Harvested crop evaluation method, harvested crop evaluation system, and harvested crop evaluation program

Abstract

To accurately evaluate the spoilage of harvested crops.SOLUTION: A harvested crop evaluation method includes the steps of: identifying each of index values indicative of the influence of each type of data on the spoilage of harvested crops on the basis of each of the plurality of data, the plurality of data being measurement data obtained by measuring harvested crops, environmental data on the environment in which the crops were cultivated, and historical data of operations or processing performed on the crops related to spoilage; calculating a first evaluation value indicative of the risk of spoilage occurrence in the harvested crops from the plurality of index values identified on the basis of each of the plurality of data; and evaluating the harvested crops using the first evaluation value.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to a method for evaluating harvested products, a system for evaluating harvested products, and a program for evaluating harvested products.

Background Art

[0002] There are harvested products such as onions and fruit trees, where internal rot cannot be detected from the outside even if the inside is rotten. If such internally rotten harvested products are shipped, there is a risk of claims from processors, general consumers, etc. against the producer, leading to a decline in the trust in the production area. Therefore, it is preferable to remove as many rotten harvested products as possible before shipment.

[0003] Conventionally, as a non-destructive detection technique for internal rot of onions, a detection technique for internal rot of onions by visible near-infrared spectroscopy is known (see, for example, Non-Patent Documents 1, 2, etc.). This non-destructive detection technique irradiates onions with electromagnetic waves in the visible to near-infrared region with wavelengths of 665 nm to 955 nm for spectrum measurement, and determines the level of rot based on the spectrum information.

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, since the technologies of Non-Patent Documents 1 and 2 detect whether the harvested product (onion) is internally spoiled at the time of spectrum measurement, it has not been evaluated whether there is a high possibility that spoilage will become apparent after spectrum measurement (during storage or after shipment). For this reason, there is a possibility that internally spoiled harvested products are mixed in at the stage of distribution to general consumers and processors.

[0006] An object of the present invention is to provide a harvested product evaluation method, a harvested product evaluation system, and a harvested product evaluation program capable of accurately evaluating the spoilage of harvested products.

Means for Solving the Problems

[0007] The first harvested product evaluation method of the present invention includes a step of specifying, for each of a plurality of data among measurement data obtained by measuring a harvested product of a crop, environmental data regarding the environment in which the crop was cultivated, and performance data of work or treatment related to spoilage performed on the crop, an index value indicating the influence that each of the plurality of data gives to the spoilage of the harvested product; a step of calculating a first evaluation value indicating the spoilage occurrence risk of the harvested product from the plurality of index values specified based on each of the plurality of data; and a step of evaluating the harvested product using the first evaluation value.

[0008] The second harvested product evaluation method of the present invention includes a step of calculating a first evaluation value indicating the spoilage occurrence risk of the harvested product based on at least one of measurement data obtained by measuring a harvested product of a crop, environmental data regarding the environment in which the crop was cultivated, and performance data of work or treatment related to spoilage performed on the crop; a step of calculating a second evaluation value indicating the likelihood of internal spoilage in the harvested product by inputting an image of the harvested product taken by a camera into a learning model; and a step of evaluating the harvested product using the first evaluation value and the second evaluation value.

Effects of the Invention

[0009] The method, system, and program for evaluating harvested products of the present invention have the effect of accurately evaluating the spoilage of harvested products.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, one embodiment will be described in detail with reference to FIGS. 1 to 12.

[0012] FIG. 1 is a diagram schematically showing the flow of shipment of harvested crops (onions in this embodiment). As shown in FIG. 1, producers harvest onions cultivated in the fields, dry them in the fields for a predetermined period, then pack them in containers and transport them to the warehouses of processing facilities such as JA. In the processing facility, after removing the soil adhering to the onions, checking for spoilage, damage, and shape, sorting them by size, and packing them in boxes, they are shipped to processors or general consumers. The harvested product evaluation system 100 of this embodiment is a system used in producers and processing facilities, and is a system for evaluating the internal spoilage of onions cultivated in the fields.

[0013] FIG. 2 is a diagram schematically showing the configuration of the harvested product evaluation system 100 according to this embodiment. As shown in FIG. 2, the harvested product evaluation system 100 includes a server 10, a producer terminal 60, an information processing device 70, and a camera 79. The server 10, the producer terminal 60, and the information processing device 70 are connected to a network 80 such as the Internet.

[0014] The server 10 is a device that acquires various data related to onions cultivated in the fields and the fields where the onions are cultivated (harvested) from the producer terminal 60, and the identification information of the fields where the onions are cultivated (referred to as field ID), and evaluates the spoilage risk of the onions cultivated in the fields. The evaluation of the spoilage risk is an evaluation of the susceptibility of the onions cultivated in the fields to spoilage. The server 10 displays the evaluation results on the producer terminal 60 or transmits them to the information processing device 70.

[0015] FIG. 3(a) shows an example of the hardware configuration of server 10. As shown in FIG. 3(a), server 10 includes a CPU (Central Processing Unit) 90, a ROM (Read Only Memory) 92, a RAM (Random Access Memory) 94, a storage (HDD (Hard Disk Drive) or SSD (Solid State Drive)) 96, a network interface 97, and a drive 99 for a portable storage medium, etc. Each component of the configuration of server 10 is connected to a bus 98. In server 10, the functions of each part shown in FIG. 4 are realized by the CPU 90 executing a program stored in the ROM 92 or the storage 96, or a program read by the drive 99 for a portable storage medium from the portable storage medium 91. Note that the functions of each part in FIG. 4 may be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). Details of each part in FIG. 4 will be described later.

[0016] Returning to FIG. 2, the producer terminal 60 is a terminal (such as a PC (Personal Computer) or a smartphone) that can be used by the producer of onions. The producer inputs various data regarding the onions cultivated in his own farm and the farm where the onions were cultivated, and the farm ID, into the producer terminal 60. The producer terminal 60 transmits the input various data and the farm ID to the server 10. Note that the producer terminal 60 may also acquire various data regarding the farm from an external device based on the information input by the producer. For example, when the producer inputs information regarding the location information of the farm and the cultivation period into the producer terminal 60, the producer terminal 60 may acquire weather data corresponding to the input information from the mesh agricultural weather data held by the Japan Meteorological Agency or the National Agriculture and Food Research Organization.

[0017] In addition, the producer terminal 60 also has a function of displaying the screen (the screen showing the evaluation result regarding the spoilage risk of onions cultivated in the field) sent from the server 10 and providing information to the producer.

[0018] FIG. 3(b) shows an example of the hardware configuration of the producer terminal 60. As shown in FIG. 3(b), the producer terminal 60 includes a CPU 190, a ROM 192, a RAM 194, a storage 196, a network interface 197, a display unit 193, an input unit 195, and a drive 199 for a portable storage medium, etc. Each of these constituent parts of the producer terminal 60 is connected to a bus 198. The display unit 193 is a liquid crystal display, an organic EL display, or the like, and the input unit 195 is a keyboard, a mouse, a touch panel, or the like. In the producer terminal 60, the functions of each part shown in FIG. 4 are realized by the CPU 190 executing a program stored in the ROM 192 or the storage 196, or a program read by the drive 199 for a portable storage medium from the portable storage medium 191.

[0019] Returning to FIG. 2, the information processing apparatus 70 is a terminal such as a PC installed in a processing facility. The information processing apparatus 70 acquires an image of an onion (see FIG. 7(a)) captured by the camera 79, and detects a confidence score regarding the prediction that the onion shown in the acquired image is internally spoiled. Further, the information processing apparatus 70 performs a comprehensive evaluation of the onion by combining the detected confidence score and the evaluation result regarding the spoilage risk transmitted from the server 10. The information processing apparatus 70 has the same hardware configuration as the producer terminal 60. That is, as shown in FIG. 3(b), the information processing apparatus 70 includes a CPU 290, a ROM 292, a RAM 294, a storage 296, a network interface 297, a display unit 293, an input unit 295, and a drive 299 for a portable storage medium, etc. Each of these constituent parts of the information processing apparatus 70 is connected to a bus 298. In the information processing apparatus 70, the functions of each part shown in FIG. 4 are realized by the CPU 290 executing a program stored in the ROM 292 or the storage 296, or a program read by the drive 299 for a portable storage medium from the portable storage medium 291.

[0020] The camera 79 is a camera capable of capturing RGB images. The camera 79 is installed, for example, on a belt conveyor that conveys onions in a processing facility, captures an image of the onions on the belt conveyor, and transmits the captured image to the information processing apparatus 70.

[0021] (Regarding the functions of each device) Hereinafter, the functions of the producer terminal 60, the server 10, and the information processing apparatus 70 will be described in detail based on FIG. 4.

[0022] (Regarding the functions of the producer terminal 60) As shown in FIG. 4, the producer terminal 60 includes a data acquisition unit 62, a data transmission unit 64, and a display processing unit 66.

[0023] The data acquisition unit 62 acquires various data related to onions cultivated in the field and the field ID input by the producer. In addition, the data acquisition unit 62 acquires various data related to onions cultivated in the field (such as meteorological data of the field) from an external device based on information about the field input by the producer (such as the location information of the field and information about the cultivation period). Details of the various data will be described later. The data acquisition unit 62 transmits the acquired various data and the field ID to the data transmission unit 64.

[0024] The data transmission unit 64 transmits the various data and the field ID acquired by the data acquisition unit 62 to the data reception unit 12 of the server 10.

[0025] The display processing unit 66 acquires the screen transmitted from the server 10 and displays it on the display unit 193 to provide information to the producer.

[0026] (Functions of the server 10) As shown in FIG. 4, the server 10 includes a data reception unit 12, a corruption risk calculation unit 14 as a specifying unit and a first calculation unit, a corruption risk transmission unit 16, and a screen generation unit 18.

[0027] The data reception unit 12 receives the various data and the field ID transmitted from the data transmission unit 64 of the producer terminal 60. The data reception unit 12 transmits the received various data and the field ID to the corruption risk calculation unit 14.

[0028] The corruption risk calculation unit 14 calculates the potential corruption risk (an index value indicating the ease of spoilage) of the onions cultivated in the field specified by the field ID using the various data transmitted from the data reception unit 12. Hereinafter, the method of calculating the potential corruption risk will be described in detail.

[0029] When calculating the potential corruption risk, the corruption risk calculation unit 14 determines the corruption risk values (index values indicating the influence of each element on the corruption of onions) for each of the three elements (plant body, environment, and ease of fungal invasion). Fig. 5 is a table summarizing the method for determining the corruption risk values for each element.

[0030] The corruption risk value for the element "plant body" means the corruption risk value that can be determined from the data (measurement data) obtained by actually measuring the onions harvested in the field. Also, the corruption risk value for the element "environment" means the corruption risk value that can be determined from the data (environment data) regarding the environment in which the onions were cultivated. Further, the corruption risk value for the element "ease of fungal invasion" means the corruption risk value that can be determined from the data (performance data of operations and treatments related to corruption) regarding the control and chemicals used during the cultivation period. Hereinafter, the specific method for determining each corruption risk value will be described.

[0031] (Regarding the corruption risk value for the element "plant body") (1) Assume that the producer measures the fresh weight and dry weight of one onion harvested in the field, calculates the moisture content from the measured values using Equation (2) described later, and inputs the calculated moisture content as the representative value of the moisture content of the onions harvested in the field. In this case, the corruption risk calculation unit 14 determines that if the moisture content is "~90%", the corruption risk value = 1; if the moisture content is "90 - 93%", the corruption risk value = 2; and if the moisture content is "93% ~", the corruption risk value = 3. Note that the upper limit of "~" can be either "above" or "greater than", and the lower limit of "~" can be either "below" or "less than", as long as the ranges corresponding to each corruption risk value do not overlap (the same applies hereinafter).

[0032] Note that the producer may also obtain the moisture content of multiple onions harvested in the field and input the average value of the obtained moisture contents as the moisture content of the onions harvested in the field. Also, the moisture content may be measured using a sensor capable of measuring the moisture amount.

[0033] (2) Suppose the producer measures the nitrogen content of one onion harvested from the field by any of the ninhydrin test, measurement by the Kjeldahl method, or measurement by a CN coder, and inputs the measured nitrogen content as a representative value of the nitrogen content of the onions harvested from the said field. In this case, if the nitrogen content is "~13 mgN / g dry matter", the spoilage risk calculation unit 14 determines that the spoilage risk value = 1; if the nitrogen content is "13 - 18 mgN / g dry matter", the spoilage risk value = 2; if the nitrogen content is "18 mgN / g dry matter ~", the spoilage risk value = 3. Note that the producer may input the average value of the nitrogen contents of a plurality of onions harvested from the field as the nitrogen content of the onions harvested from the field.

[0034] (3) Suppose the producer measures the sugar content of one onion harvested from the field by using a saccharimeter or by separating it by HPLC (high performance liquid chromatography) and then analyzing it with a detector, and inputs the measured sugar content as a representative value of the sugar content of the onions harvested from the said field. In this case, if the sugar content is "~5.5%", the spoilage risk calculation unit 14 determines that the spoilage risk value = 1; if the sugar content is "5.5 - 6.0%", the spoilage risk value = 2; if the sugar content is "6.0% ~", the spoilage risk value = 3. Note that the producer may input the average value of the sugar contents of a plurality of onions harvested from the field as the sugar content of the onions harvested from the field.

[0035] (4) Suppose the producer measures the relative intensity of calcium ions in one onion harvested from the field by X-ray irradiation and inputs it as a representative value of the said field. In this case, if the relative intensity is "60 ~", the spoilage risk calculation unit 14 determines that the spoilage risk value = 1; if the relative intensity is "40 - 60", the spoilage risk value = 2; if the relative intensity is "~40", the spoilage risk value = 3. Note that the producer may input the average value of the relative intensities of calcium ions in a plurality of onions harvested from the field as the relative intensity of calcium ions of the onions harvested from the field.

[0036] In addition, when a plurality of data among (1) to (4) are input by the producer, the spoilage risk calculation unit 14 may treat the average value, maximum value, median value, etc. of the spoilage risk values obtained from each of the input data as the spoilage risk value regarding the element "plant body".

[0037] (Regarding the spoilage risk value regarding the element "environment") (1) Assume that the producer inputs the location information of the field and the information that can identify the enlargement period of the onions cultivated in the field (for example, information indicating when the enlargement period is from when to when, information indicating the harvest time, etc.) into the producer terminal 60, and the data acquisition unit 62 acquires the precipitation amount during the enlargement period at the location of the field from the mesh agricultural meteorological data. In addition, when the producer inputs the harvest date of the onions, the data acquisition unit 62 identifies the predetermined period (for example, 3 weeks) before the harvest date as the enlargement period and acquires the precipitation amount. Alternatively, assume that the data acquisition unit 62 acquires the irrigation amount during the enlargement period in the field input by the producer into the producer terminal 60. In this case, the spoilage risk calculation unit 14 determines that if the precipitation amount (or irrigation amount) is "~12.7 mm / week", the spoilage risk value = 1; if the precipitation amount (or irrigation amount) is "12.7~25.4 mm / week", the spoilage risk value = 2; if the precipitation amount (or irrigation amount) is "25.4 mm / week~", the spoilage risk value = 3.

[0038] (2) Assume that the producer inputs the location information of the field and the information that can identify the enlargement period of the onions cultivated in the field into the producer terminal 60, and the data acquisition unit 62 acquires the temperature during the enlargement period at the location of the field from the mesh agricultural meteorological data. In this case, the spoilage risk calculation unit 14 determines that if the number of days with a maximum temperature of 30 degrees or more is "30% or less", the spoilage risk value = 1; if the number of days with a maximum temperature of 30 degrees or more is "3~60%", the spoilage risk value = 2; if the number of days with a maximum temperature of 30 degrees or more is "6~100%", the spoilage risk value = 3.

[0039] (3) When the producer inputs the nitrogen application rate in the field into the producer terminal 60, the spoilage risk calculation unit 14 determines that if the nitrogen application rate is "~16 kg / 10a", the spoilage risk value = 1; if the nitrogen application rate is "16~25 kg / 10a", the spoilage risk value = 2; and if the nitrogen application rate is "25 kg / 10a~", the spoilage risk value = 3.

[0040] (4) Suppose the producer observes the onion seedlings cultivated in the field and inputs the ratio of the formation of strip grooves along the leaf veins (the ratio of abnormal leaf shape) into the producer terminal 60. In this case, the spoilage risk calculation unit 14 determines that if the ratio of abnormal leaf shape is "~10%", the spoilage risk value = 1; if the ratio of abnormal leaf shape is "10~15%", the spoilage risk value = 2; and if the ratio of abnormal leaf shape is "15%~", the spoilage risk value = 3.

[0041] (5) When the producer inputs the presence or absence of previous cropping history in the field into the producer terminal 60, the spoilage risk calculation unit 14 determines that if the previous cropping history is "less than 3 years of continuous cropping and no disease has occurred", the spoilage risk value = 1; if the previous cropping history is "no disease has occurred but continuous cropping for 3 years or more", the spoilage risk value = 2; and if the previous cropping history is "disease has occurred regardless of continuous cropping", the spoilage risk value = 3.

[0042] In addition, when multiple pieces of data among (1) to (5) are input by the producer, the spoilage risk calculation unit 14 may handle the average value, maximum value, median value, etc. of the spoilage risk values obtained from each of the input data as the spoilage risk value regarding the element "environment".

[0043] (Regarding the spoilage risk value related to the element "ease of invasion of bacteria") (1) When the producer inputs the control interval of thrips in the field into the producer terminal 60, the spoilage risk calculation unit 14 determines that if the control interval is "10-day to 15-day interval", the spoilage risk value = 1; if the control interval is "7- to 10-day interval or 15- to 30-day interval", the spoilage risk value = 3; and in other cases, the spoilage risk value = 2.

[0044] In onion cultivation, when thrips are controlled during cultivation, the occurrence of rotten bulbs decreases. This is presumably because the feeding damage marks, which serve as entry points for bacteria, are reduced, preventing the infection of rot disease and making it difficult for rotten bulbs to occur. Therefore, when thrips are controlled at appropriate control intervals, the above-described method for determining the rot risk value is adopted from the perspective that the rot risk value becomes low.

[0045] (2) When the producer inputs information on the chemicals used in controlling thrips in the field (information on the usage ratio of low-control chemicals) into the producer terminal 60, the rot risk calculation unit 14 determines that if the ratio is "~15%", the rot risk value = 1; if the ratio is "15~70%", the rot risk value = 2; and if the ratio is "70%~", the rot risk value = 3.

[0046] In addition, when a plurality of data among (1) and (2) are input by the producer, the average value, maximum value, median value, etc. of the rot risk values obtained from each of the input data may be treated as the rot risk value regarding the element "ease of bacterial invasion".

[0047] Note that the method for determining the rot risk value (the range corresponding to each rot risk value) shown in FIG. 5 is an example. The range (judgment criterion) corresponding to each rot risk value may be appropriately changed according to the cultivation area, variety, etc. Also, the items shown in FIG. 5 are examples. Therefore, it is also possible to determine the rot risk value using items other than those shown in FIG. 5.

[0048] The spoilage risk calculation unit 14 determines the spoilage risk values for each of the three elements (plant body, environment, and ease of fungal invasion) as described above. Then, based on the determined spoilage risk values, it calculates the total spoilage risk (hereinafter referred to as "potential spoilage risk") of onions harvested in the field corresponding to the field ID. For example, if the spoilage risk value for the element "plant body" is a, the spoilage risk value for the element "environment" is b, and the spoilage risk value for the element "ease of fungal invasion" is c, and the weight of the spoilage risk value for the element "plant body" is α, the weight of the spoilage risk value for the element "environment" is β, and the weight of the spoilage risk value for the element "ease of fungal invasion" is γ, the spoilage risk calculation unit 14 can calculate the potential spoilage risk from the following formula (1). Potential spoilage risk = α × a + β × b + γ × c …(1)

[0049] For example, assume that the weight α = 5, the weight β = 2, the weight γ = 1, the spoilage risk value a for the element "plant body" = 2, the spoilage risk value b for the element "environment" = 2, and the spoilage risk value c for the element "ease of fungal invasion" = 1. In this case, the potential spoilage risk is Potential spoilage risk = 5 × 2 + 2 × 2 + 1 × 1 = 15 and can be obtained. This potential spoilage risk is the first evaluation value indicating the risk of spoilage occurring in the onions harvested in the field.

[0050] Note that since the spoilage risk value for the element "plant body" is a value determined based on measured data, it is considered good to make the weight α larger than the weights β and γ. Also, empirically, it is considered that the element "environment" has a greater contribution to the potential spoilage risk than the element "ease of fungal invasion", so it is considered good to make the weight β larger than the weight γ. However, it is not limited to this, and the values of the weights α, β, and γ may be determined based on various findings.

[0051] Note that the calculation formula for obtaining the potential spoilage risk is not limited to the above formula (1), and other calculation formulas may also be used.

[0052] The spoilage risk calculation unit 14 associates the calculated value of the potential spoilage risk with the field ID and transmits it to the spoilage risk transmission unit 16 and the screen generation unit 18.

[0053] The spoilage risk transmission unit 16 associates the value of the potential spoilage risk calculated by the spoilage risk calculation unit 14 with the field ID and transmits it to the spoilage risk reception unit 73 of the information processing device 70.

[0054] The screen generation unit 18 generates a screen for displaying the spoilage risk value and the value of the potential spoilage risk calculated by the spoilage risk calculation unit 14, and transmits it to the display processing unit 66 of the producer terminal 60. For example, if the screen generation unit 18 is the above example (the example where the spoilage risk values are a = 2, b = 2, c = 1, and the potential spoilage risk = 15), it generates a screen as shown in FIG. 6 (a radar chart showing the spoilage risk value and a screen for displaying the value of the potential spoilage risk) and transmits it to the producer terminal 60. Thereby, the producer can utilize the information displayed on the screen for improving the cultivation of onions in subsequent times. That is, the producer can perform cultivation management so as to reduce the potential spoilage risk. Note that the screen to be displayed on the display unit 193 is not limited to the screen of FIG. 6, and may be a screen for displaying the spoilage risk value and the value of the potential spoilage risk in other formats.

[0055] (Regarding the functions of the information processing device 70) As shown in FIG. 4, the information processing device 70 includes a field ID acquisition unit 76, an image acquisition unit 71, an internal spoilage detection unit 72 as a second calculation unit, a spoilage risk reception unit 73, a comprehensive evaluation unit 74 as an evaluation unit, and a display processing unit 75.

[0056] The field ID acquisition unit 76 acquires identification information (field ID) of the field where the onions being conveyed by the belt conveyor are harvested. In the present embodiment, the field ID is input by the operator of the processing facility via the input unit 295 of the information processing device 70. However, the present invention is not limited to this, and a reading device connected to the information processing device 70 may automatically read the field ID attached to the container that houses the onions conveyed by the belt conveyor. The field ID acquisition unit 76 transmits the acquired field ID to the internal spoilage detection unit 72.

[0057] The image acquisition unit 71 acquires the image of the onion captured by the camera 79 and transmits it to the internal spoilage detection unit 72. The camera 79 is disposed above the belt conveyor that conveys the onion in the processing facility as described above, and captures the onion passing through a predetermined position (within the angle of view of the camera 79). FIG. 7(a) shows an example of the image of the onion captured by the camera 79.

[0058] The internal spoilage detection unit 72 inputs (feeds) the acquired image of the onion into the machine learning model, and detects the confidence score regarding the prediction that the onion shown in the image is internally spoiled. Here, the internal spoilage detection unit 72 can use, for example, a machine learning model called VGG16. There is considered to be a correlation between the shape of the onion (especially the shape of the neck of the onion) and internal spoilage. Therefore, a machine learning model is generated (learned) using the image of the onion and the information on the presence or absence of internal spoilage in the onion as correct data, and by inputting the image of the onion before shipment (such as FIG. 7(a)) into the machine learning model, the confidence score (a value within the range of 0 to 100%) can be detected. Here, in the present embodiment, it is assumed that the higher the confidence score, the higher the possibility that internal spoilage as shown in FIG. 7(b) has occurred, and the lower the confidence score, the higher the possibility that no internal spoilage has occurred. That is, it can be said that the confidence score is a second evaluation value indicating the likelihood of occurrence of internal spoilage in the onion.

[0059] Note that the internal spoilage detection unit 72 associates the detected confidence score with the field ID and transmits it to the comprehensive evaluation unit 74.

[0060] The spoilage risk reception unit 73 receives the value of the potential spoilage risk and the field ID transmitted from the spoilage risk transmission unit 16 of the server 10, and transmits them to the comprehensive evaluation unit 74.

[0061] The comprehensive evaluation unit 74 performs a comprehensive evaluation of the photographed onions based on the information (confidence level) received from the internal corruption detection unit 72 and the value of the potential corruption risk corresponding to the farm ID received from the corruption risk reception unit 73.

[0062] Figure 8 is a diagram showing the comprehensive evaluation matrix used for comprehensive evaluation. In the comprehensive evaluation matrix of Figure 8, the confidence level (0 to 100%) is divided into ranges of 10, and to each range, values of 0.5, 1, 2, 3, 4, 5, 6, 7 and identifiers of H (sound) and D (corrupt) are assigned. Then, the product of the value indicating the range of the confidence level and the value of the potential corruption risk (the product of the values in the rows and columns of Figure 8) is stored in the column where the row and column intersect. In addition, "D" is stored in the rows where the range of the confidence level is D, and "H" is stored in the rows where the range of the confidence level is H.

[0063] The comprehensive evaluation unit 74 obtains a comprehensive evaluation value by fitting the received confidence level and the value of the potential corruption risk to the comprehensive evaluation matrix of Figure 8. For example, if the confidence level is "95%", regardless of the value of the potential corruption risk, the comprehensive evaluation value will be "D". Also, if the confidence level is "0%", regardless of the value of the potential corruption risk, the comprehensive evaluation value will be "H". Further, if the confidence level is "56%" and the value of the potential corruption risk is "18", the comprehensive evaluation value will be "90". In this embodiment, the higher the confidence level and the higher the value of the potential corruption risk, the larger the comprehensive evaluation value; the lower the confidence level and the lower the value of the potential corruption risk, the smaller the comprehensive evaluation value. Also, even if the confidence levels are the same, the higher the value of the potential corruption risk, the larger the comprehensive evaluation value.

[0064] The comprehensive evaluation unit 74 obtains the "judgment result" and "countermeasure" corresponding to the comprehensive evaluation value by applying the comprehensive evaluation value obtained using the comprehensive evaluation matrix in FIG. 8 to the comprehensive evaluation table in FIG. 9. For example, when the comprehensive evaluation value is D or is between 133 and 168, the "spoiled ball" is obtained as the judgment result, and "discard" is obtained as the countermeasure. Also, for example, when the comprehensive evaluation value is between 90 and 132, the "high-risk ball" is obtained as the judgment result, and "immediate shipment" is obtained as the countermeasure. Also, for example, when the comprehensive evaluation value is between 46 and 89, the "medium-risk ball" is obtained as the judgment result, and "refrigerated storage (short-term storage)" is obtained as the countermeasure. Also, for example, when the comprehensive evaluation value is between 9 and 45, the "low-risk ball" is obtained as the judgment result, and "long-term storage with countermeasures (gas chamber storage)" is obtained as the countermeasure. Also, for example, when the comprehensive evaluation value is H or is between 4 and 9, the "healthy ball" is obtained as the judgment result, and "long-term storage" is obtained as the countermeasure. When stored long-term, it can be shipped at the end-of-life period, and high-unit-price transactions can be expected.

[0065] The comprehensive evaluation unit 74 transmits the judgment result and countermeasure information obtained from the comprehensive evaluation table in FIG. 9 to the display processing unit 75.

[0066] Returning to FIG. 4, the display processing unit 75 displays the information (judgment result and countermeasure) received from the comprehensive evaluation unit 74 on the display unit 93 of the information processing device 70. In this case, the display processing unit 75 displays a screen as shown in FIGS. 10(a) and 10(b), for example. For example, the display processing unit 75 performs the display as shown in FIGS. 10(a) and 10(b) at the timing when the onion located in front of the worker performing the sorting operation near the belt conveyor is in position, and provides information to the worker, so that the worker can perform the sorting operation based on the provided information.

[0067] (Regarding the processing flow) Next, regarding the processing flow of the server 10 and the information processing device 70, it will be described with reference to the flowcharts in FIGS. 11 and 12 and other drawings as appropriate.

[0068] (Processing of the server 10) In server 10, the processing along the flowchart of FIG. 11 is executed. When the processing of FIG. 11 starts, first, in step S10, the data reception unit 12 waits until various data are transmitted from the producer terminal 60 (data transmission unit 64) together with the farm ID.

[0069] When the farm ID and various data are transmitted from the producer terminal 60, the process proceeds to step S12, and the data reception unit 12 acquires (receives) the farm ID and various data.

[0070] Next, in step S14, the spoilage risk calculation unit 14 calculates the spoilage risk using the various data. That is, as described above, the spoilage risk calculation unit 14 obtains the spoilage risk values for each of the three elements (plant body, environment, and ease of fungal invasion) by the method shown in FIG. 4, and substitutes the obtained spoilage risk values into the above formula (1) to calculate the value of the potential spoilage risk in the farm corresponding to the farm ID. The spoilage risk calculation unit 14 transmits the calculated value of the potential spoilage risk in the farm together with the farm ID to the spoilage risk transmission unit 16.

[0071] Next, in step S16, the spoilage risk transmission unit 16 transmits the value of the potential spoilage risk in the farm in association with the farm ID to the information processing device 70 (spoilage risk reception unit 73).

[0072] Next, in step S18, the screen generation unit 18 generates a screen (see FIG. 6) for displaying the spoilage risk values and the value of the potential spoilage risk for each element calculated by the spoilage risk calculation unit 14, and transmits it to the producer terminal 60 (display processing unit 66). In the producer terminal 60, the display processing unit 66 displays the screen of FIG. 6 on the display unit 193.

[0073] Thereafter, the process returns to step S10, and the above-described processing is repeatedly executed.

[0074] (Processing of the information processing device 70) In the information processing apparatus 70, the processes according to the flowchart of FIG. 12 are executed. When the process of FIG. 12 starts, first, in step S50, the farm ID acquisition unit 76 waits until a farm ID is transmitted. For example, when an operator of a processing facility conveys onions harvested in a certain farm via a belt conveyor, the operator inputs the farm ID via the input unit 295 of the information processing apparatus 70. In this way, when the farm ID is input, the farm ID acquisition unit 76 proceeds to step S52.

[0075] When proceeding to step S52, the farm ID acquisition unit 76 acquires the farm ID. The farm ID acquisition unit 76 transmits the acquired farm ID to the internal spoilage detection unit 72.

[0076] Next, in step S54, the image acquisition unit 71 waits until an image of an onion on the belt conveyor is transmitted from the camera 79. When the image is transmitted, the image acquisition unit 71 proceeds to step S56 and acquires the transmitted image. The image acquisition unit 71 transmits the acquired image to the internal spoilage detection unit 72.

[0077] Next, in step S58, the internal spoilage detection unit 72 detects a confidence level from the captured image using a machine learning model. The internal spoilage detection unit 72 associates the detected confidence level with the farm ID and transmits the result to the comprehensive evaluation unit 74.

[0078] Next, in step S60, the comprehensive evaluation unit 74 comprehensively evaluates the onions using the detected confidence level and the value of the potential spoilage risk in the farm corresponding to the farm ID. That is, the comprehensive evaluation unit 74 obtains a comprehensive evaluation value by applying the confidence level received from the internal spoilage detection unit 72 and the value of the potential spoilage risk corresponding to the farm ID received from the internal spoilage detection unit 72 to the comprehensive evaluation matrix of FIG. 8. In addition, the comprehensive evaluation unit 74 refers to the comprehensive evaluation table of FIG. 9 and acquires the "judgment result" and "countermeasure" corresponding to the obtained comprehensive evaluation value. The comprehensive evaluation unit 74 transmits the acquired "judgment result" and "countermeasure" to the display processing unit 75.

[0079] Next, in step S62, the display processing unit 75 generates a screen (see FIGS. 10(a) and 10(b)) for displaying the comprehensive evaluation result, and displays it on the display unit 293 of the information processing apparatus 70.

[0080] Next, in step S64, the farm ID acquisition unit 76 determines whether there is a change in the farm ID. If the determination in this step S64 is negative, the process returns to step S54, and the processes after step S54 are executed again. On the other hand, if the determination in step S64 is affirmative, the process returns to step S52, and the processes after step S52 are executed again.

[0081] (Example) Before explaining the example, the comparative example will be explained. In this comparative example, the presence or absence of internal rot was detected by inputting images of 17 rotten bulbs and 2 healthy bulbs (19 in total) of onions into the machine learning model. The detection results of this comparative example are shown in FIG. 13. In this comparative example, the image of each onion was input into the machine learning model to obtain a confidence level. If the confidence level was 50% or more, it was considered "D: rotten bulb", and if the confidence level was less than 50%, it was considered "H: healthy bulb". As a result, as shown in FIG. 13, the onions with No. = 4, 9, 10, 15, 16, 17, 19 were detected as healthy bulbs, and the other onions were detected as rotten bulbs.

[0082] Regarding this result, when compared with the actual presence or absence of internal rot of the onions, as shown in gray in FIG. 13, different results from the actual presence or absence of internal rot were detected in 5 onions. That is, the misdetection rate in this comparative example was 5 / 19 ≒ 26%.

[0083] Next, the example will be explained. In this example, for each of the 19 onions also used in the comparative example, the processes of the above-described embodiment (the processes by the server 10 and the information processing apparatus 70) were carried out.

[0084] Figure 14 is a table for explaining this embodiment. In this embodiment, first, an image evaluation value (see Figure 8) was obtained from the confidence level of the image of each onion taken.

[0085] Also, in this embodiment, for each onion, the risk values of spoilage for each of the three elements (plant body, environment, and susceptibility to fungal invasion) were determined by the following method.

[0086] (a) Spoilage risk value for the element "plant body" For each onion, the fresh weight and dry weight were measured, and the moisture content was obtained from the following formula (2). Moisture content = {(fresh weight - dry weight) / fresh weight} × 100 …(2) Then, based on the criteria in the row of the item "moisture content" in Figure 5, the spoilage risk value for the element "plant body" was determined.

[0087] (b) Spoilage risk value for the element "environment" The total precipitation during the bulb enlargement period (three weeks before the harvest date) of each onion was obtained from the mesh agricultural meteorological data, and based on the criteria in the row of the item "precipitation during the bulb enlargement period" in Figure 5, the spoilage risk value for the element "environment" was determined.

[0088] (c) Spoilage risk value for the element "susceptibility to fungal invasion" It was confirmed how often the target pesticide for thrips was sprayed when each onion was cultivated, and based on the criteria in the row of the item "control interval for thrips" in Figure 5, the spoilage risk value for the element "susceptibility to fungal invasion" was determined.

[0089] In the column of "spoilage risk value" in Figure 14, the spoilage risk values for the three elements in each onion are shown.

[0090] Also, in this embodiment, using the above formula (1), the value of the potential spoilage risk in the field where each onion was harvested was determined. Regarding the weights in the above formula (1), α = 5, β = 2, and γ = 1. In the "Potential Spoilage Risk" column of FIG. 14, the values of the potential spoilage risk of each onion are shown. Further, for each onion, the product of the image evaluation value and the potential spoilage risk (comprehensive evaluation value) was determined. In the column of "Image Evaluation Value × Potential Spoilage Risk" in FIG. 14, the product (comprehensive evaluation value) of the image evaluation value and the potential spoilage risk of each onion is shown.

[0091] And in this embodiment, the comprehensive evaluation was performed by applying the comprehensive evaluation value of each onion to the comprehensive evaluation table in FIG. 9. The result is shown in the "Comprehensive Evaluation" column of FIG. 14. In this embodiment, among the 19 onions, 3 onions (spoiled balls) were evaluated as healthy. The misevaluation rate in this embodiment (the rate where the comprehensive evaluation is D while the actual presence or absence of spoilage is H, or the comprehensive evaluation is H while the actual presence or absence of spoilage is D) is 3 / 19 ≈ 16%, which can be seen to be improved compared to the comparative example. In the comparative example, as in No. 17 and No. 19 in FIG. 13, onions (spoiled balls) with actual internal spoilage were misdetected as healthy, but in this embodiment, as shown by arrow A in FIG. 14, they are evaluated as low-risk balls or medium-risk balls.

[0092] Also, in this embodiment, as shown by arrow B in FIG. 14, the onions that were detected as healthy in the comparative example are evaluated as low-risk balls. In this way, by comprehensively evaluating the onions, the handling (shipping time and storage method) of onions with spoilage risk can be made different from that of healthy ones.

[0093] As described in detail above, according to this embodiment, the server 10 determines a corruption risk value for the element "plant body" based on the measurement data obtained by measuring the onions, determines a corruption risk value for the element "environment" based on the environmental data regarding the environment in which the onions are cultivated, and determines a corruption risk value for the element "ease of fungal invasion" based on the performance data of the work or process related to corruption performed on the onions. Further, the server 10 calculates a value of potential corruption risk (first evaluation value) in the field based on each corruption risk value. Further, the information processing device 70 detects a confidence level (second evaluation value) by inputting the image of the onions captured by the camera 79 into the machine learning model, and performs a comprehensive evaluation of the onions based on the confidence level and the value of the potential corruption risk. Thus, in this embodiment, in order to evaluate the onions in consideration of the value of the potential corruption risk and the confidence level detected from the image, a more detailed and accurate evaluation can be performed compared to the case of evaluating the onions only from the image. For example, even if it is possible to detect whether internal corruption has occurred in the onions only from the image, it is not possible to evaluate whether corruption is likely to occur in the future. In contrast, in this embodiment, by using the value of the potential corruption risk, it is possible to evaluate whether corruption is likely to occur in the future for onions in which internal corruption has not occurred. Further, in this embodiment, based on the evaluation result, the countermeasures (whether to ship, shipping time, storage method) for each onion are determined, so that the onions can be shipped so as not to receive claims from general consumers or the like. Further, in this embodiment, since the value of the potential corruption risk in the field where the onions are harvested is calculated based on the corruption risk values for each of the plurality of elements, the potential corruption risk can be calculated based on various elements.

[0094] Further, in this embodiment, the corruption risk calculation unit 14 calculates the potential corruption risk using the product of the weights (α, β, γ) set for each corruption risk value for a plurality of elements and the corruption risk values for the plurality of elements, as in the above formula (1). Thereby, the potential corruption risk can be calculated in consideration of the importance of each corruption risk value for each element and the degree of influence (contribution degree) that each element gives to the value of the potential corruption risk.

[0095] Also, in this embodiment, the camera 79 is a camera capable of taking RGB images. As a result, there is an advantage that expensive equipment such as Non-Patent Documents 1 and 2 does not need to be prepared. Further, in the case of Cited Documents 1 and 2, it is necessary to fix the onion in a predetermined posture, but in this embodiment, since it is sufficient to obtain an image in which the entire onion is shown, there is no need to fix the onion, and the onion being conveyed by the belt conveyor may be photographed using the camera 79.

[0096] Also, in this embodiment, based on the comprehensive evaluation table in FIG. 9, the comprehensive evaluation unit 74 specifies the shipment availability of the onion, the shipment time, and the storage method until shipment from the comprehensive evaluation value, and the display processing unit 75 displays the information specified by the comprehensive evaluation unit 74 on the display unit 293 of the information processing apparatus 70. As a result, the operator can know how to handle each onion.

[0097] Also, in this embodiment, the value of the risk of spoilage for each of the plurality of elements and the value of the potential risk of spoilage calculated from these are to be displayed on the display unit 193 of the producer terminal 60. Therefore, the producer can confirm what points need to be improved in the next cultivation.

[0098] Note that, in the above embodiment, the case of calculating the value of the potential risk of spoilage using the value of the risk of spoilage for each of the three elements has been described, but the present invention is not limited to this. For example, the value of the potential risk of spoilage may be calculated from the value of the risk of spoilage for two or one of the three elements. Further, the value of the potential risk of spoilage may be calculated using the value of the risk of spoilage for an element other than the above three elements (plant body, environment, ease of invasion of bacteria).

[0099] Incidentally, in the above-described embodiment, the risk value of spoilage regarding the element "plant body" may be determined from the weight (bulb weight) of each harvested onion. For example, if the weight of the onion is "~15 g", the spoilage risk value = 1; if it is "15 - 21 g", the spoilage risk value = 2; and if it is "21 g ~", the spoilage risk value = 3 can be determined. The weight of each onion may be automatically measured, for example, on a belt conveyor. In this case, in the spoilage risk calculation unit 14 of the server 10, the potential spoilage risk may not be calculated, and only the spoilage risk value regarding the element "environment" and the spoilage risk value regarding the element "ease of fungal invasion" may be calculated. In the information processing device 70, the spoilage risk value regarding the element "plant body" may be determined from the weight of the onion. In this case, the information processing device 70 (for example, the comprehensive evaluation unit 74) may calculate the potential spoilage risk using the spoilage risk value regarding the element "plant body", the spoilage risk value regarding the element "environment" determined by the server 10, and the spoilage risk value regarding the element "ease of fungal invasion". When the potential spoilage risk is calculated for each onion in this way, even if the fields are the same, different potential spoilage risk values will be used for each onion. Thereby, a comprehensive evaluation can be accurately performed for each onion.

[0100] Incidentally, in the case where the moisture content of each onion can be measured on the belt conveyor, similar to the case of using the above-described weight (bulb weight), the spoilage risk value regarding the element "plant body" may be determined from the moisture content of each onion.

[0101] Incidentally, in the above-described embodiment, the case where the harvested product evaluation system 100 includes the server 10 has been described, but it is not limited thereto, and the server 10 may be omitted. In this case, the information processing device 70 may execute the processing of the server 10 described in the above-described embodiment. Conversely, the information processing device 70 may be omitted in the harvested product evaluation system 100. In this case, the server 10 may execute the processing of the information processing device 70 described in the above-described embodiment.

[0102] In the above-described embodiment, the case where the confidence level is detected using the RGB image captured by the camera 79 has been described. However, the present invention is not limited to this, and the confidence level may be detected using an image obtained by the same method as in Non-Patent Documents 1 and 2.

[0103] In the above-described embodiment, the case where the evaluation result (judgment result and countermeasure) of the comprehensive evaluation unit 74 is displayed on the display unit 293 has been described. However, the present invention is not limited to this. For example, based on the evaluation result of the comprehensive evaluation unit 74, a tomato sorting system (a system that automatically changes the conveyance destination of tomatoes) connected to the belt conveyor may be controlled. Thereby, it becomes possible to automatically sort tomatoes based on the evaluation result.

[0104] In the above-described embodiment, the case where the comprehensive evaluation unit 74 obtains the comprehensive evaluation value from the value of the potential corruption risk and the confidence level using the comprehensive evaluation matrix of FIG. 8 has been described. However, the present invention is not limited to this. That is, the comprehensive evaluation unit 74 may obtain the comprehensive evaluation value from the value of the potential corruption risk and the confidence level using other methods (calculation formulas and calculation criteria) for obtaining the comprehensive evaluation value.

[0105] In the above-described embodiment, the case where the comprehensive evaluation unit 74 obtains the comprehensive evaluation value from the value of the potential corruption risk and the confidence level has been described. However, the present invention is not limited to this. For example, the comprehensive evaluation unit 74 may use, as the comprehensive evaluation value, a value obtained from the value of the potential corruption risk without using the value of the potential corruption risk itself or the confidence level, and perform a comprehensive evaluation using a comprehensive evaluation table similar to FIG. 9 based on this comprehensive evaluation value. In this case, the camera 79, the image acquisition unit 71, the internal corruption detection unit 72, the field ID acquisition unit 76, etc. can be omitted.

[0106] In the above-described embodiment, the case where the harvested product is an onion has been described. However, the harvested product is not limited to this, and it may be a harvested product that is prone to internal rot (for example, a melon, etc.). In this case, regarding the method (Figure 5) for obtaining the decay risk value for each of the above three elements (plant body, environment, and ease of fungal invasion), it shall be appropriately changed according to the harvested product. That is, the "items", "measurement methods", and reference values for determining the "decay risk value" in Figure 5 may be appropriately changed according to the harvested product. For example, the item "aphid control interval" and the item "chemicals used for aphid control" in Figure 5 may be changed to the control intervals for pests and diseases other than aphids, and the chemicals used for controlling pests and diseases other than aphids. Also, according to the harvested product, the method for obtaining the value of the potential decay risk (Equation (1)), the method for obtaining the comprehensive evaluation value (Figure 8), and the content of the comprehensive evaluation table (Figure 9) shall also be appropriately changed.

[0107] Note that the above processing functions can be realized by a computer. In that case, a program describing the processing content of the functions that the processing device should have is provided. By executing that program on a computer, the above processing functions are realized on the computer. The program describing the processing content can be recorded on a computer-readable storage medium (excluding carrier waves).

[0108] When distributing the program, for example, it is sold in the form of a portable storage medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. Also, 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.

[0109] A computer that executes a program stores, for example, a program recorded on a portable storage medium or a program transferred from a server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. Note that the computer can also directly read the program from the portable storage medium and execute processing according to the program. Further, each time a program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

[0110] The above-described embodiments are preferred examples of the present invention. However, the present invention is not limited thereto, and various modifications can be made without departing from the gist of the present invention.

Explanation of Signs

[0111] 10 Server 14 Spoilage risk calculation unit (specification unit, first calculation unit) 60 Producer terminal 70 Information processing device 72 Internal spoilage detection unit (second calculation unit) 74 Comprehensive evaluation unit (evaluation unit) 79 Camera 100 Harvest evaluation system

Claims

1. determining an index value indicating the effect of each of the plurality of data on spoilage of the harvested crop based on each of the plurality of data including measurement data obtained by measuring the harvested crop, environmental data on the environment in which the crop was grown, and performance data on spoilage-related work or treatment performed on the crop; calculating a first evaluation value indicating a risk of spoilage of the harvested product from a plurality of index values ​​identified based on each of a plurality of data; evaluating the harvested product using the first evaluation value; A crop evaluation method including:

2. The harvest evaluation method described in claim 1, characterized in that in the step of calculating the first evaluation value, the first evaluation value is calculated using a product of a weight set for each of the multiple index values ​​and each of the multiple index values.

3. The method includes inputting an image of the harvest captured by a camera into a learning model to calculate a second evaluation value indicating a high likelihood of internal decay occurring in the harvest; 2. The harvest evaluation method according to claim 1, wherein in the evaluating step, the harvest is evaluated using the first evaluation value and the second evaluation value.

4. The crop evaluation method according to claim 3 , wherein the image is an RGB image.

5. A harvest evaluation method according to any one of claims 1 to 4, characterized in that in the process of evaluating the harvest, at least one of the following is identified: whether the harvest can be shipped, the shipping time, and the storage method until shipment.

6. A step of calculating a first evaluation value indicating a risk of spoilage of the harvested product based on at least one of measurement data obtained by measuring the harvested product of the crop, environmental data related to the environment in which the crop was grown, and performance data of spoilage-related work or treatment performed on the crop; A step of inputting an image of the harvest captured by a camera into a learning model to calculate a second evaluation value indicating a high likelihood of internal decay occurring in the harvest; and evaluating the harvest using the first evaluation value and the second evaluation value.

7. an identification unit that identifies an index value indicating the influence of each of a plurality of data on the decay of the harvested product based on each of the plurality of data among measurement data obtained by measuring the harvested product of the crop, environmental data on the environment in which the crop was grown, and performance data of work or processing related to the decay performed on the crop; A first calculation unit that calculates a first evaluation value indicating a risk of spoilage of the harvested product from a plurality of index values ​​identified based on each of a plurality of data; an evaluation unit that evaluates the harvested product using the first evaluation value; A crop evaluation system comprising:

8. A second calculation unit calculates a second evaluation value indicating a degree of likelihood of internal decay occurring in the harvest by inputting an image of the harvest captured by a camera into a learning model, The harvest evaluation system according to claim 7 , characterized in that the evaluation unit evaluates the harvest using the first evaluation value and the second evaluation value.

9. a first calculation unit that calculates a first evaluation value indicating a risk of spoilage of the harvested product based on at least one of measurement data obtained by measuring a harvested crop, environmental data related to an environment in which the crop was grown, and performance data of spoilage-related work or treatment performed on the crop; A second calculation unit that calculates a second evaluation value indicating a high likelihood of internal decay in the harvest by inputting an image of the harvest captured by a camera into a learning model; A harvest evaluation system comprising: an evaluation unit that evaluates the harvest using the first evaluation value and the second evaluation value.

10. determining an index value indicating the effect of each of the plurality of data on the decay of the harvested product based on each of the plurality of data including measurement data obtained by measuring the harvested product of the crop, environmental data on the environment in which the crop was grown, and performance data on the work or treatment related to the decay performed on the crop; Calculating a first evaluation value indicating a risk of spoilage of the harvested product from a plurality of index values ​​identified based on each of the plurality of data; Evaluating the harvested product using the first evaluation value. A crop evaluation program that causes a computer to execute processing.

11. a computer executes a process of inputting an image of the harvest captured by a camera into a learning model to calculate a second evaluation value indicating a degree of likelihood of internal decay occurring in the harvest; 11. The harvest evaluation program according to claim 10, wherein the evaluation process is a process of evaluating the harvest using the first evaluation value and the second evaluation value.

12. calculating a first evaluation value indicating a risk of spoilage of the harvested product based on at least one of measurement data obtained by measuring the harvested product of the crop, environmental data relating to an environment in which the crop was grown, and performance data of spoilage-related work or treatment performed on the crop; The image of the harvest captured by the camera is input into a learning model to calculate a second evaluation value indicating the likelihood of internal decay occurring in the harvest; evaluating the harvested product using the first evaluation value and the second evaluation value; A crop evaluation program that causes a computer to execute processing.

Images