=x, and its coefficient of determination R 2 The value was 0.67, which was found to be an appropriate value as an estimate of the apparent modulus of elasticity.

[0106] (Model for estimating shelf life) Furthermore, in this embodiment, assuming that the apparent modulus of elasticity of the fruit is 7.5 MPa or less, the desire to purchase decreases due to the deterioration of tactile feel associated with softening, and the shelf life is determined from the following equation (7'), which is a modified version of equation (7) above.

[0107] Quality retention period =[7.5-{-0.176368×(storage temperature t)-0.551896×(average temperature between flowering and harvest)-0.025821×(cumulative solar radiation between flowering and harvest)+26.933166}]÷{-0.026237×(storage temperature t)-0.030324}…(7') Using the above equation (7'), it becomes possible to arbitrarily control the shelf life by controlling the storage temperature t, which can be controlled artificially.

[0108] Furthermore, when cultivating the strawberry variety "Koiminori," there is a positive correlation between the cumulative temperature from flowering to harvest and the apparent modulus of elasticity (immediately after harvest: 1 day of storage), as shown in Figure 35(a). Also, when cultivating the strawberry variety "Koiminori," there is a positive correlation between the average humidity from flowering to harvest and the apparent modulus of elasticity (immediately after harvest: 1 day of storage), as shown in Figure 35(b). In addition, when cultivating the strawberry variety "Koiminori," there is a negative correlation between the average solar radiation from flowering to harvest and the apparent modulus of elasticity (immediately after harvest: 1 day of storage), as shown in Figure 36(a). Furthermore, when cultivating the strawberry variety "Koiminori," there is a positive correlation between the average daytime CO2 concentration from flowering to harvest and the apparent modulus of elasticity (immediately after harvest: 1 day of storage), as shown in Figure 36(b). Therefore, in the estimation models of equations (7) and (7') above, at least one of the following variables may be used instead of at least one of the flowering-harvest average temperature and flowering-harvest cumulative solar radiation: flowering-harvest cumulative temperature, flowering-harvest average humidity, flowering-harvest average solar radiation, and flowering-harvest average daytime CO2 concentration.

[0109] 《Second Embodiment》 The following describes an agricultural support system according to the second embodiment. Figure 26 schematically shows the configuration of the agricultural support system 200 according to the second embodiment. Similar to the quality control system 100 of the first embodiment, the agricultural support system 200 includes a server 110, an administrator terminal 60, and a user terminal 70, and the devices are connected to each other via a network 80.

[0110] The server 110 in this second embodiment is a device that, when a user (such as a farmer) inputs information (desired date information) about the period during which they want to maintain the quality of the harvested produce, estimates and presents what kind of cultivation environment the produce should be grown in in order to maintain its quality for the input period.

[0111] Figure 27 shows a functional block diagram of the server 110. As can be seen by comparing it with Figure 3 (server 10 of the first embodiment), the server 110 of this second embodiment is characterized in that it has an environmental control information estimation unit 150 instead of the hardness transition estimation unit 50 and the quality retention period estimation unit 52 shown in Figure 3.

[0112] The following describes the processing performed by server 110, following Figures 28 and 29.

[0113] Figure 28 shows a flowchart of the environmental control information estimation model generation process. It should be noted that the process in Figure 28 assumes that the history information DB30 contains a large amount of cultivation and storage history information (Figure 4) for each variety.

[0114] When the process shown in Figure 28 begins, first, in step S110, the preprocessing unit 42 waits until it is time to generate or update the environmental control information estimation model. The timing for generating or updating the environmental control information estimation model may be at predetermined intervals, each time a predetermined number of new cultivation / storage history entries are added to the history information DB 30, or at any other time.

[0115] When it is time to generate or update the environmental control information estimation model, the process moves to step S112, where the preprocessing unit 42 performs preprocessing of the cultivation and storage history information. Specifically, it generates the cultivation and storage history information (after preprocessing) shown in Figure 5 from the cultivation and storage history information shown in Figure 4. The preprocessing unit 42 stores the generated cultivation and storage history information (after preprocessing) in the preprocessed information DB 32.

[0116] Next, in step S114, the estimation model generation unit 44 generates an environmental control information estimation model by multivariate analysis. Specifically, the estimation model generation unit 44 reads the cultivation and storage history information (after preprocessing) of one variety from the preprocessed information DB 32, and uses the read cultivation and storage history information (after preprocessing) to generate an environmental control information estimation model using the same formula as the estimation model for estimating the hardness of the inner peel. The estimation model generation unit 44 also generates environmental control information estimation models for other varieties in the same manner.

[0117] For example, for the variety "CF Momotaro York," the following equation (8), which is the same as equation (1') above, is generated, and for the variety "Tomimaru Mucho," the following equation (9), which is the same as equation (3') above, is generated.

[0118] (CF Momotaro York's environmental control information estimation model) Endocarp hardness: Hi(N) = -0.041287 × (Storage days n) - 0.050377 × (Fruit setting - Average EC at harvest) + 0.002744 × (Fruit setting - Cumulative temperature at harvest) - 0.505659 …(8)

[0119] (Tomimaru Mucho's Environmental Control Information Estimation Model) Endocarp hardness Hi(N) = -0.041287 × (Storage days n) + 0.072348 × (Fruit setting - Average EC at harvest) + 0.002744 × (Fruit setting - Cumulative temperature at harvest) - 0.380886 …(9)

[0120] Next, in step S116, the estimation model generation unit 44 stores the generated environmental control information estimation model in the estimation model DB 34.

[0121] With the above steps completed, the entire process of generating the environmental control information estimation model shown in Figure 28 is finished.

[0122] (Estimated process) Next, the estimation process using the environmental control information estimation model will be explained following the flowchart in Figure 29.

[0123] In the process shown in Figure 29, first, in step S130, the input information acquisition unit 46 waits until input information is transmitted from the user terminal 70. Once input information is transmitted from the user terminal 70, the input information acquisition unit 46 proceeds to step S132.

[0124] When the system moves to step S132, the input information acquisition unit 46 acquires the input information (see Figure 30). The input information includes tomato variety information, cumulative temperature from fruiting to harvest, and storage period (days). For example, the user inputs the tomato variety they are cultivating and information on how long (in days) they want to maintain the quality after harvest (desired date information) into the user terminal 70. The user also inputs either the cumulative temperature from fruiting to harvest during the cultivation period or the average EC from fruiting to harvest. For example, if the cultivation environment to be controlled when maintaining quality until the desired date is EC, the user inputs the cumulative temperature from fruiting to harvest information. If the user knows the value of the cumulative temperature from fruiting to harvest during the cultivation period, they input that value into the user terminal 70. If they do not know it, they input the information necessary to calculate the cumulative temperature from fruiting to harvest from average values, etc. (information on fruiting date and harvest date). The input information acquisition unit 46 transmits the acquired input information to the estimation model acquisition unit 48 and the environmental control information estimation unit 150.

[0125] Next, in step S136, the estimation model acquisition unit 48 acquires (reads out) an environmental control information estimation model corresponding to the variety from the estimation model DB 34. For example, as shown in Figure 30, if the variety information included in the input information is "CF Momotaro York", the estimation model acquisition unit 48 reads out the above equation (8) as the environmental information estimation model.

[0126] Next, in step S138, the environmental control information estimation unit 150 uses the acquired environmental control information estimation model to estimate environmental control information (fruit-harvest average EC) to match the quality retention period with the desired date. Specifically, the environmental control information estimation unit 150 substitutes the values ​​shown in Figure 30 into equation (8) above, Endocarp hardness: Hi(N) = -0.041287 × 10 - 0.050377 × (Fruit set - Average EC at harvest) + 0.002744 × 1114.188194 - 0.505659 In addition to obtaining the following equation, the average fruit-to-harvest EC when the endocarp hardness Hi is outside the predetermined range (less than the predetermined value (e.g., 1.5N)) is calculated from the following equation (8'). 1.5>-0.041287×10 -0.050377×(fruit set - average EC at harvest) +0.002744×1114.188194 -0.505659 …(8') ⇔(Fruit set - Average EC at harvest) = (1.5 + 0.041287 × 10 - 0.002744 × 1114.188194 + 0.505659) / (-0.050377) ⇔(Fruit set - Average EC at harvest)>12.6804574376

[0127] The environmental control information estimation unit 150 transmits the calculated fruit-set-harvest average EC (=12.6804574376) to the output unit 54.

[0128] Next, in step S142, the output unit 54 outputs the estimation result of the environmental control information estimation unit 150 to the user terminal 70. The display unit of the user terminal 70 displays the average EC value from fruit setting to harvest to set the quality retention period to 10 days, so that the user can appropriately adjust the EC to set the quality retention period to 10 days. If there is an adjustment device that automatically adjusts the EC, the average EC value from fruit setting to harvest may be output to the adjustment device and the EC may be adjusted automatically.

[0129] As described in detail above, according to this second embodiment, the estimation model generation unit 44 generates an environmental control information estimation model for each variety that estimates second cultivation environment information to match the day on which the hardness after harvest falls outside a predetermined range (quality retention period) with the desired date, based on variety information, first cultivation environment information (e.g., cumulative temperature from fruit setting to harvest) and second cultivation environment information (e.g., average EC from fruit setting to harvest) during the tomato cultivation period, and information on the change in hardness of the produce after harvest. Furthermore, when the input information acquisition unit 46 acquires input information (variety information, desired date information (storage period), and first cultivation environment information (e.g., cumulative temperature from fruit setting to harvest)), the environmental control information estimation unit 150 estimates second cultivation environment information to match the quality retention period with the desired date by inputting the desired date information and the first cultivation environment information into the environmental control information estimation model corresponding to the input variety, and outputs it to the user terminal 70 via the output unit 54. This allows users to precisely control the environment to match the desired shelf life.

[0130] In the second embodiment described above, if the environmental control information estimation model is an expression in which m (m≧3) environmental information variables are used (for example, in the above expressions (3') and (4')), then (m-1) environmental information variables should be substituted into the environmental control information estimation model.

[0131] (Variation 1) In the second embodiment described above, the case in which the environmental control information estimation model is an equation obtained by multivariate analysis was explained, but it is not limited to this. For example, as shown in Figure 31, the estimation model generation unit 44 may generate a trained model (environmental control information estimation model) by machine learning the cultivation and storage history information (after preprocessing) in the environmental control information estimation model generation process. In this case, the environmental control information estimation unit 150 can estimate the environmental information necessary to set the quality retention period to the desired date by inputting the input information to the trained model. Even in this way, the same effects as in the second embodiment can be obtained.

[0132] (Modification 2) Furthermore, the environmental control information estimation model of the second embodiment described above may also be a model that estimates and outputs the hardness of fruits and vegetables according to the number of days after harvest, based on information about the cultivation environment and the storage conditions after harvest, similar to the modified example 2 of the first embodiment described above.

[0133] In the second embodiment and its modifications described above, the case where the fruit is a tomato was explained, but it is not limited to this, and other fruits may be used, as in the first embodiment and its modifications.

[0134] In the first and second embodiments and their variations described above, the hardness estimation model and environmental control information estimation model are generated in the server 10 and used in the server 10. However, the invention is not limited to these cases. For example, the hardness estimation model and environmental control information estimation model may be used in the user terminal 70 to estimate the hardness progression, quality retention period, and environmental control information.

[0135] The above processing functions can be implemented by a computer. In this case, a program describing the processing content of the functions that the processing unit should have is provided. By executing this program on a computer, the above processing functions are implemented on the computer. The program describing the processing content can be recorded on a computer-readable recording medium (excluding carrier waves).

[0136] When distributing a program, it may be sold in the form of a portable recording medium such as a DVD (Digital Versatile Disc) or CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. Alternatively, the program can be stored in the storage device of a server computer and transferred from the server computer to other computers via a network.

[0137] A computer executing a program stores the program, for example, on a portable storage medium or transferred from a server computer, in its own memory. The computer then reads the program from its memory and executes the processing according to the program. Alternatively, the computer can directly read the program from the portable storage medium and execute the processing according to that program. Furthermore, the computer can sequentially execute the processing according to the program received each time it is transferred from a server computer.

[0138] The embodiments described above are preferred examples of the present invention. However, the invention is not limited thereto, and various modifications are possible without departing from the spirit of the invention. [Explanation of Symbols]

[0139] 10 servers 30. History Information Database 32 Preprocessed Information Database 34 Estimated Model Database 40. Cultivation and Storage History Acquisition Department 42 Pre-processing section 44 Estimation Model Generation Unit 46 Input Information Acquisition Unit 48 Estimation Model Acquisition Unit 50 Hardness transition estimation unit 52 Quality retention period estimation section 54 Output section 60 Administrator terminals 70 User terminals 150 Environmental control information estimation unit

Claims

1. A method for managing the quality of fruits and vegetables after harvest, Based on information about the varieties of cultivated produce, information about the cultivation environment during the cultivation period, and information about the changes in the hardness of the produce after harvest, an estimation model is generated for each variety that estimates the hardness of the produce according to the number of days after harvest based on the information about the cultivation environment during the cultivation period. Information on the variety of the first fruit or vegetable to be estimated, and information on the cultivation environment during the cultivation period of the first fruit or vegetable are obtained. By inputting information about the cultivation environment during the cultivation period of the first fruit or vegetable into the estimation model corresponding to the variety of the first fruit or vegetable, the hardness of the first fruit or vegetable is estimated according to the number of days after harvest. A method for quality control of fruits and vegetables, characterized in that the processing is performed by a computer.

2. The system identifies and outputs the number of days after harvest when the hardness of the first type of produce falls outside a predetermined range. The method for quality control of fruits and vegetables according to claim 1, characterized in that the processing is performed by the computer.

3. The first fruit and vegetable has an endocarp and an exocarp, In the above generation process, a first estimation model for estimating the hardness of the endocarp and a second estimation model for estimating the hardness of the exocarp are generated. In the estimation process described above, the hardness of the inner pericarp of the first fruit and vegetable according to the number of days since harvest is estimated using the first estimation model, and the hardness of the outer pericarp of the first fruit and vegetable according to the number of days since harvest is estimated using the second estimation model. In the output process described above, the earlier of the following two values ​​is identified and output: the number of days after harvest when the hardness of the inner peel falls outside a predetermined first range, and the number of days after harvest when the hardness of the outer peel falls outside a predetermined second range. The method for quality control of fruits and vegetables as described in feature 2.

4. In the above generation process, based on information about the variety of the cultivated produce, information about the cultivation environment during the cultivation period, information about the changes in the hardness of the produce after harvest, and information about the storage environment after harvest, an estimation model is generated for each variety that estimates the hardness of the produce according to the number of days after harvest and the storage conditions after harvest, based on the information about the cultivation environment during the cultivation period. In the estimation process described above, the hardness of the first produce is estimated according to the number of days after harvest by inputting information on the cultivation environment during the cultivation period of the first produce and the storage conditions after harvest of the first produce into the estimation model corresponding to the variety of the first produce. The method for quality control of fruits and vegetables as described in feature 1.

5. The method for quality control of fruits and vegetables according to claim 1, characterized in that the estimation model is a multivariate analysis model or a machine learning model.

6. The aforementioned produce is strawberries. The hardness of the aforementioned produce is the apparent modulus of elasticity of the strawberry fruit. The method for quality control of fruits and vegetables as described in feature 1.

7. The method for quality control of fresh produce according to any one of claims 1 to 6, characterized in that the information on the cultivation environment during the cultivation period includes at least one piece of information on the amount of sunlight during the cultivation period, temperature, and EC of the nutrient solution used for hydroponic cultivation.

8. A method for managing the quality of fruits and vegetables after harvest, Information on the cultivation environment during the cultivation period of the aforementioned fruits and vegetables is obtained, Using acquired information on the cultivation environment and an estimation model for estimating the hardness of the produce according to the number of days after harvest based on the cultivation environment information of the produce, the hardness of the cultivated produce according to the number of days after harvest is estimated. A method for quality control of fruits and vegetables, characterized in that the processing is performed by a computer.

9. A quality control program for fruits and vegetables that estimates the quality of fruits and vegetables after harvest, Based on information about the varieties of cultivated produce, information about the cultivation environment during the cultivation period, and information about the changes in the hardness of the produce after harvest, an estimation model is generated for each variety that estimates the hardness of the produce according to the number of days after harvest based on the information about the cultivation environment during the cultivation period. Information on the variety of the first fruit or vegetable to be estimated, and information on the cultivation environment during the cultivation period of the first fruit or vegetable are obtained. By inputting information about the cultivation environment during the cultivation period of the first fruit or vegetable into the estimation model corresponding to the variety of the first fruit or vegetable, the hardness of the first fruit or vegetable is estimated according to the number of days after harvest. A quality control program for fruits and vegetables, characterized by having a computer perform the processing.

10. A quality control program for fruits and vegetables that manages the quality of fruits and vegetables after harvest, Information on the cultivation environment during the cultivation period of the aforementioned fruits and vegetables is obtained, Using acquired information on the cultivation environment and an estimation model for estimating the hardness of the produce according to the number of days after harvest based on the cultivation environment information of the produce, the hardness of the cultivated produce according to the number of days after harvest is estimated. A quality control program for fruits and vegetables, characterized by having a computer perform the processing.

11. A method of agricultural support that provides cultivation conditions for fruits and vegetables, Based on information about the variety of the cultivated produce, information about the first and second cultivation environments during the cultivation period of the produce, and information about the changes in the hardness of the produce after harvest, an estimation model is generated for each variety to estimate the second cultivation environment information necessary to match the day when the hardness after harvest falls outside a predetermined range with a desired date. The information of the first type of produce, the information of the desired date, and the information of the first cultivation environment are obtained. By inputting the desired date information and the first cultivation environment information into the estimation model corresponding to the first variety of produce, the system estimates and outputs second cultivation environment information to match the date on which the post-harvest hardness of the first produce falls outside a predetermined range with the desired date. An agricultural support method characterized by having a computer perform the processing.

12. A method of agricultural support that provides cultivation conditions for fruits and vegetables, We will obtain information on the desired date and information on the first cultivation environment. Using the acquired information on the desired date and the first cultivation environment, and an estimation model that estimates second cultivation environment information to match the date on which the post-harvest hardness of the produce falls outside a predetermined range with the desired date, the system estimates and outputs second cultivation environment information to match the date on which the post-harvest hardness of the produce falls outside a predetermined range with the desired date. An agricultural support method characterized by having a computer perform the processing.

13. This is an agricultural support program that provides cultivation conditions for fruits and vegetables, Based on information about the variety of the cultivated produce, information about the first and second cultivation environments during the cultivation period of the produce, and information about the changes in the hardness of the produce after harvest, an estimation model is generated for each variety to estimate the second cultivation environment information necessary to match the day when the hardness after harvest falls outside a predetermined range with a desired date. The information of the first type of produce, the information of the desired date, and the information of the first cultivation environment are obtained. By inputting the desired date information and the first cultivation environment information into the estimation model corresponding to the first variety of produce, the system estimates and outputs second cultivation environment information to match the date on which the post-harvest hardness of the first produce falls outside a predetermined range with the desired date. An agricultural support program characterized by having a computer perform the processing.

14. This is an agricultural support program that provides cultivation conditions for fruits and vegetables, We will obtain information on the desired date and information on the first cultivation environment. Using the acquired information on the desired date and the first cultivation environment, and an estimation model that estimates second cultivation environment information to match the date on which the post-harvest hardness of the produce falls outside a predetermined range with the desired date, the system estimates and outputs second cultivation environment information to match the date on which the post-harvest hardness of the produce falls outside a predetermined range with the desired date. An agricultural support program characterized by having a computer perform the processing.

Images