JP2020054289A - Harvest prediction system for facility cultivated fruits - Google Patents

Abstract

To provide harvest prediction system for facility cultivated fruits with a high product value due to securing inexpensive and reliable harvesting labor and excellent and reliable shipping destinations by accurately predicting the harvest date, yield, size, etc. of fruits.SOLUTION: The present invention comprises: a fruit harvest date prediction unit 3 for determining the flowering date of strawberry, a fruit, from the time-lapse image 20 of the fruit taken by the imaging camera 114, and setting the harvest date from the estimated integrated temperature in a house after the flowering date; a fruit yield prediction unit 4 for predicting the yield of strawberry by accumulating information on the number of the strawberry flowers; a flower bed volume calculation unit 5 for calculating the flower bed volume of the strawberry; and a fruit size predicting unit 6 for predicting strawberry size from the flower bed volume.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a system for predicting the harvest of facility-grown fruits, for example, regarding the harvest time, harvest amount, size, etc., of facility-grown fruits such as strawberries, and predicts the harvest time, amount, and size for each individual fruit. The present invention relates to a system for predicting harvest of institutionally grown fruits. In addition, the present invention relates to a system for predicting the harvest time of a facility cultivated fruit, which more accurately predicts the harvest time, amount, and size using artificial intelligence. In the present specification, the cultivation of "strawberry" will be described as an example of the target of facility-grown fruit, but the present invention is not limited to this, and is also applied to other facility-cultivated fruits such as tomato, publica, tangerine, and grape. You. In addition, the strawberry varieties exemplified in the present specification are intended for all strawberry varieties such as “red hop” and “tochiotome”.

  FIG. 6 shows one embodiment of the greenhouse 100 to which the present harvest prediction system 1 is applied. The harvest prediction system 1 is applied to a strawberry 2 which is a fruit 23 cultivated on a cultivation bed 108 provided in a greenhouse 100 in which an indoor environment is controlled. In this greenhouse 100, a transparent covering material 107 such as a fluorine film, glass, or the like is attached to a frame composed of a column member 101 and a beam member 102, for example. On the roof, a skylight 103 with an insect net and a stirring fan 104 are provided, and on the ceiling, an automatic light-shielding curtain 105 for controlling solar radiation and an automatic heat-insulating curtain 106 for averaging indoor temperature are provided. The indoor environment of the greenhouse 100 is controlled. The greenhouse 100 shown in FIG. 6 is an example of a facility that performs so-called “house cultivation”, and the greenhouse 100 of the present invention is not limited to this example.

  In the greenhouse 100, a cultivation bed 108 for cultivating the strawberry 2 as the cultivated fruit 109 is juxtaposed, and a seedling of the strawberry 2 is provided on the cultivation bed 108. Then, the automatic traveling robot 113 travels along the traveling path 110 between the cultivation beds 108, and photographs the strawberry 2 with the imaging camera 114 mounted on the automatic traveling robot 113. The self-propelled robot 113 is provided with an image monitor 112 and a transmission / reception antenna 111, displays information on the cultivated strawberry 2 on the image monitor 112, and displays information on the cultivated strawberry 2 on a centralized monitor or the like. Send and receive. In addition, the automatic traveling robot 113 is provided with a notification means such as a warning sound generator 115, and when the cultivated strawberry 2 has a defective product such as deformed fruits or when the strawberry 2 is mutated. A warning sound is generated when a person in charge is present, and the person in charge is notified.

  Since the strawberry 2 is easily affected by the natural environment such as a typhoon or a long rain, in recent years, it is often cultivated in the greenhouse 100 where the indoor environment is controlled. As described above, cultivation in the greenhouse 100 makes it possible to control the harvest time. However, once a pest or disease represented by "mildew" spreads easily in the greenhouse 100, it is necessary to perform quality control of the strawberry 2 manually.

  Patent Literature 1 discloses a harvest prediction device and method that can reduce time and cost and that predict the yield of crops in a wide area. Here, the image analysis unit acquires spectrum data from the image data in which the crop is captured, compares the acquired spectrum data with the spectrum data stored in the spectrum DB, identifies the type of crop, and the growing stage, and crops. Calculate the area. Further, the crop yield is predicted from the crop type, the growing stage, the cropped area obtained by the image analysis unit, the crop growing period stored in the crop information DB, and the yield per unit area. Is described.

  Patent Literature 2 discloses a harvesting device capable of automatically and quickly harvesting yellow-green crops. Here, it has a movable body that has a harvesting unit and a traveling control unit and is capable of self-traveling, and a position detecting device of the movable body from a transmitter and a receiver, and the movable body blooms the harvest target. An image processing device for recognizing the situation and the fruiting situation, a memory for storing the flowering position and time recognized by the image processing device, and a database for calculating the fruiting time from the flowering situation; It is described that a calculation unit is provided that predicts the time and position of the fruiting based on the content and the detection result of the image processing apparatus.

  In Patent Document 3, based on data obtained by sensing with an unmanned aerial vehicle such as a drone, the management of farmland, etc., and the prediction of the growth and harvest of crops are automatically performed, and the appropriate management of farmland and crops is performed. The agricultural management prediction system which implement | achieves is disclosed. Here, an unmanned aerial vehicle, a communication unit that receives image data sent from the management terminal in a server device that can freely communicate with the management terminal, a vegetation indexing unit that performs vegetation indexing based on the image data, A standard deviation converting section for converting information into a standard deviation, and a semantic section for converting the standard deviation into meaning are provided, and the route and photographing of the unmanned aerial vehicle are remotely controlled by the management terminal, and based on image data obtained by the unmanned aerial vehicle. It is described that the server apparatus side performs management prediction of agriculture.

JP-A-2003-6612 JP-A-6-133624 JP 2018-46787 A

  Institutional cultivation of strawberries has the problem of increasing the loss rate due to overmaturity because strawberries etc. are crops that do not last for a long time. This has been a problem in that labor costs have to be increased.

  Moreover, in the cultivation of strawberries in a facility, there is a problem that the growing period and the yield of fruits cannot be accurately predicted or must rely on intuition of a skilled person. In addition, there is a problem that it is not possible to accurately predict the size of a fruit that greatly affects the selling price.

  The purpose of the present application is to solve such a problem and accurately predict the harvest date, harvest amount, size, etc. of the fruit, thereby ensuring a low-priced and reliable harvest labor force and securing a good and reliable shipping destination, thereby increasing the commercial value. The purpose of the present invention is to provide a system for predicting the yield of institutionally grown fruits.

  In addition, in facility cultivation of fruits, individual features of each fruit are extracted using artificial intelligence neural networks, etc., and each fruit is individually managed, and the harvest date, amount, size, etc. of the fruits are predicted without human intervention. In addition, it is possible to provide a facility cultivation fruit harvest prediction system with enhanced commercial value.

  Furthermore, in the institutional cultivation of fruits, deep learning of artificial intelligence, etc., makes it possible to more accurately predict the harvest date, amount, size, etc. of fruits by referring to past big data. Harvest prediction system can be provided.

  In order to achieve the above object, the system for predicting the harvest of greenhouse cultivated fruits according to the present invention captures fruits cultivated in a greenhouse in which the indoor environment is controlled by an imaging camera, and obtains the fruits from the time-lapse images captured by the imaging camera. A harvest date prediction unit that determines the flowering date and sets the harvest date from the scheduled accumulated temperature in the greenhouse after the flowering date, and a fruit harvest number prediction unit that integrates the flowering number information of the fruit and predicts the harvest number of the fruit And a flower bed volume calculation unit that calculates the flower bed volume of the fruit captured by the imaging camera, and a fruit size prediction unit that predicts fruit size information from the flower bed volume.

  With the above configuration, for each fruit cultivated in a greenhouse where the indoor environment is controlled, the date of harvest, the number of harvested fruits, and the size of the fruit can be accurately predicted from the time-lapse image taken by the imaging camera. For example, it is possible to determine the flowering date of the fruit from a regular temporal image taken by the imaging camera, and to set the harvest date of the fruit from the scheduled integrated temperature in the greenhouse after the flowering date. In addition, by integrating the prediction of the harvest date of the fruits for all the fruits in the greenhouse, it is possible to predict the number of fruits harvested in the entire greenhouse. In addition, the weight of fruits grown in a facility, such as strawberries, is approximately proportional to the flower bed volume. Therefore, by measuring the flower bed volume of a fruit, it is possible to predict in advance the yield of fruits such as large strawberries having a high sales unit price.

  In addition, the facility cultivated fruit harvest prediction system includes an identification number assigning unit that assigns an identification number to each fruit, a feature amount extracting unit that analyzes a temporal image of the fruit to extract individual feature amounts, and associates the identification amount with the identification number. It is preferable to provide In this manner, a time-lapse image of a fruit such as a strawberry is analyzed by an artificial intelligence neural network or the like, and individual feature amounts of the image are extracted and assigned identification numbers. As described above, by individually recognizing and cultivating fruits such as strawberry to be cultivated, more accurate quality control can be performed.

  Further, in the facility cultivation fruit harvest prediction system, it is preferable that the individual feature amount of each fruit photographed by the feature amount extraction unit is stored as an individual feature amount in the individual fruit management database together with the identification number. In this way, fruits such as strawberries are individually separated by individual characteristic amounts such as shapes that are different for each fruit, for example, the peripheral shape related to the strawberries themselves such as “slabs”, or “separation”. Can be identified. And based on this individual fruit management database, more precise quality control becomes possible.

  In addition, the system for predicting the yield of institution-grown fruits can identify the fruits by comparing the individual characteristic amounts of the fruits taken by the imaging camera with the characteristic amount extraction unit with the individual characteristic amounts of the fruits stored in the individual fruit management database. preferable. As a result, fruits such as strawberries imaged by the imaging camera grow periodically and their shapes change, but each time the fruits are retrieved from the individual feature amount of the fruits and individually imaged by the imaging camera, they are individually identified and identified. Can be managed.

  Further, it is preferable that the harvest prediction system for the institutional grown fruits stores history information of a temporal image of the flower bed of each fruit in the individual fruit management database. Thus, a temporal image of a flower bed of a fruit such as a strawberry can be stored as big data, and can be used in prediction of a harvest date, a harvest number, and a size of fruit cultivation every year.

  Further, in the harvest prediction system for institutionally grown fruits, it is preferable that the harvest date prediction unit corrects the harvest date set from the temporal image of the fruits stored in the individual fruit management database. As a result, when the harvest date set from the flowering date of fruits such as strawberries and the estimated accumulated temperature in the house after the flowering date deviates from the initial harvest date due to other factors such as sunshine conditions, for example, The past time-lapse image of fruits such as strawberries stored in the fruit management database is reflected, and the harvest date set by artificial intelligence deep learning or the like can be corrected.

  In addition, the system for predicting the yield of institution-grown fruits requires the fruit size prediction unit to predict the size of fruits based on comparisons of past big data stored in the individual fruit management database with historical images of the fruits. Is preferred. As a result, the final size can be accurately predicted by artificial intelligence from past progress information in which the flower bed of a fruit such as a strawberry is enlarged and becomes a fruit. The “size related to fruit such as strawberry” may be the weight of fruit such as strawberry.

  In the system for predicting the harvest of institutionally grown fruits, it is preferable that the imaging camera is mounted on an automatic traveling robot and runs in parallel along a growing bed adjusted to a height that allows close-up photography. Accordingly, the automatic traveling robot can take close-up shots of the flower bed, fruits, and the like of fruits such as strawberries provided on the cultivation bed, and can perform detailed analysis with artificial intelligence.

  Furthermore, the system for predicting the harvest of institution-grown fruits further includes a fruit quality check unit, and if an abnormality has occurred in the fruit from a photographed image of the fruit by the automatic traveling robot, an alarm may be issued as a defective product. preferable. By this, abnormalities of fruits such as strawberries that rely on visual inspection until now, for example, if fruits such as strawberries do not grow due to disease, etc. It can be detected in advance by artificial intelligence. It is also possible to detect whether or not a disease such as "mildew" has occurred not only in fruits but also in leaves.

  As described above, according to the system for predicting the harvest of institutionally grown fruits according to the present invention, by accurately predicting the date of harvest of fruits, the amount of harvest, the size, and the like, an inexpensive and reliable harvest labor and an excellent and reliable harvest labor By securing the shipping destination, it is possible to provide a system for predicting the yield of institutionally grown fruits having high commercial value.

  In addition, in facility cultivation of fruits, individual features of each fruit are extracted using artificial intelligence neural networks, etc., and each fruit is individually managed, and the harvest date, amount, size, etc. of the fruits are predicted without human intervention. In addition, it is possible to provide a facility cultivation fruit harvest prediction system with enhanced commercial value.

  Furthermore, in facility cultivation of fruits, deep learning of artificial intelligence, etc., makes it possible to predict the date of harvest, amount of harvest, size, etc. of fruits with higher accuracy by referring to past big data, and high-quality facility cultivation A fruit harvest prediction system can be provided.

BRIEF DESCRIPTION OF THE DRAWINGS It is a block diagram which shows schematic structure of one Embodiment of the harvest prediction system of the facility cultivation fruit which concerns on this invention. It is explanatory drawing which shows the individual fruit management database which manages fruits, such as a strawberry grown in a greenhouse, individually. It is explanatory drawing which shows the structure which estimates a harvest date by this harvest prediction system. It is explanatory drawing which shows the structure which estimates the amount of harvest by this harvest prediction system. It is explanatory drawing which shows the structure which estimates the size of a fruit by this harvest prediction system. It is a top view which shows one Example of the greenhouse which applies this harvest prediction system. FIG. 2 is an explanatory diagram illustrating a schematic structure of one of the automatic traveling robots equipped with an imaging camera in a greenhouse. It is explanatory drawing which shows the name of the flower of strawberry which is one Example of a fruit, and a fruit.

(Structure of fruit harvest prediction system)
Hereinafter, the harvest prediction system 1 for institutionally grown fruits according to the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of one embodiment of the harvest prediction system 1. In the present specification, the case where the facility-cultivated fruit 23 is “strawberry” will be described. However, the present invention is not limited to this, and is also applicable to other facility-cultivated strawberries 2 such as, for example, tomato, publica, tangerine, and grape. You. In addition, the variety of the strawberry 2 exemplified in this specification covers all types of the strawberry 2, such as “red hop” and “tochiotome”.

(Automatic traveling robot)
FIG. 7 shows a schematic structure of one of the automatic traveling robots 113 on which the imaging camera 114 is mounted. The automatic traveling robot 113 traveling inside the greenhouse 100 automatically travels around the cultivation bed 108. The automatic traveling robot 113 has an imaging camera 114 mounted thereon so that the height of the cultivation bed 108 can be photographed in close-up. Thereby, the automatic traveling robot 113 can take a close-up image of all the strawberries 2 in the greenhouse 100. The imaging camera 114 periodically captures the temporal image 20 (see FIG. 1) of all the strawberries 2 grown in the greenhouse 100.

(Structure of a system for predicting harvest of institutionally grown fruits)
FIG. 1 shows a configuration of the harvest prediction system 1. Each strawberry 2 cultivated on the cultivation bed 108 is periodically photographed as a temporal image 20 by the imaging camera 114. The time-lapse image 20 is basically assumed to be a point in time when flowering is completed once a day, but is not limited to this point and may be taken as needed. Then, the harvest prediction system 1 predicts a harvest date, a harvest number, and a fruit size of the strawberry 2 based on the photographed temporal image 20 of each strawberry 2. The harvest prediction system 1 includes a fruit harvest date prediction unit 3, a fruit harvest number prediction unit 4, a flower bed volume calculation unit 5, and a fruit size prediction unit 6.

(Fruit harvest date prediction, harvest number prediction, fruit size prediction)
FIG. 8 shows the names of flowers and fruits of the strawberry 2 to be cultivated. FIG. 8A shows a strawberry flower, and FIG. 8B shows a strawberry fruit 23. The fruit harvest date prediction unit 3 determines the flowering date of the strawberry 2 from the temporal image 20 captured by the imaging camera 114. After the flowering date, the number of days from the flowering date of the strawberry 2 to the harvest date is set based on the scheduled accumulated temperature 27 of the greenhouse 100. The unit of the scheduled integrated temperature 27 is represented by (° C./day). In the case of the strawberry 2, the number of days from flowering to harvest can be determined based on varieties such as “red hop” and “tochiotome” and the accumulated temperature in the greenhouse 100. That is, in a greenhouse where the temperature and the like are controlled, the harvest date can be predicted almost accurately. The fruit yield prediction unit 4 predicts the yield of the strawberry 2 by integrating the information on the flowering date of the strawberry 2 predicted by the fruit harvest date prediction unit 3, that is, the flowering number information 12. The number of healthy fruits that can be harvested from a flowering flower varies depending on varieties, climate, diseases, pests, bees and the like, but can be generally predicted from past data and the like. As one calculation formula, the number of healthy fruits = the number of flowers × the fruiting rate × (1−the defective rate) can be expressed. Here, the defective rate refers to the probability of causing pest damage or deformed fruits. The flower bed volume calculation unit 5 calculates the volume of the flower bed 19a of the strawberry 2 captured by the imaging camera 114. The flower bed volume 14 (V) is obtained by V = radius (R) of flower bed 19a × π × height / 3. In other words, the fruit is attached to the flower bed 19a and includes one seed. As the number of seeds on the flower bed 19a increases, the strawberry 2 grows larger. Thus, it can be said that there is a correlation between the flower bed volume 14 and the size of the strawberry 2. The fruit size prediction unit 6 predicts the size information 13 of the strawberry 2 from the volume of the flower bed 19a calculated by the flower bed volume calculation unit 5. This is because, as described above, "the flower bed volume 14 of the strawberry 2 and the weight of the harvested fruit are substantially proportional to each other".

(Individual recognition and individual management of fruits)
FIG. 1 shows an image analysis of a time-lapse image 20 captured by the imaging camera 114, an identification number assigning unit 7 that assigns an identification number 18 to a strawberry 2, and a feature amount extracting unit that extracts an individual feature amount 15 of the strawberry 2. 8 is shown. Also, an individual fruit management database 10 for individually recognizing and managing strawberries 2 grown in the greenhouse 100 is shown. In the individual fruit management database 10, an identification number 18 and an individual feature value 15 of the strawberry 2 cultivated in the greenhouse 100 are stored.

  As described above, the harvest prediction system 1 includes the identification number assigning unit 7 and the feature amount extracting unit 8, and the identification number assigning unit 7 assigns the identification number 18 to each strawberry 2 for identification. The individual feature value 15 of the strawberry 2 is, for example, the number, position, and shape of the pistils 19b and stamens 19c of the flower bed 19a, and the soka 19d, petal 19g, sepal 19e, fruit handle 19f, and the like shown in FIG. The number, position, and shape of petals 19g, etc., and a feature that allows each strawberry 2 to be distinguished from other strawberries 2. The individual characteristic amount 15 is either the individual characteristic amount 15 at the stage of the “flower bed 19a” of the strawberry 2 or the individual characteristic amount 15 at the stage when the flower bed 19a is enlarged and developed into flesh. Is also included. The feature amount extraction unit 8 extracts these individual feature amounts 15 and associates them with the identification number 18.

  Each time the image capturing camera 114 takes an image, the feature amount extraction unit 8 compares the individual feature amount 15 of the strawberry 2 with the individual feature amount 15 of the strawberry 2 stored in the individual fruit management database 10, and is the strawberry 2. Identify that. In other words, the strawberry 2 is periodically photographed by the imaging camera 114, and the strawberry 2 can be identified from the individual characteristic amount 15 by searching the individual fruit management database 10 each time. Thereby, the latest image of the strawberry 2 is obtained every time the strawberry 2 is imaged by the imaging camera 114. In this manner, the history information 17 for the time-lapse image 20 of each strawberry 2 is stored in the individual fruit management database 10.

  FIG. 2 shows the contents of the individual fruit management database 10 that individually manages the strawberries 2 grown in the greenhouse 100. As shown in the upper part of FIG. 2, in the greenhouse 100, the strains of the strawberry 2 are indicated as “strain 1”, “strain 2”,. In each strain, the petals 19 g of the strawberry 2 become pulp, and the strawberry 2 is individually managed by giving the identification application number 18 thereto. For example, strain 1 is assigned strains 1-1 to 1-4 as identification number 18, strain 2 is assigned strains 2-1 to 2-3 as identification number 18, and strain n is assigned identification number. As strain 18, strains n-1 to n-3 are given. Then, the individual fruit management database 10 stores an individual feature quantity 15 associated with the identification number 18 for each strawberry 2.

  As an example, the individual fruit management database 10 stores the individual characteristic amount 15 for the strawberry 2 with the identification number 18 (strain 2-3), and also displays the time-lapse image 20 of the flowering date of the strawberry 2 and the fruit set of the strawberry 2 The chronological image 20 of the day and the chronological image 20 of the strawberry 2 at the time of harvest are stored together with the shooting date and time. Then, information indicating that the flower has bloomed is recorded in the time-lapse image 20 of the flowering date of the strawberry 2 along with the record of the year, month, day, and time. Information indicating that the fruit has been set is written together with the recording of the time, and the chronological image 20 of the strawberry 2 at the time of harvest is stored together with the shooting date and time. Then, information indicating that the strawberry 2 has been harvested is written in the temporal image 20 of the harvest date of the strawberry 2 along with the record of the year, month, day and time. From these information, the day information (N1) from the flowering date to the fruiting date of all the strawberries 2 and the day information (N2) from the fruiting date to the harvest date are calculated and recorded in the individual fruit management database 10. You.

(Calculation of modified harvest date)
FIG. 3 is a block diagram showing a configuration in which the harvest date of each strawberry 2 is predicted by the fruit harvest date prediction unit 3. The fruit harvest date prediction unit 3 corrects the harvest date set from the time-lapse image 20 of the strawberry 2 stored in the individual fruit management database 10. That is, as described above, the fruit harvest date prediction unit 3 determines the flowering date information 25 based on the cultivation plan 26 of the strawberry 2 from the temporal image 20 captured by the imaging camera 114. Then, after the flowering date, the number of days from the flowering date of the strawberry 2 to the expected harvest date 28a is set from the scheduled accumulated temperature 27 in the greenhouse 100. In order to set the scheduled integrated temperature 27, the type data 31a is required. This method is based on the fact that, if the greenhouse 100 is cultivated in a controlled environment, the number of maturation days from the flowering date to the harvest date can be predicted to some extent. However, the set harvest date may fluctuate due to climate change related to the growth of the strawberry 2, the occurrence of diseases and pests, the working condition of bees and the like. For example, since the strawberry 2 is cultivated in the greenhouse 100 where the indoor environment is controlled, the indoor temperature should be constant, but the strawberry 2 may be affected by climate change due to a typhoon, a cold wave, or the like. On the other hand, the corrected harvest date 28b can be calculated by performing the harvest date prediction 30a by deep learning of artificial intelligence or the like.

  For the calculation of the corrected harvest date 28b, as shown in FIG. 2, information on the number of days (N1) from the past flowering date to the fruiting date and the past fruiting date stored in the individual fruit management database 10 A method of predicting the day information (N2) from the harvest date to the harvest date as big data by artificial intelligence can be adopted. That is, the fruit harvest date prediction unit 3 can correct the harvest date set from the time-lapse image 20 of the strawberry 2 stored in the individual fruit management database 10. That is, the number of mature days from the flowering date of the strawberry 2 to the expected harvest date 28a is set by the scheduled integrated temperature 27 within 100 from the flowering date to the harvest date. However, as described above, the number of ripening days may fluctuate due to factors other than the scheduled integrated temperature 27, and this is corrected by the time-lapse image 20 of the strawberry 2 stored in the individual fruit management database 10. For example, when the maturity of the strawberry 2 is advanced or delayed at a certain point in time based on past data, the corrected harvest date 28b is calculated in consideration of the error (X) of the number of days. .

(Example of using artificial intelligence)
The fruit harvest date prediction unit 3 uses the artificial intelligence deep learning to determine the maturity of the strawberry 2 from the time-lapse images 20 of all the strawberry 2 stored in the individual fruit management database 10 to determine the strawberry 2 in the greenhouse 100. The state can be arranged in a plane. For example, the degree of maturity of the strawberry at which coordinate position or height position is advanced or delayed with respect to past data is analyzed by deep learning. As a result, it is possible to determine whether the error of the number of days (X) is caused by climate change, the occurrence of a disease or a pest, the working condition of a bee or the like, or any other reason.

  With respect to the time-lapse image 20 of the strawberry 2 taken periodically, the artificial intelligence can detect a change in the number of maturation days due to factors other than the scheduled accumulated temperature 27 without overlooking a subtle change. Then, by accumulating the fluctuation, the final harvest date can be accurately corrected. When cultivation is performed outside the greenhouse 100, it is generally possible to predict the harvest time to some extent if there is standard data and average weather data of the crop to be cultivated. According to the method, it is possible to predict the harvest time in consideration of the case where the number of maturation days fluctuates due to factors other than the scheduled integrated temperature 27.

(Calculation of the number of corrected crops)
FIG. 4 is a block diagram showing a configuration in which the yield of each strawberry 2 is predicted by the fruit yield predictor 4. The fruit harvest number prediction unit 4 corrects the harvest number set from the time-lapse image 20 of the strawberry 2 stored in the individual fruit management database 10. That is, as described above, the fruit harvest number prediction unit 4 determines the flowering date information 25 of the strawberry 2 from the temporal image 20 captured by the imaging camera 114. Then, after the flowering date, the fruit set date of the strawberry 2 is predicted from the scheduled integrated temperature 27 in the greenhouse 100, and the expected harvest number 29a is set. To set the expected harvest number 29a, a fruit set rate is required as the variety data 31b. This method is that "if the greenhouse 100 is cultivated in a controlled environment, the number of maturation days from the flowering date to the harvest date can be predicted to some extent". However, the set harvest date fluctuates due to climate change, the occurrence of diseases and pests, the working condition of bees and the like related to the growth of the strawberry 2, and the number of harvests may fluctuate accordingly. In such a situation, the corrected harvest number 29b can be calculated by performing the harvest number prediction 30b in consideration of the error (Y) of the harvest number by the artificial intelligence described above.

  With respect to the time-lapse image 20 of the strawberry 2 taken periodically, the artificial intelligence can detect a change in the number of maturation days due to factors other than the scheduled accumulated temperature 27 without overlooking a subtle change. Then, by accumulating this variation, the final harvest number can be accurately corrected. In addition, when cultivating outside the greenhouse 100, in general, it is possible to predict a certain number of harvests if there is standard data of the cultivated crop and average weather data, but the prediction is made by artificial intelligence. According to the method, the number of harvests can be predicted in consideration of the case where the number of maturation days fluctuates due to factors other than the scheduled integrated temperature 27.

(Fruit size prediction system)
FIG. 5 is a block diagram showing a configuration in which the size of each strawberry 2 is predicted by the fruit size prediction unit 6. The fruit size prediction unit 6 predicts the size information 13 of the strawberry 2 from the volume of the flower bed 19a calculated by the flower bed volume calculation unit 5 based on the temporal image 20 of each strawberry 2. This is because the flower bed volume 14 and the weight of the harvested fruit are in a proportional relationship. However, the harvest date and amount set according to the climate change, the occurrence of diseases and pests, the working condition of bees and the like related to the growth of the strawberry 2 fluctuate, and the fruit size may fluctuate accordingly. In such a situation, the corrected size information 13 of the strawberry 2 can be calculated by performing the fruit size prediction 30c using artificial intelligence.

  As shown in FIG. 5, the fruit size prediction unit 6 compares the past strawberry harvest size data 32 stored in the individual fruit management database 10 for each strawberry 2 with the time-lapse image 20 of the strawberry 2 and Can be predicted by correcting the size information 13 of. For example, regarding the maturity of the strawberry 2, if the size information 13 at a certain point in time is large or delayed from the past data, the size information corrected in consideration of the error (Z) of the size information 13 is is expected.

  The fruit size prediction unit 6 uses artificial intelligence deep learning to arrange the size of the strawberry 2 from the time-lapse images 20 of all the strawberry 2 stored in the individual fruit management database 10 in a planar manner in the greenhouse 100. be able to. For example, the degree of maturity of the strawberry at which coordinate position or height position is advanced or delayed with respect to past data is analyzed by deep learning. As a result, it is possible to determine whether the error (Z) in the size information 13 is caused by climate change, the occurrence of a disease or a pest, the working condition of a bee or the like, or any other reason. .

  In the case of strawberries, the selling destination differs depending on the size information 13. Therefore, it is important for a sales strategy to know in advance how much strawberry 2 can be harvested and in what quantity. As described above, the size of the strawberry 2 can be predicted to some extent from the flower bed volume 14, and the size distribution of the crop other than the strawberry 2 can be estimated to some extent. The size can be predicted in consideration of the case where the number of maturation days fluctuates due to other factors.

(Fruit quality control by automatic traveling robot)
The harvest prediction system 1 further includes a fruit quality confirmation unit 9. That is, the automatic traveling robot 113 has a feature that, in principle, all the strawberries 2 in the greenhouse 100 are regularly photographed without leakage. Moreover, it has the characteristic that the quality of the strawberry 2 can be managed individually. Taking advantage of these characteristics, the abnormality of the strawberry 2 which has been relying on the visual observation until now, for example, when the strawberry 2 does not grow due to a disease, when the strawberry 2 has a defective product such as deformed fruit, the strawberry 2 suddenly Strawberries 2 that cannot be shipped, such as when they have a mutation, can be eliminated in advance by artificial intelligence.

  In addition, the harvest prediction system 1 detects whether or not signs of a disease such as "mildew" have occurred not only in the strawberry 2 but also in the leaves of the strawberry 2, and if so, measures are taken at an early stage. Can take. When the occurrence of a disease or the occurrence of a defective product is detected, the automatic traveling robot 113 determines the type of the problem together with the identification number 18 of the strawberry 2 on the monitor of the automatic traveling robot 113 or the centralized monitor for quality control. Can be displayed. The quality control of these strawberries 2 can reduce labor costs by substituting the work performed visually so far with the automatic traveling robot 113, and also minimize omission of checks.

  The configuration, shape, size, and arrangement relationship of the institution-cultivated fruit harvest prediction system 1 described in the above embodiments are merely schematic diagrams to the extent that the present invention can be understood and implemented. Therefore, the present invention is not limited to the described embodiments, and can be modified in various forms without departing from the scope of the technical idea described in the claims.

1. Harvest forecasting system (for institutionally grown fruits), 2. Strawberry, 3. Fruit harvest date forecasting unit, 4. Fruit harvest number forecasting unit, 5. Flower bed volume calculation unit, 6. Fruit size forecasting unit, 7. Identification number assigning unit, 8. Feature value Extraction unit, 9 Fruit quality confirmation unit, 10 Individual fruit management database, 12 Flowering number information, 13 Size information, 14 Flower bed volume, 15 Individual feature amount, 17 History information, 18 Identification number, 19a Flower bed, 19b Meseki, 19c Stamen, 19d Soka, 19e Sepal, 19f Fruit pattern, 19g Petal, 20 images over time, 23 fruits, 25 Flowering date information, 26 Cultivation plan, 27 Scheduled integrated temperature, 28a Expected harvest date, 28b Corrected harvest date, 29a Expected Harvest number, 29b Corrected harvest number, 30a Harvest date prediction by artificial intelligence, 30b Harvest number prediction by artificial intelligence, 30c Fruit size prediction by artificial intelligence, 31a products Data (integrated temperature), 31b Varieties data (fruit set rate), 32 Past fruit harvest size data, 100 greenhouse, 101 pillars, 102 beams, 103 skylight with insect-proof net, 104 stirring fan, 105 automatic shading curtain, 106 Automatic warming curtain, 107 transparent covering material, 108 cultivation bed, 109 cultivated fruit (strawberry), 110 (automatic traveling robot) traveling path, 111 transmission / reception antenna, 112 image monitor, 113 automatic traveling robot, 114 imaging camera, 115 alarm sound generation Vessel, 116 drainage port, 117 drip irrigation tube, 118 medium, N1 information on the number of days from the flowering date of the fruit to the fruiting date, N2 information on the number of days from the fruiting date of the fruit to the harvest date, error in X days, Y Error in harvest number, error in Z size.

Claims (9)

Shooting the fruits grown in the house where the indoor environment is controlled by the imaging camera,
A fruit harvest date prediction unit that determines the flowering date of the fruit from the time-lapse image taken by the imaging camera, and sets a harvest date from a scheduled integrated temperature in the house after the flowering date,
A fruit harvest number prediction unit for integrating the flowering number information of the fruit and predicting the harvest number of the fruit,
A flower bed volume calculation unit that calculates the flower bed volume of the fruit imaged by the imaging camera,
Fruit size prediction unit for predicting the size of the fruit from the flower bed volume,
A harvest prediction system for facility-grown fruits, comprising:   The facility cultivation fruit harvest prediction system according to claim 1, wherein an identification number assigning unit that assigns an identification number to each fruit, and the temporal image of the fruit is analyzed to extract an individual characteristic amount of the fruit. And a feature quantity extracting unit that associates the identification number with the identification number.   3. The system for predicting harvest of facility grown fruits according to claim 1 or 2, wherein the individual characteristic amount of each fruit photographed by the characteristic amount extraction unit is included in the individual fruit management database together with the identification number as an individual characteristic amount. A harvest prediction system for institutionally grown fruits, which is preserved.   4. The system for predicting the harvest of institutionally grown fruits according to claim 1, wherein the characteristic amount extracting unit stores the individual characteristic amounts of the fruits photographed by an imaging camera in an individual fruit management database. 5. A system for predicting harvest of institutionally grown fruits, wherein the fruits are identified by comparing the individual characteristic amounts of the fruits.   5. The system for predicting the yield of institutionally grown fruits according to any one of claims 1 to 4, wherein the individual fruit management database stores history information of a temporal image of a flower bed of each fruit. System for predicting the harvest of institutional grown fruits.   The harvest prediction system for institutionally grown fruits according to any one of claims 1 to 5, wherein the fruit harvest date prediction unit sets a harvest date set from a temporal image of the fruits stored in an individual fruit management database. A system for predicting the yield of institutionally grown fruits, wherein the system is modified.   The harvest prediction system for institutionally grown fruits according to any one of claims 1 to 6, wherein the fruit size prediction unit includes past big data regarding the size of the fruits stored in an individual fruit management database, A harvest prediction system for facility-cultivated fruits, wherein a size of the fruits is predicted from a comparison with a temporal image of the fruits.   8. The system according to claim 1, wherein the imaging camera is mounted on an automatic traveling robot, and is arranged along a cultivation bed adjusted to a height at which close-up photography is possible. 9. A system for predicting the harvest of institutionally grown fruits, characterized by running side by side in a house. 9. The system for predicting the harvest of institutionally grown fruits according to claim 1, further comprising a fruit quality check unit, wherein an abnormality occurs in the fruits from a photographed image of the fruits by an automatic traveling robot. 10. If so, an alert is issued as a defective product, and a system for predicting the harvest of institutionally grown fruits is provided.

Images